Download AN ABSTRACT OF THE THESIS OF

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Document related concepts

Ridge (biology) wikipedia, lookup

Biology and consumer behaviour wikipedia, lookup

Gene expression programming wikipedia, lookup

RNA-Seq wikipedia, lookup

Y chromosome wikipedia, lookup

Neocentromere wikipedia, lookup

Pathogenomics wikipedia, lookup

Site-specific recombinase technology wikipedia, lookup

Genetic engineering wikipedia, lookup

Gene expression profiling wikipedia, lookup

Gene wikipedia, lookup

Epigenetics of human development wikipedia, lookup

Genomic imprinting wikipedia, lookup

Polyploid wikipedia, lookup

Artificial gene synthesis wikipedia, lookup

Minimal genome wikipedia, lookup

X-inactivation wikipedia, lookup

Designer baby wikipedia, lookup

History of genetic engineering wikipedia, lookup

Genome evolution wikipedia, lookup

Microevolution wikipedia, lookup

Genetically modified crops wikipedia, lookup

Public health genomics wikipedia, lookup

Genome (book) wikipedia, lookup

Quantitative trait locus wikipedia, lookup

Transcript
AN ABSTRACT OF THE THESIS OF
Theerayut Toojinda for the degree of Doctor of Philosophy in Crop Science presented on
December 9, 1998. Title: Mapping and Introgression of Disease Resistance Genes in Barley
(Hordeum vulgare L.).
Redacted for privacy
Abstract approved:
Patrick M. Hayes
Molecular tools, coupled with unique germplasm stocks and rigorous phenotyping, are
useful for developing a better understanding of qualitative and quantitative disease resistance
genes in plants. The identification of molecular markers linked to all types of resistance
genes provides opportunities for implementing a range of resistance breeding strategies,
ranging from gene pyramiding to gene deployment. This thesis consists of two chapters. The
first describes a disease resistance gene mapping effort and the second describes a disease
resistance gene introgression effort. The number, location, and effects of genes determining
resistance to stripe rust, leaf rust and Barley Yellow Dwarf Virus were determined using a
population of doubled haploid (DH) lines from the cross of Shyri x Galena. Resistance to leaf
rust was qualitatively inherited, and the locus was mapped to the long arm of chromosome 1.
Resistance to stripe rust and BYDV was quantitatively inherited. Multiple QTLs were
detected for each type of resistance. The principal stripe rust resistance QTL was on the
short arm of chromosome 5 and the principal BYDV resistance QTL was on the long arm of
chromosome 1, linked in repulsion phase with the leaf rust resistance gene. Additional QTLs
and QTL x QTL interactions were detected. The majority of the qualitative and quantitative
resistance loci detected in the Shyri x Galena population coincided with Resistance Gene
Analog Polymorphisms (RGAPs) mapped in the same population. These RGAPs were based
on degenerate primers derived from cloned resistance gene sequence motifs. These
associations should be useful for efficient resistance gene mapping and provide an approach
for ultimately isolating and describing quantitative and qualitative resistance genes. The
second chapter describes a molecular marker assisted selection (MMAS) effort to introgress
stripe rust resistance QTLs on chromosomes 4 and 7 into susceptible germplasm. DH lines
were derived form a MMAS backcross-one (BC-1) population, extensively phenotyped for
stripe rust resistance, and genotyped for the introgressed QTLs and background genome.
The resistance QTLs that were introgressed were significant determinants of resistance in the
new genetic background. Additional resistance QTLs were also detected. Together, these
chapters describe an integrated approach to disease resistance gene characterization and
utilization.
Mapping and Introgression of Disease Resistance Genes in Barley
(Hordeum vulgare L.)
by
Theerayut Toojinda
A THESIS
submitted to
Oregon State University
In partial fulfillment of
the requirement for the
degree of
Doctor of Philosophy
Presented December 9, 1998
Commencement June 1999
Doctor of Philosophy thesis of Theerayut Toojinda presented on December 9, 1998
APPROVED:
Redacted for privacy
Major Professor, representing Crop
fence
Redacted for privacy
ead of Department of Crop and Soil Science
Redacted for privacy
can of Gradua 1 chool
I understand that my thesis will become part of the permanent collection of Oregon State
University libraries. My signature below authorizes release of my thesis to any reader upon
request.
Redacted for privacy
Theerayut Toojinda, Author
ACKNOWLEDGEMENTS
I would like to thank my major professor, Dr. Patrick M. Hayes, for his advice,
encouragement and support during my Ph.D. program. I thank the members of my graduate
committee, Dr. Steve J. Knapp, Dr. Christopher E. Mundt, Dr. Tony Chen, Dr. Hugo Vivar,
Dr. Warren Kronstad, and Dr. Leslie Fuchigami for their advice and assistance. I would also
like to thank my colleagues at OSU, CIMMYT and Scottish Crop Research Institute (SCRI)
for their help and friendship.
I would like to thank the Barley Project at Oregon State University for sponsoring my
Ph.D. studies. I would like to thank the North American Barley Genome Mapping Project,
and the Oregon, Idaho, and Washington Barley Commissions for their support of my
research. Many thanks to Dr. Somvong Tragoonrung, Dr. Chalermlarb Chuiprasit, Ajan
Anong Junsrikul and all my dear friends at LARTC for encouraging me to pursue this degree.
Most of all, I extend special thanks to my father (Prajob Toojinda), my mother (Sawang
Toojinda), my aunt (Samran Juntanee) and my sisters (Panjaree, Busaba and Katekanda
Toojinda), not only for love and understanding, but also for their constant support and
encouragement. Finally, I would like to express my appreciation to my wife, Kanokrat
Tiyapun, for her love, patience, and companionship.
CONTRIBUTION OF AUTHORS
This dissertation has benefited from advice and suggestions kindly offered by Dr. Patrick
M. Hayes, my major professor and advisor. Dr. Steve J. Knapp, Dr. Christopher E. Mundt,
Dr. Tony Chen and Dr. Hugo Vivar provided detailed critiques of a draft of this dissertation.
Dr. Patrick M. Hayes proposed the subject of this research and was involved in all
genetic and statistical analysis. Dr. Hugo Vivar and his colleagues at CIMMYT were
involved in phenotype data collection. Dr. A. Kleinhofs and D. Kudrna were involved in
generating RFLPs. Xianming Chen was involved in generating RGAPs. John Korte was
involved in providing all materials and expertise needed in the lab and also helped in
development of the SSR and RGA techniques used in these experiments.
TABLE OF CONTENTS
Page
CHAPTER 1.
INTRODUCTION
Molecular Markers
1
1
Quantitative and Qualitative Disease Resistance
4
Molecular Breeding
6
CHAPTER 2.
ASSOCIATION OF QUANTITATIVE AND QUALITATIVE DISEASE
RESISTANCE GENES IN BARLEY (Hordeum vulgare) WITH
RESISTANCE GENE ANALOG POLYMORPHISMS (RGAPs)
8
Abstract
9
Introduction
9
Materials and Methods
Plant Material
Genotyping and Map Construction
Disease Resistance Phenotyping
Stripe Rust
Leaf Rust
Barley Yellow Dwarf Virus (BYDV)
QTL Analysis
Results and Discussion
13
13
13
18
18
19
20
20
21
Linkage Map Construction
21
QTL Detection
22
References
38
TABLE OF CONTENTS (Continued)
Page
CHAPTER 3.
INTROGRESSION OF QUANTITATIVE TRAIT LOCI (QTLs) DETERMINING
STRIPE-RUST RESISTANCE IN BARLEY:
45
AN EXAMPLE OF MARKER-ASSISTED LINE DEVELOPMENT
Abstract
46
Introduction
46
Materials and Methods
50
Germplasm
Genotyping
Phenotyping
50
52
55
Data Analysis
56
Results
58
Discussion
64
References
71
CHAPTER 4.
CONCLUSION
74
BIBLIOGRAPHY
78
LIST OF FIGURES
Page
Figure
2.1a
2.1b
2.1c
Linkage maps of chromosomes 1, 2, and 3 based on 94 DH
progeny from the cross of Shyri x Galena
16
Linkage maps of chromosome 4, 5, and 6 based on 94 DH
progeny from the cross of Shyri x Galena
17
Linkage maps of chromosome 7 based on 94 DH progeny
from the cross of Shyri x Galena
18
Average stripe-rust severity (%) in the DH progeny of
Shyri x Galena cross
22
SIM test statistic from the QTL analysis of stripe-rust severity (%)
on chromosome 2 and 3 in DH population of Shyri x Galena
24
SIM test statistic from the QTL analysis of stripe-rust severity (%)
on chromosome 5 and 6 in DH population of Shyri x Galena
24
Mean and 95% LSD intervals for the two locus interactions
between the QTLs on chromosomes 2 and 5 determining
stripe-rust severity (%)
27
Mean and 95% LSD intervals for the two locus interactions
between the QTLs on chromosomes 2 and 6 determining
stripe-rust severity (%)
28
2.7
Average leaf-rust index in the DH progeny of Shyri x Galena
30
2.8
SIM test statistic from the QTL analysis of leaf-rust index
on chromosome 1 in the DH population of Shyri x Galena and
showing the coincident QTL peak and position of
the Rphx, locus
31
Plant height reduction (cm) after infection with the
BYDV-MAV serotype in the DH progeny of Shyri x Galena
32
Dwarfness score (0-9) in the DH progeny of Shyri x Galena
infected with the BYDV-MAV serotype
33
Tillering score (0-9) in the DH progeny of Shyri x Galena
infected with BYDV-MAV serotype
33
2.2
2.3
2.4
2.5
2.6
2.9
2.10
2.11
LIST OF FIGURES (Continued)
Figure
2.12
2.13
2.14
2.15
3.1
3.2
3.3
Page
Dwarfness score (0-9) in the progeny of Shyri x Galena
infected with the BYDV-PAV serotype
34
Tillering score (0-9) in the progeny of Shyri x Galena
infected with the BYDV-PAV serotype
34
SIM test statistics from the QTL analysis of plant height
reduction (cm), dwarfness score, and tillering score after
infected with the BYDV-MAV and BYDV-PAV serotypes
on chromosome 1 in DH population of Shyri x Galena
35
SIM test statistics from the QTL analysis of plant height
reduction (cm), dwarfness score, and tillering score after
infection with the BYDV-MAV and BYDV-PAV serotypes
on chromosome 4 and 5 in DH population of
Shyri x Galena
36
Derivation of double haploid (DH) germplasm from
a marker-assisted selection program for adult plant stripe rust
resistance
51
Average stripe rust disease severity (%) in DH lines derived
from one cycle of marker-assisted backcrossing using BRB41
and Steptoe as the donor and recurrent parents, respectively
58
Percentage of donor parent genome in DH lines derived from
one cycle of marker assisted backcrossing as measured by
120 markers
63
LIST OF TABLES
Table
2.1
2.2
Page
Resistance Gene Analog (RGA) primers used to develop
Resistance Gene Analog Polymorphism (RGAP) markers
for the Shyri x Galena cross
15
Chromosomal location, allele phase, and effect (expressed as
percentage of phenotypic variance explained) for stripe
rust, leaf rust, and BYDV resistance loci in the Shyri x Galena
population
25
Percentage of phenotypic variance explained (R2p) in multilocus
models involving main effect QTLs and their interactions for stripe
rust and BYDV resistance in the Shyri x Galena population
29
Chromosome location, slope, p-value, and R2 for markers
significantly associated with stripe rust severity in single locus
regression. Chromosome locations in italics are putative
60
Markers with significant effects in multi-locus regression models
of stripe rust severity in individual environments and in the
average of four environments
61
Average stripe rust severities in 1995 and 1996 for the ten most
resistant and susceptible doubled haploid lines derived from one
cycle of marker assisted selection
62
the Rep
2.3
3.1
3.2
3.3
MAPPING AND INTROGRESSION OF DISEASE RESISTANCE GENES IN
BARLEY (Hordeum vulgare)
CHAPTER 1
INTRODUCTION
Molecular markers are tools for simultaneously advancing our understanding of
plant genome organization and increasing the efficiency of plant breeding. Quantitatively
and qualitatively inherited disease resistance genes are logical targets for applying
molecular marker technologies because these genes are of tremendous economic
importance. While recent advances have revealed commonalties in cloned qualitative
resistance genes, little is known about the structure and function of quantitative resistance
genes. However, quantitative resistance is attractive from the standpoint of probable
durability. Barley is an excellent model system for molecular marker analysis. This diploid
species has seven cytologically distinct chromosomes and a haploid genome size of
approximately 5 x 109 by (Bennet and Smith, 1976). The economic importance of the crop
makes it a model system of practical utility.
Molecular Markers
Molecular markers are abundant and phenotypically neutral. As a consequence,
molecular markers have been used in an array of organisms, including crop plants in
general, and barley in particular, for DNA fingerprinting (Bassam and Bentley 1994;
Becker et al. 1995), phylogenetic analysis (Saghai-Maroof et al. 1995), genome mapping
(Hayes et al. 1996), map-based cloning (Zhang et al. 1994; Tanksley et al. 1995; Dean
and Schmidt 1995; Kilian et al. 1997; Bennetzen and Freeling 1997), candidate gene
2
analysis (Touzet et al. 1995), and markerassisted breeding (Mazur and Tingey 1995;
Han et al. 1997; Toojinda et al. 1997). With particular reference to traits showing
complex inheritance, molecular markers can be used to dissect the genetic basis of
complex phenotypes into Mendelian components in order to obtain information about
gene dosage, epistasis, pleiotropy and genotype x environment interaction (Hayes et al.
1993; Bezant et al. 1997; Mather et al. 1997). Information regarding the number, effect,
and chromosomal location of genes determining a phenotype is of immediate practical
utility. This information can be used to design matings and to select target genotypes in
segregating progeny. In barley, QTL analysis has moved beyond the descriptive phase to
validation (Hayes et al. 1996) and use in cultivar development (Toojinda et al. 1997).
The four most common types of markers used in applied plant genomics in
general, and implemented in barley, are Restriction Fragment Length Polymorphisms
(RFLPs), Random Amplified Polymorphic DNAs (RAPDs), Amplified Fragment Length
Polymorphisms (AFLPs), and Simple Sequence Repeats (SSRs). RFLPs have been the
mainstay of North American Barley Genome Mapping Project (NABGMP) and other
maps (Kleinhofs A. 1991; Heun et al. 1991; Graner et al. 1991). RFLPs are based on the
differential hybridization of cloned DNA to DNA fragments in a sample of restriction
enzyme- digested DNA. The marker is specific to a single clone/restriction enzyme
combination. RFLPs are usually codominant and often multiallellic. They are excellent
framework markers, as they have locus identity. They are excellent markers for
comparative mapping, across divergent taxonomic groups, such as barley: wheat,
barley:rice, etc. However, RFLP are tedious and inefficient to assay.
3
RAPD markers are based on the differential PCR amplification of a random
sample of DNA with short oligonucleotide sequences. RAPDs are usually dominant.
The primary advantages of RAPDs are simplicity and cost, and have been used widely in
barley genome mapping (Barua et al. 1993; Giese et al. 1994; Noli et al. 1997; Dahleen et
al. 1997). However, these advantages can be outweighed by low reproducibility.
Furthermore, RAPDs do not have defined locus identity, and it can be difficult to relate
RAPD loci between different experimental populations of the same species.
AFLP markers are generated by a combination of restriction digestion and PCR
amplification. They are visually dominant, biallelelic and extremely high throughput.
AFLPs have been useful in developing several barley linkage maps (Becker et al. 1995;
Powell et al. 1997; Qi et al. 1998) and for whole genome screening in marker-assisted
selection experiments. A principal drawback to AFLPs is that the assay is time
consuming, as it is based on sequencing protocols. The issue of locus identity has been
explored with AFLPs (Waugh et al. 1997; Qi et al. 1998). However, identity needs to be
established on a case by case basis. An additional drawback to AFLPs is that they are
reported to cause map expansion (Becker et al. 1995). Linkage map expansion is usually
attributable to poor quality data.
SSR markers are a molecular marker based on the PCR amplification of tandem
repeats of one to six nucleotide motifs. The polymorphism among individuals is due to
the variation in the number of repeat units. SSR markers are codominant, often
multiallelic which allow the unambiguous identification of alleles. The multi-allelism
can be impressive (Maroof et al. 1994; Russell et al. 1997). SSRs are excellent
4
framework markers (Becker and Heun 1995; Liu et al. 1996) with locus identity. They
can be multiplexed to achieve higher throughput (Mitchell et al. 1997). Two
disadvantages of SSRs are the high cost of discovery and the lack of transferability across
genera and even distantly-related species (Roder et al. 1995).
Quantitative and Qualitative Disease Resistance
Disease resistance is commonly the result of a biochemical pathway from
perception to defense response (Beynon 1997). H.H. Flor (1955) demonstrated the
interaction of host and pathogen genes in determining resistance and susceptibility in what
is known as the gene-for-gene concept. J.E. Vanderplank introduced the ideas of horizontal
and vertical resistance (Thresh 1998). However, the genetic basis and relationship between
genes determining vertical (qualitative) resistance and those determining horizontal
(quantitative) resistance is still unclear. Advances in molecular biology are revealing that
many qualitative resistances following the gene-for-gene systems are actually quite
complex (Crute 1985; Jorgensen 1992).
On the other hand, only one or few loci have been shown to determine quantitative
resistance (Parlevliet 1978; Geiger and Heun 1989; Chen et al. 1994; Thomas et al. 1995).
The genes that are determinants of quantitative resistance, and quantitative phenotypes in
general, are termed quantitative trait loci (QTLs). QTL analysis has been used for
dissecting quantitative characters into discrete genetic loci (Tanksley 1993; Wang et al.
1994; Thomas et al. 1995; Powell et al. 1997) and for defining the roles of individual genes
in quantitative disease resistance (Hayes et al. 1996). To date, the primary outcomes of
QTL resistance mapping projects have been informed on the number and location of
principal determinants (Steffenson et al. 1996; Moharramipour et al. 1997). The
5
availability of reference linkage maps is allowing for systematic intra-specific mapping of
resistance genes and comparative mapping of resistance genes in related taxa (Young
1996). These efforts will ultimately lead to a clearer picture of gene number, relationship,
structure, and function.
With particular reference to qualitative resistance genes, genes that are induced in
response to pathogens have been identified in several host:pathogen systems (Ferreira et al.
1995; Lefebvre and Palloix 1996). In many cases, genes conferring resistance to different
specificities of the same pathogen, or resistance to different pathogens, are found in clusters
(Michelmore 1995; Ellis et al. 1998). This led to the hypothesis that some resistance genes
are related in function and evolution and that individual member of these multigene
families diverged to confer different specificities (Michelmore et al 1987; Pryor 1987). The
presence of common motifs among cloned genes the leucine rich repeat (LRR), nucleotide
binding sites (NBS), serine/threonine kinase- supports this hypothesis. Based on structure,
disease resistance genes can be classified into five groups: detoxifying enzymes, kinases,
NBS/LRR proteins, extracellular receptors and receptor kinases (Lamp 1994; Dangl 1995;
Martin 1996; Buschges et al. 1997).
Although many candidate disease resistance genes have been identified, it has
been difficult to demonstrate that these are truly defense genes (Bowles 1990). The
candidate gene approach combining genome mapping and sequence information to
integrate molecular analysis of host-pathogen interactions and gene mapping (Concibido
et al. 1996; Dahleen 1997; Han et al. 1997). The conserved motifs of cloned resistance
genes have been used to design degenerate primers known as "Resistance Gene Analogs"
(RGAs) for isolating unknown disease resistance gene homologous in different plant
6
taxa. The RGA approach has proven to be an efficient tool for isolating resistance genes
(Kanazin et al 1996; Yu et al 1996; Leister et al.1996; Marek and Shoemaker RC; Leister
et al.1998). The RGA approach can also be used to generate Resistance Gene Analog
Polymorphisms (RGAPs) for linkage mapping in populations (Chen et al.1998). Linkage
or coincidence of RGAPs with resistance genes would support the hypothesis that these
polymorphisms are revealing disease resistance loci. To date, these molecular tools have
not been focussed on quantitative disease resistance.
Molecular Breeding
Molecular markers can be used to increase the efficiency with which qualitative
and quantitative disease resistance genes are manipulated in breeding programs (Horvath
et al. 1995; Toojinda et al. 1997). Backcrossing, a technique that has been used
extensively to introgress disease resistance loci into adapted backgrounds, can be fraught
with linkage drag (Tanksley et al. 1989).
Molecular markers can increase the efficiency of the process in several ways.
Flanking markers can be used to identify the backcross lines that are heterozygous for
resistance loci regions. Advancing only these selected lines will also have the effect of
reducing linkage drag (Young and Tanksley 1989; Tanksley and Nelson 1996). Singlecopy, or low copy, markers with defined map locations, such as RFLPs, SSRs, and
RGAPs, are ideal for this step.
Molecular markers could also increase the efficiency of backcrossing by allowing
for selection of genotypes with maximum percentage of the recurrent parent genome.
Markers with higher information content per reaction, such as AFLPs and RAPDs, are
ideal for this step (Waugh et al. 1997). However, manipulation of QTLs can be
7
problematic due to loss of target loci through recombination, incorrect information
regarding the location of the QTLs, and/or negatively altered expression of the QTLs in
new genetic background (Hayes et al. 1996). Therefore, a marker-assisted QTL
backcrossing scheme might (1) use flanking markers to select progeny with a probability
of earring the target QTL allele (s), (2) confirm the target phenotype in the selected
progeny, and (3) use multiplex markers to identify those selections with the maximum
percentage of the recurrent parent genome.
A series of experiments were conducted to integrate molecular markers,
qualitative and quantitative disease resistance, and molecular breeding. These
experiments were conducted in the context of the very serious stripe rust problem facing
barley growers from the Pacific Northwest of the United States to the Andean region of
South America.
Quantitative resistance may be the best tool for dealing with stripe rust, caused by
Puccinia striiformis f. sp. hordei. The first chapter of this thesis describes mapping
qualitative and quantitative disease resistance loci using a range of molecular markers,
including RFLPs, AFLPs, SSRs, and RGAPs. The second chapter describes markerassisted introgression of QTLs determining stripe rust resistance.
8
CHAPTER 2
ASSOCIATION OF QUANTITATIVE AND QUALITATIVE DISEASE
RESISTANCE GENES IN BARLEY (Hordeum vulgare) WITH RESISTANCE
GENE ANALOG POLYMORPHISMS (RGAPs)
Theerayut Toojinda, Xianming Chen, Patrick M. Hayes, Hugo Vivar
A. Kleinhofs
9
Abstract
Stripe rust, leaf rust, and Barley Yellow Dwarf Virus (BYDV) are important
diseases of barley (Hordeum vulgare L). Using 94 doubled haploid lines (DH) from the
cross of Shyri x Galena, multiple disease phenotype data sets, and a 99-marker linkage
map, we determined the number, genome location, and effects of genes conferring
resistance to these diseases. We also demonstrated associations of these resistance genes
with Resistance Gene Analog Polymorphism (RGAP) loci based on degenerate motifs of
cloned disease resistance genes. Leaf rust resistance was determined by a single gene on
chromosome 1. QTLs on chromosomes 2, 3, 5, and 6 were the principal determinants of
resistance to stripe rust. Two-locus QTL interactions were significant determinants of
resistance to this disease. Resistance to the MAV and PAV serotypes of BYDV was
determined by coincident QTLs on chromosomes 1, 4, and 5. QTL interactions were not
significant for BYDV resistance. The associations of RGAPs with qualitative and
quantitative resistance loci provide a powerful tool for efficiently mapping disease
resistance genes. These relationships should be useful in answering fundamental
questions regarding quantitative and qualitative disease resistance genes.
Introduction
The genetic resistance-of plants to diseases is an area of intense and important
research activity. Genetic resistance is the most cost-effective and environmentally
appropriate approach to disease management. Plant breeders have, in general, made
excellent use of genetic resistance. Disease resistance breeding remains a principal
objective of many breeding programs. While genetic variation for disease resistance to
many diseases is still available within the cultivated germplasm pool of many crop species,
10
in many cases restricted genetic variance has led to searches for new resistance genes in
crop ancestors and relatives (Conner et al. 1989; Ordon and Friedt, 1993; Reuvein et al.
1997; Veremis et al. 1997). Introgression of exotic genes is an expensive and difficult
process (Hoffbeck et al. 1995; Tanksley and Nelson 1996). Resistance genes are,
accordingly, precious commodities. In this context, the durability of resistance is of great
importance. While durability can only be demonstrated in hindsight, theory and some
historical evidence support the contention that quantitative resistance is often more durable
than qualitative resistance (Browning et al. 1977; Johnson 1981; and Line 1993).
Qualitative resistance genes have been extensively studied in terms of genome
location (Giese et al. 1993; El-Kharbotly et al. 1994; Graner and Tekauz 1996), specificity
(Thomas et al. 1995; Ori et al. 1997), and most recently structure and function (Lamp 1994;
Martin 1996; Dangl 1995; Buschges et al. 1997). Until the recent development of
quantitative trait locus (QTL) analysis tools, the study of quantitative resistance genes
focused on biometrics and epidemiology. QTL tools allow for the systematic dissection of
quantitative resistance into estimates of locus number, location, effect, and interaction
(Michelmore 1995; Young 1996; Powell et al. 1997). Disease resistance QTLs have been
described for a number of host pathogen systems (Williamson et al. 1994; El-Kharbotly et
a. 1994; Maisonneuve et al. 1994; Zaitlin et al. 1993), including barley (Barua et al. 1993;
Giese et al. 1993; Graner and Bauer 1993; Chen et al. 1994; Hayes et al. 1996). However,
the structure and function of quantitative resistance genes is still a matter of conjecture.
They could represent the effects of alternative alleles at qualitative resistance loci
(Dingerdissen et al. 1996) or they could represent an entirely different class of genes (Pryor
and Ellis 1993).
11
Advances in molecular biology are revealing that many "simple" gene-for-gene
systems are actually quite complex (Crate, 1985; Jorgensen, 1992). QTL studies are
revealing in many systems that one or a few loci are principal determinants of trait
expression (Thomas et al. 1995; Chen et al. 1994). The availability of reference linkage
maps is allowing for systematic intra-specific mapping of resistance genes and comparative
mapping of resistance genes in related taxa, and these efforts will ultimately lead to a
clearer picture of gene number, relationship, structure, and function. Genes conferring
resistance to different specificities of the same pathogen, and to different pathogens, are
known to cluster in a range of plants (Saxena et al. 1968; Giese et al. 1981; Hooker 1985;
Paran et al. 1991; Michelmore 1995; Ellis et al. 1998). These clusters are particularly
dynamic regions of the genome, although the mechanisms leading to variation are still a
matter of debate (Borst and Greaves, 1987; Pryor 1987; Pryor and Ellis, 1993).
The discovery of common motifs in cloned resistance genes leucine rich repeats
(LRR), nucleotide binding sites (NBS), serine/threonine kinase has served as a basis for a
generalized approach to resistance gene analysis. Degenerate primers based on these
motifs can be used to amplify specific genomic DNA sequences known as resistance gene
analogs (RGAs). The RGA approach has proven to be an efficient tool for isolating
resistance genes (Kanazin et al 1996; Yu et al 1996; Leister et al 1996 ; Leister et al. 1998).
The technique can also be used to generate polymorphisms (RGAPs) for linkage mapping
in populations (Chen et al. 1998).
Qualitative and quantitative resistance to rust fungi (Puccinia sp.) has been an area
of extensive study in the Triticeae. In barley (Hordeum vulgare L.), there is an especially
rich literature on resistance to leaf rust (Puccinia hordei) (Alemayehu and Parlevliet 1996;
12
Feuerstein et al. 1990; Jin et al. 1996; Qi 1998). The impetus for this research was the fact
that race-specific qualitative resistance genes lacked durability to this pathogen of
worldwide importance. There is less information on the basis of genetic resistance to stripe
rust (Puccinia striiformis, fsp. hordei). Due to the relatively recent arrival and importance
of this pathogen in the Americas, we have been systematically mapping resistance in a
range of germplasm (Chen et al. 1994; Hayes et al. 1996). Barley Yellow Dwarf Virus
(BYDV) is an aphid-vectored luteovirus of worldwide importance (Harder and Harber
1992; D'Arcy 1995; Collin et al. 1996). These diseases tend to be episodic in response to
environmental conditions, but in any given environment at least one disease is usually a
principal production constraint. In many situations, the use of cultivars resistant to multiple
diseases is a necessity. Thus, the ideal variety should be resistant to multiple diseases,
assuming resistance genes per se do not have a cost. Development of such varieties can be
expedited by information on the number, location and effect of the determinants of
resistance.
We reasoned that if many disease resistance genes contain conserved motifs, then
the RGAP approach could serve as tool for efficient mapping of multiple resistance genes
in a single population. A portion of the total number of RGAPs scored in a linkage
mapping population should correspond to resistance genes. Furthermore, if quantitative
resistance genes are related to qualitative resistance genes, then resistance QTLs should be
detected in association RGAPs. To test these hypotheses, we mapped RGAPs in a doubled
haploid population of barley, together with a qualitative resistance gene conferring
resistance to leaf rust (caused by the fungus Puccinia hordei), QTLs associated with
resistance to stripe rust (caused by the fungus Puccinia striifarmis fsp. hordei), and QTLs
13
associated with resistance to two serotypes (MAV and PAV) of Barley Yellow Dwarf
Virus (BYDV).
Materials and Methods
Plant Material
One hundred doubled haploid (DH) lines were derived from the F1 of the cross
Shyri/Galena, using the Hordeum bulbosum technique, as described by Chen and Hayes
(1989). Shyri is a two-rowed feed barley developed by ICARDA/CIMMYT (Mexico) and
released by INIAP (Ecuador). Shyri is a source of resistance to a range of diseases,
including stripe rust, leaf rust, scald and net blotch. Galena is proprietary two-rowed
malting barley belonging to the Coors Brewing Company Inc.
Genotyping and Map Construction
Ninety-four of the DH lines were genotyped with 41 Restriction Fragment Length
Polymorphism (RFLP), 51 Simple Sequence Repeat (SSR), 562 Amplified Fragment
Length Polymorphism (AFLP), 144 Resistance Gene Analog Polymorphism (RGAP)
markers, and one morphological marker. The RFLP marker nomenclature follows that
employed by the North American Barley Genome Mapping Project (Kleinhofs et al., 1993;
Hayes et al., 1996). RFLPs were assayed as described by Chen et al. (1994) and Kleinhofs
et al. (1993). Two sources of SSRs were used: database-derived repeats (described by
Becker and Heun 1995; Liu et al. 1996) and repeats derived from an enriched genomic
library. Fifty-one SSR markers (designated as BMAC, EBMAC, HVM, HVC, HVP, HTT
plus an arbitrary number) were assayed as described by Morgante et al. (1994); Liu et al.
(1996) and Russell et al. (1997).
14
The AFLP assays were performed using 16 Pstl/Msel and 16 EcoRl/Msel primer
combinations as described by Zabeau and Vos (1993). A total of 562 AFLP markers
(designated as E M for EcoRI /Msel, T M for Pstl/Msel) were scored. The RGAP
markers were generated with a set of degenerate primers derived from resistance gene
homologs (Table 2.1) as described by Chen et al. (1998). PCR amplification was
performed in a Perkin Elmer 9600 thermal cycler. The reaction mixture and the
polyacrylamide gel electrophoresis procedures were as described by Liu et al (1996) and
Chen et al (1998). The morphological marker, rachilla hair length (mSrh), was scored
under a dissecting microscope.
The eight-hundred and nine markers were used for linkage map construction. The
base map (Figure 2.1) was constructed using a subset of the total markers to achieve a
target interval distance of 10-15 cM. The procedure for constructing the base map was as
follows. Markers in common with published maps were retained, when possible, in order
to facilitate map integration. Markers showing significant segregation distortion, markers
with missing observations, and markers causing map expansion were discarded. Linkage
analysis was performed on the remaining subset of 138 markers using Gmendel 3.0
(Hollaway and Knapp, 1994).
Linkage groups were first calculated using a maximum allowable recombination
percentage (rmax) of 0.25 and a LOD score of 7. Cosegregating and tightly linked markers
were then dropped. The markers with the most complete data were retained. Linkage
groups were then calculated using LOD 3.8 and rmax 0.35. Marker order was checked by
Monte Carlo and Bootstrap simulations, using annealing temperatures of 300 inner and 200
outer.
15
Table 2.1 Resistance Gene Analog (RGA) primers used to develop Resistance Gene
Analog Polymorphism (RGAPs) markers for the Shyri X Galena cross.
Primer'
Sequence (5' - 3')°
No. of
polymorphic
markers
LM637
LM638
ARIGCTARIGGIARICC
GGIGGIGTIGGIAAIACIAC
NBS -F1
GGAATGGGNGGNGTNGGNAARAC
NBS-R1
YCTAGTTGTRAYDATDAYYYTRC
NLRR-for
TAGGGCCTCTTGCATCGT
NLRR-rev
TATAAAAAGTGCCGGACT
NLRR-IN1
TGCTACGTTCTCCGGG
NLRR-IN2
TCAGGCCGTGAAAAATAT
Nkin2
GTAACTAAGGATAGA
Nploop
TCAATTAATGTTTGAGTTATTGTA
Ptokinl
GCATTGGAACAAGGTGAA
Ptokin3
TAGTTCGGACGTTTACAT
Ptokin2
AGGGGGACCACCACGTAG
Ptokin4
AGTGTCTTGTAGGGTATC
PtoFen-S
ATGGGAAGCAAGTATTCAAGGC
Pto Fen-AS
TTGGCACAAAATTCTCATCAAGC
RLK-for
GAYGTNAARCCIGARAA
RLK-rev
TCYGGYGCRATRTANCCNGGITGICC
RLRR-for
CGCAACCACTAGAGTAAC
RLRR-rev
ACACTGGTCCATGAGGTT
SI
GGTGGGGTTGGGAAGACAACG
AS3
IAGIGCIAGIGGIAGICC
S2
GGIGGIGTIGGIAAIACIAC
AS3
IAGIGCIAGIGGIAGICC
S2-INV
CAICAIAAIGGITGIGGIGG
AS3-INV
CCIGAIGGIGAICGIG
XLRR-for
CCGTTGGACAGGAAGGAG
XLRR-rev
CCCATAGACCGGACTGTT
XLRR-INV1
TTGTCAGGCCAGATACCC
XLRR-INV2
GAGGAAGGACAGGTTGCC
Total polymorphic markers
4
3
11
13
9
13
12
6
1
13
12
17
16
7
7
144
The primers LM637 and LM638 was designed by Kanazin et al. (1996). LM637 was based on a second region of amino acid
identity which in the RPS2 protien is proposed to reside in a transmembrane region and LM638 was based on the conserved Ploop sequence of genes N, RPS2, and L6. The primer pair, NBS Fl and NBS R1, were designed by Yu et al. (1996) based on
the amino acid sequences of two highly conserved motifs of the nucleotide-binding site in N and RPS2 genes. The primer
Nkin2 was based on the second kinase region and primer NPloop was based on the P-loop of the N gene (Naweed Naqvi,
IRRI). The primers Ptokinl, Ptokin2, Ptokin3, and Ptokin4 were designed based on the DNA sequence encoding for protein
kinase in the tomato Pto gene conferring resistance to Pseudomonas syringae pv tomato (Naweed Naqvi, IRRI). The
primers Pto/Fen-S and Pto/Fen-AS were based on a member of Pto gene family conferring sensitivity to fenthion (Leister et al.
1996). The primers Si, S2, AS1, and AS3 were designed by Leister et al. (1996) based on the resistance genes RPS2 of
Arabidopsis thaliana and N of tobacco. The primer pairs, RLK for and RLK rev, were designed by Feuillet et al. (1997) to
amplify serine/threonine kinase sequence subdomains II to VIII of the wheat Lr10 gene conferring resistance to Puccinia
recondita. The primer pairs, NLRR-for and NLRR-rev, RLRR-for and RLRR-rev, and XLRR-for and XLRR-rev were
designed based on leucine-rich repeat regions of genes RPS2, Xa21, N, and C19, respectively (Naweed Naqvi, IRRI). The
primers NLRR-INV1, NLRR-INV2, S2-INV, AS3-INV, XLRR-INV1, and XLRR-INV2 were designed based on the inverse
sequences of NLRR-for, NLRR-rev, S2, AS3, XLRR-for, and XLRR-rev, respectively. All primers were made by Operon
(Alameda, California, USA).
b Codes for mixed bases: Y = C/T, N = A/G/C/T, R. = A/G, and D = A/G/
16
The final map, consisting of 99 markers, is shown in Figure 2.1 and was used, together
with the phenotype data sets, for QTL analysis. The assignment of linkage groups to
chromosomes was based on markers in common with previously published maps (Heun et
al. 1991; Kleihofs et al. 1993; Qi et al. 1996).
Chromosome 1
Chrormsome 2
ABG704
Owormsonm 3
BCD307
F1R2-5
ABO:68
5.8
RoFc6-3
EACIvb2f
7.7
223
HvIAC1861
XLRF62
FLFR-10
288
>21:12112-3
XLRRIN4
H1PFB
S2AS31N.5
271
144:3889
HvI.C222a
F2121.1
113
789
1-1 KRIN40
14 CMA
123
BMAC187
15.7
122
BMA054
31.8
E42M17e
1F21.37/8-4
87
S1AS3-16
S2AS311440
EBMAC525
Ra/3-1
BMAC144f
S2AS311,L1
SR
HVIvE0
FK1/3-9
KRIM
3).1
P104
20.12-5
N.R8.8
M.28.7
E33M12e
14112..7
224
4.8
2U1145
200
KFP190
BYDV
BMaC144a
10.0
11.1
ABC253
1118.2
RLR4
S2ASIN-3
106
E44M3611
RON
FK13-10
142
RI.R84
B88431,11
82122.7
13346
HVMD
RI.Rfr-a
17.8
1.6/1537/8-1
S2A..811144
R.LK-2
184
EHMAC539b
HV1AC218a
SIAS3-7
DAK642
/4Bt-8
RIAS3-10
SI-1A.%41
SlAS3-L2
81RIR-1
122M521(
<1 F1R24
FLRIN3
SR
BYDV
5.3
31
14.3
19.8
S2AS31N-11
LR
-2
HVM54
BCD135
87
12YR-3
HVM49
Thal
HVPRP
10.5
7k0
22 =
1
FIC2146
8441/4-5
IMT3G
10.7
XLRRFR-1
BG1Zb
203.3 cM
233 cM
193.1 cM
Figure 2.1a Linkage maps of chromosome 1, 2, and 3 based on 94 doubled haploid (DH)
progeny from the cross of Shyri x Galena. Marker loci indicated on the left
side of each map were used for linkage map construction. Distances are in
Kosambi cM units. Markers on the right side of each linkage group are
RGAPs mapping to the region within the corresponding brackets. QTLs
for stripe rust (SR), leaf rust (LR), and barley yellow dwarf virus (BYDV)
resistance are shown "bold".
17
Clinansotre4
2440334
Chumsonr5
F2R26
Ilv137/8-3
=8
33.7
MEV
MAG137
Ea44C213
145
32
5.6
-/
_\
thrusososre6
)1142H7.3
MA440
1,1R1N6
N_RR.45
WINS
98
N3-1
IWW°
N3SM1
43
EM4C316
ASS-7
XLM1R2
15.4
amorx
OX1542
1103 -2
2L4
S2AS3-9
S2AS3-8
HV1VB
N(19
B3MCh 11.8
139/3.6
P102
55
1\111N7
PKV3-3
13M4C30
106
165
106
AEC160
N.111N1
E42M34o
21.5
S2ASMN6
80
STAMM
KsAlH2
F1R26
ECRU
S2ASV12
line
S1AS33
HVM74
111126-3
19.2
/4111N9
147
RIM
1L6
WIVES
MAG552A
34 42rz 80
T22Ivf2f
Parent
98
SIA.S3-9
BM/
FM-3
143
N.BRO
68
ER2-3
131 218b 69
66
P2R2-5
HEM
BIZvB4d
RE1W-10
FIRM
PIOS
FA44C3C8z1
A1EG472
167
BM4C1446 9'1
/4.112.-6
119/3-7
as
S1AS36
Rcan-1
PK2/4-1
121
206
KFF221
F19/3-11
17.5
111414
HvHVA1
35.0
IsKR2
AF3397
15.1
NLBP.-4
360
SMIN13
aryl
SYS3-17
S2AS34
IvINGNS
SASS
SZAS3B415
81
R1.RFR.6
T2Z 1vP4g
Ficy4-11
SZAS34
RLRE-12
45
S2AS3-3
RI11-11
Ftfen4
120
AE 387
1662dvl
1769cM
1427cM
Figure 2.1b Linkage maps of chromosome 4, 5, and 6 based on 94 DH progeny from the
cross of Shyri x Galena. Marker loci indicated on the left side of each map
were used for linkage map construction. Distances are in Kosambi cM
units. Markers on the right side of each linkage group are RGAPs mapping
to the region within the corresponding brackets. QTLs for stripe rust (SR),
leaf rust (LR), and barley yellow dwarf virus (BYDV) resistance are shown
"bold".
18
Chromosome 7
HMSO
21.2
F2R2-2
21.8
XLRAFR-7
/4111N-4
ABC302
5.4
BCD298
199
mErh
F2R2-4-1
7.8
ABG59
24.4
DEMAC539a
HVEHN7
1VCEIN9
9.5
RIREC2a
24.1
N.RIN12
£3354124
25.3
H.F.114-12
24.F -3
N.1114.13
2321%641
10.8
S2ARD3-2
243
201.3 cM
Figure 2.1c Linkage maps of chromosome 7 based on 94 DH progeny from the cross of
Shyri x Galena. Marker loci indicated on the left side of each map were used
for linkage map construction. Distances are inKosambi cM units. Markers
on the right side of each cartoon are RGAPs mapping to the region within the
corresponding brackets.
Disease Resistance Phenotyping
Stripe Rust
The 94 DH lines and parents were assessed for adult plant resistance in four tests at
Toluca, Mexico and one test at Celaya, Mexico. At Celaya, the DH lines and parents were
grown in uni-replicate hill plots in the winter of 1995. A field epidemic was initiated by
19
inoculating spreader rows (formed from a mixture of 15 extremely susceptible genotypes)
with a stripe rust isolate whose virulence pattern corresponds to the race 24 VarundaMazurka type described by Dubin and Stubbs (1986).
Stripe rust severity was rated at DGS59 (Feekes stage 10.5) as percent severity on
a plot basis by the modified Cobb Scale (Melchers and Parker, 1992) and reaction type
was scored using a 0 to 9 scale (0= immune, 9 = susceptible). At Toluca, the DH lines
and the parents were planted in one-row, 3-m plots in 1994, 1995 and at two planting
dates in 1996. Spreader rows, planted at 5.25-m intervals and consisting of a mixture of
15 susceptible genotypes, were inoculated twice with infected plants placed in the
foliage, and with applications of spores suspended in oil. Infected plants and spores were
collected locally. The race composition of this inoculum was not determined. Stripe rust
was rated as percent severity on a plot basis.
Leaf Rust
The 94 DH lines and parents were assessed for adult plant leaf rust resistance in
field tests at Ciudad Obregon, Mexico in 1994, 1995 and 1996 using two-row, 3-m plots.
Epidemics were initiated by inoculating spreader rows, a mixture of several very
susceptible lines, with fresh spores of a mixture of leaf rust races 8, 19, and 30 collected
from a susceptible variety. The inoculum was applied by two methods: (1) a water and
surfactant suspension was injected by syringe into the stems of boot-stage plants in
spreader rows and (2) a talc carrier was blown into the nursery with a backpack duster
multiple times during the growing cycle. Disease was rated by infection type - resistant
(R), moderately resistant (MR), moderately susceptible (MS) and susceptible (S) - and
percent severity on a plot basis. For mapping purposes, a rust index was calculated by
20
equating the infection type rating with a numerical score (R =1, MR = 2, MS = 3 and S = 4)
and multiplying this value by the percentage severity.
Bark, Yellow Dwarf Virus (BYDV)
The population and parents were assessed for resistance to the BYDV-MAV-Mex
and BYDV-PAV-Mex serotypes. In all tests, one replication of two row, 1-m plots was
used. The infected plots were infested twice with 5-10 aphids harboring the MAV and PAV
serotypes respectively, at three weeks intervals. Aphids were killed by applying an
insecticide one week after the second inoculation. Control plots were chemically protected
against aphid attack during the entire growing cycle. MAV resistance was described with a
dwarfness score (0-9) and a tillering score (0-9) in four tests at Toluca, Mexico (two
planting dates in each of 1995 and 1996). In 1996, plant height reduction (control -
infected) was also calculated. PAV resistance was measured in one test at Toluca in 1996.
The same dwarfing and tillering scores used for MAV were used to measure PAV
resistance.
OTL Analysis
QTLs were mapped using the interval mapping (SIM) and simplified composite
interval mapping (sCIM) procedures of MQTL (Tinker and Mather, 1995) and regression
procedures. For MQTL, each data set was analyzed with 1,000 permutations, a 5-cM walk
speed, and a Type I error rate of 5%.
For sCIM, eighteen background markers with approximately even spacing were
specified, with a maximum of three background markers per linkage group. Approximate
estimates of heritability were computed by substituting environments for replications as:
21
H2
a2
,
2
g/ 15
62e/r where 62g is the variance among DH lines, 62e is the error variance
among DH lines, and r is the number of environments.
Results and Discussion
Linkage Map Construction
The base map of 1316.5 cM is shown in Figure 2.1. Single linkage groups were
assigned to each chromosome, except for chromosomes 3 and 7. There are also gaps on the
long arm of chromosomes 3; the long and short arms of 7; and the long arm of
chromosomes 5 and 6. Locus ordering and distance are in agreement with published maps
(Heun et al. 1991; Graner et al. 1991; Kleinhofs et al. 1993; Qi et al. 1996).
Based on comparisons of markers in common with the Steptoe/Morex map
(Kleinhofs et al. 1993) and the merged barley map (Qi et al. 1996), the gap on the short arm
of chromosome 7 is approximately 40 cM long, and the gap on chromosome 3 is
approximatly 39 cM long. Telomeric RFLP loci on chromosome 5 and 6 (ABG387 and
MWG798, respectively) were not associated with the corresponding linkage groups. Twolocus recombination distances between these telomeric markers and nearest markers
(HvHVA1 and PK2/4-1) were 36 and 35 cM respectively.
RFLPs and SSRs served as anchor markers and the higher-throughput AFLPs and
RGAPs were used to fill gaps. However, only 16 AFLPs are included in the final base
map, due to AFLP clustering, a phenomenon reported in the literature (Becker et al. 1995).
Most (87%) of the RGAP markers followed expected segregation ratios. One-hundred and
ten RGAPs mapped to high density (<5 cM) clusters and were therefore excluded from the
base-map. Therefore, only 12 RGAPs are included. Markers used for base-map
22
construction are shown on the left-hand side of each linkage group, while additional
RGAPs are shown on the right hand side of each linkage group (Figure 2.1a, 2.1b, 2.1c
QTL Detection
The phenotypic distribution of stripe rust severity, averaged over the five
environments, did not show discrete classes allowing for Mendelian analysis (Figure 2.2).
Similar distributions were observed for each of the individual environments (data not
shown). Only five of the DH lines were as resistant as Shyri (1.5 + 0.9%) while there were
33 susceptible transgressive segregants with severities higher than Galena (62.5±7.4).
These data suggest that unique configurations of multiple alleles may be required for high
levels of resistance and that the susceptible parent has some resistance alleles.
20
16
0
-o
12
8
4
1
0
20
40
60
80
100
Disease severity (% )
Figure 2.2 Average stripe-rust disease severity (%) in the DH progeny of Shyri x Galena
Cross
The consistency of the disease severity ratings (h2= 94%) confirms that resistance
was stable in the face of the stripe rust virulence present during the years these tests were
conducted. While we do not have extensive data on pathogen virulence in these tests,
studies in the U.S.A. have shown considerable variation in stripe rust populations (Chen et
23
al. 1995). We have observed comparable levels of resistance in Shyri in over ten years of
testing at multiple locations throughout the Americas (data not shown). This may be
preliminary evidence for durability of stripe rust resistance.
The QTL data confirm the multi-locus control of stripe rust resistance in Shyri and
provide some evidence for a resistance allele in Galena. Shyri contributed resistance alleles
at QTLs on chromosomes 3, 5, and 6, while Galena contributed the resistance allele at a
QTL on chromosome 2 that approached, but did not reach, the significance threshold
(Figures 2.3 and 2.4). The largest-effect QTL on the short arm of chromosome 5 (act8BMAC213) was significant in all five tests.
This QTL accounted for 28% to 50% of the variation in phenotypic expression
(PVE) in the individual environments and 47% PVE in the average of the five tests (Table
2.2). We mapped this QTL to approximately the same position as the Yr4 locus (von
Wettstein-Knowles, 1992) and a stripe rust resistance QTL reported by Thomas et al.
(1995).
The relationship of the Shyri chromosome 5 QTL to these other loci remains to be
determined, although the Yr4 locus is reported to confer resistance to race 23, while the
virulence in the Americas is broadly defined as race 24 (Marshall and Sutton 1995; Chen
et al. 1995). The QTLs on chromosomes 3 and 6 mapped to the ABG004-T22M3 1 i and
KsuAl H2-Linka intervals, respectively. The QTL main effect on chromosome 3 was
significant only in the Celaya data set, while the main effect on chromosome 6 was
significant only in the Toluca 1994 and Toluca 1995 data sets.
24
Chromosome 2
Chromosome 3
Environments
- Cclaya 1995
----- Toluca 1994
Toluca 1995
10
----- Toluca 1996-1
Toluca 1996-2
E42.111470-EBMAC525
ABG004-T22M3
Figure 2.3 SIM test statistics from the QTL analysis of stripe-rust severity (%) on
chromosomes 2 and 3 in DH population of Shyri x Galena. QTL peak
intervals are shown on the x-axis. The horizontal bar indicates the maximum
significant threshold (P = 0.05).
Environments
70
- Cclaya 1995
60
Toluca 1994
50
Toluca 1995
Toluca 1996-1
40
Toluca 1996-2
F.
30
20
10
acte-BMAC213
Kso A 1K2 -Links
Figure 2.4 SIM test statistics from the QTL analysis of stripe-rust severity (%) on
chromosomes 5 and 6 in DH population of Shyri x Galena. QTL peak
intervals are shown on the x-axis. The horizontal bar indicates the maximum
significant threshold (P = 0.05).
Table 2.2 Chromosome location, allele phase, and effect (expressed as the R2p - percentage of phenotypic variance explained) for
stripe rust, leaf rust, and BYDV resistance loci in the Shyri x Galena population.
Resistance
Phenotype
Environments or
Measurments
Type of
inheritance
Chromosome
Stripe rust
Celaya 1995
Toluca 1996-1
Celaya 1995
All tests
Toluca 1994
Toluca 1995
All tests
Height reduction
Dwarfness score
Tillering score
Height reduction
Height reduction
Dwarfness score
Height reduction
Dwarfness score
Tillering score
QTL
2
E42M47e-EBMAC525
QTL
QTL
QTL
3
Stripe rust
Stripe rust
Stripe rust
Leaf rust
MAV
MAV, PAV
MAV, PAV
MAV
MAV
MAV, PAV
MAV
MAV, PAV
MAV, PAV
Qualitative
QTL
Closest RGAP
Resistance allele
R2p
NLFR4
G+
12%,
ABG004-T22M3 I i
act8-BMAC213
KsuA 1H2-Linka
XLRR-5
S2AS3-7
XLRRFR3,
NLR1N6
S
S
S
12%
Rphxs
ABC253-HVM49
S2AS31N- 1 I
S2AS31N-11
marker
5
6
I
1
,,
QTL
QTL
Marker interval
48++
4b
QTL
5
If
II
,,
9%
S
G
If
KFP221-ABG397
MWG634-CD0542
ABC160-BMAC303d
NKP-2
F2R2-6
G
G
RLRFR- I 0
S
PP
1>
PP
+ "G" refers to resistance allele from Galena and "S" refers resistance allele from to Shyri
++ 4a and 4b denote the two distinct QTLs on the short and long arms of chromosome 4, respectively
,,
,,
28%-50%
23%,
15%
91-96%
18%
20%, 11%
22%, 11%
10%
7%
7%, 6%
15%
13%, 3%
10%, 7%
26
The individual effects of these QTLs were much smaller than the chromosome 5
QTL. The maximum PVE's for the chromosome 3 and 6 QTLs were 12% and 23%
respectively. The relationships of these QTLs to the chromosome 3 QTLs reported by
Toojinda et al. (1998) and the Yr4 locus on chromosome 6B of wheat (Triticum aestivum)
reported by Chen et al. (1995) will need to be resolved by comparative mapping. Galena
contributed the resistance allele at a QTL that approached the significance threshold on
chromosome 2 (E42M47e-EBMAC525). This trend was observed in only two of the data
sets (Celaya 1995 and Toluca 1996-1). The significant main-effect QTLs, when
considered in multi-locus models, accounted for 39-52% and 41-55% of the variation in
phenotypic and genotypic expression, respectively. These QTLs should be of
considerable value and utility, as they are different from those mapped on chromosomes
4 and 7 by Chen et al. (1994) and Hayes et al. (1996).
The genetic variance that is not accounted for by the QTLs have mapped in this
population could be due to additional QTLs with small-effects and/or to QTL interactions
(epistasis). Some of these QTLs could be located in regions of the genome not covered by
our current linkage maps. As pointed out by Melchinger et al. (1998), our estimates of
QTL effect may be biased by the use of the same population for mapping and estimating
QTL effects. These authors also point out that larger populations are required for
estimating higher order QTL x QTL interactions. Recognizing the limitations of our
population of 94 DH lines, we did, however, proceed to test for interactions between
significant or nearly significant, main effects QTLs.
Significant interactions between the chromosome 2 and chromosome 5 QTLs, and
between the chromosome 2 and chromosome 6 QTLs, were detected in all five
27
environments. The PVE's for the QTLch2 x QTLch5 interaction, after accounting for QTL
main effects, ranged from 3 to 12 % in the analysis of individual environment data. In this
interaction, the source of the allele on chromosome 2 was not important, but Shyri always
contributed the resistance allele on chromosome 5.
80
60
40
20
0
SS
SG
GS
GG
Two locus interaction of QTL and QTLch5
Figure 2.5 Mean and 95% LSD intervals for the two locus interactions between
the QTLs on chromosomes 2 and 5 determining stripe-rust severity
(%). SS, SG, GS, and GG refer to the allelic composition of DH lines at QTLs on
chromosomes 2 and 5, respectively.
The SQTLch2 x SQTLch5 and Gotch2 x SQ1Lch5 phenotypes showed the same resistance
levels (Figure 2.5). The PVE for the QTL62 x QTLch6 interaction, after accounting for
QTL main effects, was 4%. In the case of this interaction, Galena contributed the
resistance allele on chromosome 2 and Shyri contributed the resistance allele on
chromosome 6. The average disease severity of the GQTheh2 x SQTLch6 genotypes was 28%
28
(Figure 2.6). The significant main-effect QTLs and their two locus interactions, when
considered in multi-locus models accounted for 44-59% and 47-63% of the variation in
phenotypic and genotypic expression, respectively (Table 2.3).
80
04
60
4/1
40
20
0
SS
SG
GS
GG
Two locus interaction of QTLch, and QTL,s6
Figure 2.6 Mean and 95% LSD intervals for the two locus interactions between
the QTL on chromosomes 2 and 6 for stripe-rust severity (%). SS, SG,
GS, and GG refer to the allelic composition of DH lines at QTLs on
chromosomes 2 and 6, respectively.
As shown in Figure 2.1 and Table 2.2, RGAP markers mapped in proximity with
the large-effect QTL on chromosome 5, the chromosome 6 QTL, and the region on
chromosome 2 where the susceptible parent (Galena) may have a resistance allele. The
NLFR4 locus mapped to the region on chromosome 2.
The XLRR-5 locus, 9 cM distal to T22M31i, mapped near the resistance QTL on
chromosome 3. The S2AS3-7 locus co-segregated with act8, which flanked the
chromosome 5 resistance QTL. The QTL peak on chromosome 6 mapped to the KsuAl H2NLRIN9 interval. These associations of RGAPs and stripe rust resistance QTLs have
immediate utility for marker assisted selection and germplasm characterization.
29
Table 2.3 Percentage of phenotypic variance explained (R2p) in multilocus models
involving main effect QTLs and their interactions (if significant) for stripe
rust and BYDV resistance in the Shyri x Galena population.
Disease resistance
phenotype
Stripe rust
Environment
Main effect QTLs
Toluca 1994
QTLchs
Two locus interaction
(P<0.01)
QT1,,h2xgrLas
Stripe rust
Celaya 1995
QTLch3, QT1,65
QTLch2xQTLa6
QTL,h2xQT1,65
QTLch2xQIIch6
Stripe rust
Toluca 1995
QTLchs
QT1,,h2xQTLchs
Multilocus R2
55°10
44%
52%
QTL,h2xgrLch6
Stripe rust
Toluca 1996-1
QTLchs
QT1,c1,2xqn-Ths
59%
QTLch2xQnch6
Stripe rust
Stripe rust
Toluca 1996-2
Average
QTL,h5
QT1,62xQlichs
53%
QTLchs
QTLch2xQTLch6
QT1,,h2xQTLchs
QTLch2xQT1,66
61%
BYDV-MAV
Toluca 1996
QTLchI, QTLch4a ,
Height reduction
QTLchab, QTLchs
BYDV-MAV
Toluca 1996
QTLchi, glichab,
Dwarfness score
QT1,65
BYDV-MAV
Toluca 1996
QTLchi, QTLchs
Tillering score
BYDV-PAV
Toluca 1996
QTLchi, Q11.-ch4b,
Dwarfness score
QTLchs
BYDV-PAV
Toluca 1996
QTLchi, QTLchs
Tillering score
ch = chromosome and #, see Table 2 for details
++4a, 4b refer to the KFP221-ABG397 and MWG634-CD0542 intervals, respectively
43%
39%
32%
18%
17%
Assuming a subset of the RGAPs correspond to disease resistance genes (Kanazin et al
1996; Yu et al 1996; Leister et al 1996; Leister et al. 1998; Chen et al. 1998), it is intriguing
to find qualitative and quantitative resistance genes in proximity.
In contrast to stripe rust resistance, leaf rust resistance in this population is clearly
qualitatively inherited. The phenotypic distribution for the leaf rust severity index,
averaged over the three environments, is discrete (Figure 2.7). This 1:1 ratio (chi square =
0.68 with P-value < 0.01) is clear evidence for monogenic inheritance. The two parents lie
at opposite ends of the frequency distribution. The heritability of the rust severity index
30
was 97%. When the quantitative severity index data were mapped using the procedures of
MQTL, a single QTL mapped to the long arm of chromosome 1 (7H) in the ABC253-Tha2
interval. Shyri contributed the resistance allele.
50
ca
40
30
20
10
0
50
100
150
200
250
300
Leaf rust index
Figure 2.7 Average leaf-rust index in the DH progeny of Shyri x Galena.
The single locus accounted for 84 % of the PVE. When the data were treated as
bivariate scores (1,0), the single locus (Rphxs), mapped 3.2 cM proximal to That (Figure
2.8). Rphxs may be allelic with the Rph3 locus reported by Jin et al (1993) and the Rphxc
locus reported by Hayes et al (1996). Rph genes have been used extensively in barley
breeding programs, but these genes often lack durability (Feuerstein et al. 1990; Jin et al.
1996; Qi 1998).
31
400
Environments
- Ciudad
0 brcgon 1994
---- Ciudad
0 bregon 1995
Ciudad
0 brcgon 1996
Figure 2.8 SIM test statistics from the QTL analysis of leaf-rust index on
chromosome 1 in DH population of Shyri x Galena and showing the
coincident QTL peak and position of the Rphxs locus. The horizontal
bar indicates the maximum significant thresholds (P = 0.05).
When Shyri was released in Ecuador in 1989, it was resistant to leaf rust, with a
maximum rating of 20MS. It now has ratings as high 90S (Vivar personal
communication), underscoring the danger of reliance on single qualitative resistance genes.
The RGAP marker S2AS3IN-11 was 5.8 cM proximal to the Rphxs locus. Thus, RGAP
markers were found in association with both quantitatively and qualitatively inherited
genes conferring resistance to Puccinia species. We also found RGAP markers in
association with quantitatively inherited resistance to two serotypes of the aphid-vectored
viral pathogen, BYDV. The phenotypic distributions for BYDV-MAV and BYD-PAV did
not show discrete classes for any of the three traits used to measure resistance - plant height
reduction, dwarfing score or tillering score (Figures 2.9-2.13).
32
Shyri
Galena
30
25
20
15 7
10
5
0
0
10
20
30
40
Plant H eight Reduction (cm )
Figure 2.9 Plant height reduction (cm) after infection with the BYDV-MAV serotype
in the DH progeny of Shyri x Galena.
These distributions suggest that resistance in this population is not due to the Ryd2
locus on chromosome 3, which gives a clear distribution of resistance vs. susceptible
classes when resistant and susceptible alleles are segregating in a DH mapping population
(Hayes et al. 1996). There were large numbers of positive phenotypic transgressive
segregants for all measures of resistance, suggesting that both parents contributed
resistance alleles.
The QTL data support the presence of resistance genes other than Ryd2, the
contribution of resistance alleles from both parents, a common genetic basis for the three
measures of resistance, and a common basis of resistance to the two serotypes.
QTLs exceeded significance threshold only in the Toluca 96 date 1 data set. Although
trends were apparent in the other data sets, the following discussion is based on 1
environment of MAV and PAV data. Four QTLs were detected for BYDV-MAV
resistance and three for PAV resistance. None mapped to the centromeric region of
chromosome 3, the site of the Ryd2 locus (Collins et al. 1996).
33
Shyri Galena
50
40
30
20
10
0
0
1
2
3
4
5
Dwarfness Score ( 0-9 )
Figure 2.10 Dwarfness score (0-9) in the DH progeny of Shyri x Galena infected
with the BYDV-MAV serotype.
50
40
30
20
z
10
0
0
2
3
4
5
Tillering Score ( 0-9 )
Figure 2.11 Tillering score (0-9) in the DH progeny of Shyri x Galena infected with
BYDV-MAV serotype
34
Shyri
G alena
40
30
0
.0
£
20
0
z
10
0
0
2
4
6
10
8
Dwarfness Score (0-9)
.Figure 2.12 Dwarfness score (0-9) in the progeny of Shyri x Galena infected with
the BYDV-PAV serotype.
alen a
24
S h y ri
20
16
.0
z
8
4
0
0
2
4
6
T illering Score (0-9)
10
Figure 2.13 Tillering score (0-9) in the progeny of Shyri x Galena infected with
the BYDV-PAV serotype.
Coincident large-effect QTLs for all three measures of MAV and the two measures of PAV
resistance mapped to chromosome 1 in the ABC253 -HVM49 interval (Figure 2.14).
35
Phenotypes
3
M AY- height
25
redaction
M A V -dw arta ess
store
20
scare
----- PA V -dw arfaese
15
111
BC 233- H V31 49
Figure 2.14 SIM test statistics from the QTL analysis of plant height reduction
(cm), dwarfness score, and filleting score after infected with the
BYDV-MAV and BYDV-PAV serotypes on chromosome 1 in DH
population of Shyri x Galena. QTL peak intervals are shown on the
x-axis. The horizontal bar indicates the maximum significant
thresholds (P = 0.05).
In all cases Galena contributed the resistance allele. The PVE values for MAV
resistance ranged from 18% to 21% (Table 2.2). The PVEs for the two measures of PAV
resistance were 11%. Two regions on chromosome 4 were associated with MAV resistance
and one with PAV resistance. In all cases Galena contributed the resistance allele. A MAV
dwarfing score QTL mapped to KFP221-ABG397 interval (PVE = 10%). Plant height
reduction and dwarfing score QTLs for MAV and a PAV dwarfing score QTL coincided in
the MWG634-CD0542 interval. The MAV resistance QTLs had PVE's of 7%, while the
PAV QTL had a PVE of 6% (Table 2.2). Shyri contributed resistance alleles for all
measures of resistance to the two serotypes at coincident QTLs on chromosome 5 in the
ABC160-BM_AC303d interval (Figure 2.15).
36
20
Phenotypes
1
Chromosome 4
Chromosome 5
M AV-height
reduction
MAV - dwarfness
15
score
../. \
M AV- tillering
r,..
10
\
---___ PAV- tillering
score
..../
%
....
VA,
f-N..-ki ,\
I.
0
_____. PA Y-dw arfness
score
1
\I
5
score
1.-,.
.
.
i ;71i
.. A \
'
',.
\
\s,...../7
'-';',1:
....., 1/
\-:.
:
ABC 160-BMAC303d
Figure 2.15 SIM test statistics from the QTL analysis of plant height reduction
(cm), dwarfness score, and tillering score after infected with the
BYDV-MAV and BYDV-PAV serotypes on chromosomes 4 and 5 in
DH population of Shyri x Galena. QTL peak intervals are shown on
the x-axis. The horizontal bar indicates the maximum significant
thresholds (P = 0.05).
The MAV resistance QTLs accounted for PVEs ranging from 10 - 15%, while the
PAV resistance QTL accounted for 3 - 7% of the PVE. The multi-locus PVE values for
MAV dwarfing score, plant height reduction, and tillering score were 43, 39, and 32%,
respectively. The multi-locus PVE values for PAV dwarfing score and tillering score were
17% and 18%, respectively (Table 2.3). No two-locus QTL interactions were significant
for resistance to either serotype, indicating that the phenotypic variance that remains
unaccounted for may be due to genes in unmapped regions of the genome or to higher
order interactions.
The chromosome 1 BYDV-MAV/PAV resistance QTL may be homoeologous
with Bdvl, a BYDV resistance gene in wheat. This gene, mapped to chromosome 7D,
37
was reported to be tightly linked to the Lr34 and Yr18 genes for resistance to leaf rust and
stripe rust, respectively (Singh, 1993; Sharma et al 1995). Barley chromosome 1 (7H)
and wheat chromosome 7 are homoeologous. In this population, the BYDV resistance
QTL and the Rphxs loci are linked, although in repulsion phase. Additional evidence for
homoeology of these resistance clusters is the presence of a qualitative stripe rust
resistance gene in the same region of the genome in CI10587 (Hayes personal
communication).
As in the cases of qualitative and quantitative resistance to fungal pathogens, we
found RGAPs associated with quantitative resistance to the viral BYDV pathogen. As
described in the preceding section on leaf rust resistance, two RGAPs mapped to the
same region on chromosome 1 as the BYDV resistance QTL. RGAP clusters mapped to
the same regions of the genome as both BYDV resistance QTLs on chromosome 4. The
RLRFRJO mapped distal to the chromosome 5 BYDV resistance QTL.
In summary, we mapped a qualitatively inherited gene conferring resistance to a
fungal disease (leaf rust) and determinants of quantitative resistance to a fungal disease
(stripe rust) and a viral pathogen (BYDV). The leaf rust resistance gene may be allelic
with a gene mapped in other germplasm (Jin et al. 1993; Hayes et al. 1996), and its
linkage relationship with a BYDV resistance QTL mirrors a homoeologous relationship
in wheat (Singh, 1993; Sharma et al. 1995).
The largest-effect stripe rust resistance QTL coincides with a previously reported
QTL in other germplasm (Thomas et al. 1995) and a qualitative resistance gene (von
Wettstein-Knowles 1992). We have demonstrated linkage relationships of RGAPs with
genes conferring qualitative and quantitative resistance to two fungal pathogens and
38
quantitative resistance to a viral pathogen. RGAPs are also linked with QTLs for
resistance to two other fungal pathogens - scald (caused by Rhyncosporium secalis) and
net blotch (caused by Pyrenophora teres) in this same population, as will be shown in a
subsequent report. RGAPs are clearly a powerful tool for mapping qualitative and
quantitative disease resistance genes. These linkage relationships support clustering of
disease resistance genes in barley, as reported in other species (Saxena et al. 1968; Giese
et al. 1981; Hooker 1985; Paran et al. 1991; Ellis et al. 1998) and may also be evidence
for commonalties of qualitative and quantitative resistance genes. The availability of
abundant RGAP markers should facilitate isolation and characterization of both
qualitative and quantitative resistance and provide tools for answering fundamental
questions regarding the genetic basis of the two classes of resistance.
References
Alemayehu F, Parlevliet JE (1996) Variation for resistance to Puccinia hordei in
Ethiopian barley landraces. Euphytica 90:365-370.
Barua UM, Chalmers KJ, Hackett CA, Thomas WTB, Powell W, Waugh R (1993)
Identification of RAPD markers linked to a Phynchosporium secalis resistance
locus in barley using near-isogenic lines and bulked segregant analysis.
Heredity 71:177-184.
Becker J, Heun M (1995) Barley microsatellites: allele variation and mapping.
Plant Mol Biol 27:835-845.
Becker J, Vos P, Kuiper M, Salamini F, Huen M (1995) Combined mapping of
AFLP and RFLP markers in barley. Mol. Gen. Genet. 249:65-73.
Borst P, Greaves DR (1987) Programmed gene rearrangements altering expression.
Science 235:658-667.
39
Browning JA, Simons MD, Torres E (1977) Managing host genes:epidemiologic
and genetic concepts. In Horsfall JG, Cowling EB (eds). Plant Disease:an
advanced treatise. Vol.l. Academic Press, New York.
Buschges R, Hollricher K, Panstruga R, Simons G, Wolter M, Frijters A, van
Daelen R, van der Lee T, Diergaarde P, Groenendijk J, Topsch S, Vos P,
Salamini F, Schulze-Lefert P (1997) The barley Mlo gene: A novel control
element of plant pathogen resistance. Cell 88:695-705.
Chen F, Hayes PM (1989) A comparison of Hordeum bulbosum-mediated haploid
production efficiency in barley using in vitro floret and tiller culture. Theor.
Appl. Genet. 77:701-704.
Chen F, Prehn D, Hayes PM, Mulrooney D, Corey A, Vivar H (1994) Mapping
genes for resistance to barley stripe rust (Puccinia striiformis f.sp hordei).
Theor Appl Genet 88:215-219.
Chen XM, Line RF, Leung H (1995) Virulence and polymorphic DNA
relationships of Puccinia striiformis f.sp. hordei to other rusts. Phytopathology
85:1335-1342.
Chen XM, Line RF, Leung H (1998) Genome scanning for resistance-gene analogs
in rice, barley, and wheat by high-resolution electrophoresis. Theor. Appl.
Genet 97:345-355.
Crute IR (1985) The genetic basic of relationships between microbial parasites and
their host. Pages 62 in: Mechanisms of resistance in plant diseases. Kluwer
Academic Press.
Collins NC, Paltridge NG, Ford CM, Symons RH (1996) The Yd2 gene for barley
yellow dwarf virus resistance maps close to the centromere on the long arm of
barley chromosome 3. Theor. Appl. Genet 92:858-864.
Conner RL, Whelan EDP, MacDonald MD (1989) Identification of sources of
resistance to common root rot in wheat-alien amphiploid and chromosome
substitution lines. Crop Science 29:916-919.
D'Acry CJ (1995) Symptomatology and host range of barley yellow dwarf. Pages
9-28 in: Barley Yellow Dwarf: 40 Years of Progress. D'Acry CJ and Burnett
PA, eds. The American Phytopathologal Society, St. Paul, MN.
Dangl JL (1995) Piece de resistance: novel classes of plant disease resistance genes.
Cell 80:363-366.
40
Dingerdissen AL, Geiger HH, Lee M, Schechert A, We lz HG (1996) Interval
mapping of genes for quantitative resistance of maize to Setoria turcica, cause of
norther leaf blight, in a tropical environment. Molecular Breeding 2:143-156.
Dubin HJ, Stubbs RW (1986) Epidemic spread of barley stripe rust in South
America. Plant Dis 70:141-144.
El-Kharbotly A, Leonards-Schippers C, Huigen DJ, Jacobsen E, Pereira A (1994)
Segregation analysis and RFLP mapping of the RI and R3 alleles conferring
race-specific resistance to Phytohthora infestans in progeny of dihaploid potato
plants. Mol.Gen.Genet. 242:749-754.
Ellis JG, Lawswnce GJ, Peacock WJ, and Pryor AJ (1998) Approaches to cloning
plant genes conferring resistance to fungal pathogens. Annual Review of
Phytopathology 26:245-263.
Feuerstein U, Brown AHD, Burdon JJ (1990) Linkage of rust resistance genes from
wild barley (Hordeum spontaneum) with isozyme markers. Plant Breed
104:318-324.
Graner A, Jahoor A, Schondelmaier J, Siedler H, Pillen K, Fischbeck G, Wensel G,
Herrmann G (1991) Construction of an RFLP map of barley. Theor.Appl.
Genet. 83:250-256.
Graner A, Bauer E (1993) RFLP mapping of the ym4 virus resistance gene in
barley. Theor.Appl.Genet. 86:689-693.
Garner A, Tekauz A (1996) RFLP mapping in barley of a dominant gene
conferring resistance to scald (Rhynchosporium secalis). Theor. Appl. Genet.
93:421-425.
Giease H, Jorgensen J, Jensen HP, Jensen J (1981) linakge relationships of ten
powdery mildew resistance genes on barley chromosome 5. Hereditas 95:43-50.
Giease H, Holm-Jensen AG, Jensen HP, Jensen J (1993) Localization of the
Laevigatum powdery mildew resistance gene to barley chromosome 2 by the use
of RFLP markers.Theor.Appl.Genet. 85:897-900.
Harder DE, Harber S (1992) Barley yellow dwarf virus. Pages 379 in: Oats Science
and Technology. Marshall HG, Sorrells ME, eds. American Society of
Agronomy, Inc, Madison, WI.
Hayes PM, Prehn D, Vivar H, Blake T, Comeau A, Henry I, Johnston M, Jones B,
Steffenson B (1996) Multiple disease resistance loci and their relationship to
agronomic and quality loci in a spring-barley population. J.QTL
http: / /probe.nalusda.gov: 8000 /otherdocs /j QTL/index.htm
41
Heun M, Kennedy AE, Anderson JA, Lapitan NLV, Sorrels ME, Tanks ley SD
(1991) Construction of a restriction fragment length polymorphism map for
barley (Hordeum vulgare). Genome 34:437-447.
Hoffbeck MD, Openshaw SJ, Geadelmann JL, Peterson RH, Stuthman DD (1995)
Backcrossing and intermating in an exotic X adapted cross of maize. Crop
Science 35:1359-1364.
Holloway J, Knapp SJ (1994) Gmendel 3.0 Users Guide. [email protected]
Hooker AL (1985) Corn and sorghum rust. In "The cereal rusts. Volume 2.
Diseases, Distribution, Epidemiology, and Control" (Roelfs AP and Bushnell
WR, eds) pp 207-233.
Jin Y, Statler GD, Franckowiak JD, Steffenson BJ (1993) Linkage between leaf rust
resistance genes and morphological markers in barley. Phytopathology 83:230233.
Jin Y, Cui GH, Steffenson BJ, Franckowiak JD (1996) New leaf rust resistance in
barley and their allelic and linkage relationships with other Rph genes.
Phytopathology 86:887-890.
Johnson R (1981) Durable resistance: definition of, genetic control, and attainment.
Phytopathology 71:567-568.
Jorgensen JH (1992) Discovery, characterization and exploitation of Mbo powdery
mildew resistance in barley. Euphytica 63:141-152.
Kanazin V, Marex LF, Shoemaker RC (1996) Resistance gene analogs are
conserved and clustered in soybean. Proc. Natl. Acad. Sci. USA 93:1174611750.
Kleinhofs A, Kilian A, Saghai Maroof M, Biyashev R, Hayes PM, Chen F,
Lapitan A, Fenwick A, Blake T, Kanazin V, Ananiev E, Dahleen L, Kudrna D,
Bollinger J, Knapp SJ, Liu B, Sorrells M, Heun M, Franckowiak J, Hoffman D,
Skadsen R, Steffension B (1993) A molecular, isozyme, and morphological
map of the barley (Hordeum vulgare) genome. Theor. Appl. Genet. 86:705-712.
Lamb CJ (1994) Plant disease resistance genes in signal perception and
transduction. Cell 76:419-422.
Leister D, Ballvora A, Salamini F, Gebhardt C (1996) A PCR-based approach for
isolsting pathogen resistance genes from potato with potential for wide
application in plants. Nature Genetics 14:421-429.
42
Leister D, Kurth J, Laurie DA, Yano M, Sasaki T, Devos K, Graner A, SchulzeLefert P (1998) Rapid reorganization of resistance gene homologues in cereal
genomes. Proc. Natl. Acad. Sci. USA 95:370-375.
Line RF (1993) Durability of resistance to Puccinia striiformis in North American
wheat cultivars. In: Jacobs TH, Parlevliet JE (eds) Durability of disease
resistance, Kluwer Academic Press. p 332.
Liu ZW, Biyashev RM, Saghai Maroof MA (1996) Development of simple
sequence repeat markers and their integration into a barley linkage map. Theor.
Appl. Genet. 93:869-876.
Maisonneuve B, Bellec Y, Anderson P, Michelmore RW (1994) Rapid mapping of
two genes for resistance to downy mildew from Lactuca serriola to existing
clusters of resistance genes. Theor.Appl.Genet. 89:96-104.
Martin GB (1996) Molecular cloning of plant disease resistance genes. In: Plant
microb interactions vol.1. Martin GB, Herrera-Hestrella L, Rosales LS, RiveraBustamante R, Neuenschwander U, Lawton K, Ryals J, Schardl CL, Kronstad
JW, Thomashow LS, Weller DM, Philips DA, Streit WR, Prome JC, Demont N,
Stacey G (eds).Chapman & Hall; New York; USA.
Marshall D, Sutton RL (1995) Epidemiology of stripe rust, virulence of Puccinia
striiformis f. sp. hordei, and yield loss in barley. Plant Dis 70:732-737.
Melchers LE, Parker JH (1992) Rust resistance in winter wheat varieties. US Dep
Agric Bull 1046
Melchinger AE, Utz HF, Schon CC (1998) Quantitative trait locus (QTL) mapping
using testers and independent population samples in maize reveals low power of
QTL detection and large bias in estimates of QTL effects. Genetics 149:383403.
Michelmore RW (1995). Molecular approaches to manipulation of disease
resistance genes. Ann. Rev. Phytopath 33:393-428.
Morgante M, Rafalski A, Bidle P, Tingey S, Oliveri AM (1994) Genetic mapping
and variability of seven soybean simple sequence repeat loci. Genome 37:763769.
Ordon F, Friedt W (1993) Mode of inheritance and genetic diversity of BaMMV
resistance of exotic barley germplasms carrying genes different from ym4.
Theor. Appl. Genet. 86:229-233.
43
Ori N, Eshed Y, Paran I, Presting G, Aviv D, Tanks ley S, Zamir D, Fluhr R (1997)
The 12C family from the wilt disease resistance locus 12 belongs to the
nucleotide binding, leucine-rich repeat superfamily of plant resistance genes.
Plant Cell 9:521-532.
Paran I, Kesseli R, Michelmore R (1991) Identification of restriction fragment
length polymorphism and random amplified polymorphic DNA markers linked
to downy mildew resistance genes in lettuce, using near-isogenic lines. Genome
34:1021-1027.
Powell W, Thomas WTB, Baird E, Lawrence P, Booth A, Harrower B, McNicol
JW, Waugh R (1997) Analysis of quantitative traits in barley by the use of
Amplified Fragment Length Polymorphisms. Heredity 79:48-59.
Pryor T (1987) The origin and structure of fungal disease resistance genes in plants.
Trends in Genetics 3:157-161.
Pryor T, Ellis J (1993) The complexity of fungal resistance genes in plants.
Advances in Plant Pathology 10:281-304.
Qi X, Stam P, Lindhout P (1996) Comparison and integration of four barley
genetic maps. Genome 39:379-394.
Qi X (1998) Identification and mapping of genes for partial resistance to
Puccinia hordei Otth in barley. Thesis Wageningen Agricultural University.
Reuvein R, Dudai N, Putievsky E, Elmer WH, Wick RL (1997) Evaluation and
identification of basil germplasm for resistance to Fusarium oxysporum f sp.
Basilicum. Plant Disease 81:1077-1081.
Russell J, Fuller J, Young G, Thomas B, Taramino G, Macaulay M, Waugh R,
Powell W (1997) Discriminating between barley genotypes using microsatellite
markers. Genome 40:442-450.
Saxena KMS, Hooher AL (1968) On the structure of a gene for disease resistance
in maize. Proc. Natl. Acad. Sci. USA 61:1300-1305.
Sharma H, Ohm H, Goulart L, Lister R, Appels R, Benlhabib 0 (1995)
Introgression and characterization of barley yellow dwarf virus resistance from
Thinopyrum intermedium into wheat. Genome 38:406-413.
Singh RP (1993) Genetic association of gene Bdvl for tolerance to barley yellow
dwarf virus with genes Lr34 and Yr18 for adult plant resistance to rusts in bread
wheat. Plant Dis 77:1103-1106.
44
Tanks ley 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. Theor. Appl. Genet. 92:191-203.
Thomas WTB, Powell W, Waugh R, Chalmers KJ, Barua UM, Jack P, Lea V,
Forster BP, Swanston JS, Ellis RP, Hanson PR, Lance RCM (1995) Detection of
quantitative trait loci for agronomic, yield, grain and disease characters in
spring barley (Hordeum vulgare L.). Theor Appl Genet 91:1037-1047.
Tinker NA, Mather D (1995) Methods for QTL analysis with progeny replicated in
multiple environments. J.QTL:http://probe.nalusda.gov:8000/otherdocs/jqt1/
Toojinda T, Baird E, Booth A, Broers L, Hayes PM, Powell W, Thomas W, Vivar
H, Young G (1997) Introgression of quantitative trait loci (QTLs) determing
stripe rust resistance in barley: an example of marker-assisted line development.
Theor Appl Genet 96:123-131.
Veremis JC, Cap GB, Robert PA (1997) A search for resistance in Lycopersicon
spp. to Nacobbus aberrans. Plant Disease 81:217-221.
Von Wettstein-Knowles P (1992) Barley: genetics, biochemistry, molecular biology
and biotechnology. CAB Int, Wallingord, UK, pp 73-98.
Williamson VM, Ho JY, Wu FF, Miller N, Kaloshian I (1994) A PCR-based
marker tightly linked to nemathod resistance gene, Mi, in tomato.
Theor.Appl.Genet. 87:757-763.
Young ND (1996) QTL mapping and quantitative disease resistance in plants.
Aim. Rev. Phytopath. 34:479-501.
Yu YG, Buss GR, Saghi Maroof MA (1996) Isolation of a superfamily of candidate
disease-resistance genes in soybean based on a conserved nucleotide-binding
site. Proc. Natl. Acad. Sci. USA 93:11751-11756.
Zabeau M, Vos P (1993) European patent application publication no. EP 0534858
Zaitlin D, DeMars S, Ma Y (1993) Linkage of Rhm, a recessive gene for resistance
to southern corn leaf blight, to RFLP marker loci in maize (Zea mays) seedlings.
Genome 36:555-564.
45
CHAPTER 3
INTROGRESSION OF QUANTITATIVE TRAIT LOCI (QTLs)
DETERMINING STRIPE-RUST RESISTANCE IN BARLEY: AN EXAMPLE
OF MARKER-ASSISTED LINE DEVELOPMENT
Theerayut Toojinda, Patrick M. Hayes, Hugo Vivar
46
Abstract
Genome analysis tools are useful for dissecting complex phenotypes and
manipulating determinants of these phenotypes in breeding programs. Quantitative Trait
Locus (QTL) analysis tools were used to map QTLs conferring adult plant resistance to
stripe rust (caused by Puccinia striiformis f.sp. hordei) in barley. The resistance QTLs
were introgressed into a genetic background unrelated to the mapping population with
one cycle of marker-assisted backcrossing. Doubled haploid lines were derived from
selected backcross lines, phenotyped for stripe rust resistance, and genotyped with an
array of molecular markers. The resistance QTLs that were introgressed were significant
determinants of resistance in the new genetic background. Additional resistance QTLs
were also detected. The susceptible parent contributed resistance alleles at two of these
new QTLs. We hypothesize that favorable alleles were fixed at these new QTLs in the
original mapping population. Genetic background may, therefore, have an important role
in QTL transfer experiments. A breeding system is described that integrates single copy
and multiplex markers with confirmation of the target phenotype in doubled haploid lines
phenotyped in field tests. When resources are limited, this approach may be useful for
simultaneously producing agronomically useful germplasm and contributing to an
understanding of quantitatively inherited traits.
Introduction
Genome analysis based on DNA polymorphisms can reveal the genetic
determinants of complex phenotypes and provide tools for manipulating these
determinants to maximize selection response. We are using molecular markers in barley
(Hordeum vulgare L.) in an attempt to rapidly develop cultivars adapted to the Pacific
47
Northwest of the United States that are resistant to barley stripe rust. Barley stripe rust is
caused by Puccinia striiformis f.sp. hordei.
The disease was first reported in the Americas in 1975 (Dubin and Stubbs 1985).
It was first reported in the United States in 1991 (Marshall and Sutton 1995). By 1995,
the disease was reported throughout the western United States, where localized epidemics
have caused severe losses in yield and quality. While the disease can be controlled by
fungicides, economic and environmental considerations favor genetic resistance.
There is limited information on the genetics of resistance to stripe rust in barley.
Bakshi and Luthra (1971) described a dominant resistance gene in Indian germplasm but
did not map it. Three recessive resistance genes (Yr, Yr2, and Yr3) are reported in
European spring barley germplasm, and one dominant resistance gene is reported in
European winter barley. These genes have not been mapped (Lehmann et al., 1975).
The only stripe rust resistance gene showing Mendelian inheritance that has been mapped
in barley is Yr4. This locus is on located on chromosome 5 (1H). It does not confer
resistance to race 24 (von Wettstein-Knowles 1992).
The virulence of stripe rust in the Americas was first described in terms of race 24
(Dubin and Stubbs 1986). Chen et al. (1994) and Hayes et al. (1996c) reported adult
plant stripe rust resistance QTLs on chromosomes 4 (4H) and 7 (5H). The race
composition of the field inoculum was not known. Thomas et al. (1995) reported QTLs
for adult plant stripe rust resistance to uncharacterized field inoculum on chromosomes 1
(7H), 5 (1H), and 7 (5H). The chromosome 5 (1H) QTL was hypothesized to be due to
the Yr4 locus and the chromosome 7 (5H) QTL to be the same QTL reported by Chen et
al. (1994) and Hayes et al. (1996c). In contrast, the genetics of stripe rust in wheat is an
48
area of extensive study (reviewed by Line et al. 1993). Based on the homoeology of the
two crop species, there are likely to be parallels in the two host: pathogen interaction
systems. For example, quantitative, adult plant resistance will probably be more durable
than race-specific resistance (Line 1993).
We have, therefore, focused our attention on introgressing resistance genes from
genotypes that, under field conditions, allow limited disease development on adult plants.
This type of disease reaction may indicate durable, adult plant resistance (Parlevliet and
Van Ommeren 1975). When such genotypes, developed by the International Center for
Agricultural Research in the Dry Areas (ICARDA) program based at the International
Maize and Wheat Improvement Center (CIMMYT) in Mexico, are crossed with
susceptible genotypes, the progeny show a range of disease reaction phenotypes that do
not fall into discrete classes.
This quantitative inheritance can be studied through the techniques of quantitative
trait locus (QTL) analysis (Hayes et al., 1996c). Based on restriction fragment length
(RFLP) linkage data and adult plant disease reaction phenotype data, we reported stripe
rust resistance QTLs on chromosomes 4 (4H) and 7 (511) of Calicuchima-sib (Chen et al.
1994; Hayes et al. 1996c).
Backcrossing is an approach to introgressing target loci, such as stripe rust
resistance QTLs, into adapted backgrounds. The contribution of the donor parent is
reduced by half with each generation of backcrossing, assuming no linkage. Molecular
markers can increase the efficiency of the process in several ways. Flanking markers can
be used to identify the backcross lines that are heterozygous for target genome regions.
Advancing only these selected lines will also have the effect of reducing linkage drag
49
(Young and Tanks ley 1989; Tanks ley and Nelson 1996). Single-copy, or low-copy,
markers with defined map locations, such as RFLPs and simple sequence repeats (SSRs) ,
are ideal for this step. Molecular markers could also increase the efficiency of
backcrossing by allowing selection of genotypes with the maximum percentage of
recurrent parent genome. Markers with higher information content per reaction, such as
amplified fragment length polymorphisms (AFLPs), are ideal for this step (Waugh et al.
1997).
Manipulation of QTLs can be problematic due to loss of target loci though
recombination, incorrect information regarding the location of the QTLs, and/or
negatively altered expression of the QTLs in new genetic backgrounds (Hayes et al.
1996b). Therefore, a marker-assisted QTL backcrossing scheme for a self-pollinated
crop, such as barley, might (i) use flanking markers to select progeny with a probability
of carrying the target QTL allele(s), (ii) confirm the target phenotype in the selected
progeny, and (iii) use multiplex markers to identify those selections with the maximum
percentage of recurrent parent genome. In crops where a rapid approach to homozygosity
(such as a doubled haploid technique) is possible, the efficiency of the second step can be
increased by phenotyping on a plot, rather than an individual line basis. See Powell et al.
(1996) for additional detail on the idea of integrating single and multi-locus markers in
barley breeding.
Our long-term practical objective is to develop six-row, spring habit germplasm
adapted to the Pacific Northwest of the United States that has durable resistance to stripe
rust. In doing so, we sought to (i) validate the effects of mapped stripe rust resistance
QTLs, (ii) determine if there were different resistance QTLs in an unrelated genotype,
50
and (iii) pilot a marker-assisted backcrossing scheme incorporating RFLP, AFLP, and
random amplified polymorphic DNA (RAPD) markers, doubled haploids, and field
phenotyping.
Materials and Methods
Germplasm
The germplasm derivation process is shown in Figure 3.1. BSR41, a 6-row
mapline from the Calicuchima-sib x Bowman mapping population (Chen et al. 1994;
Hayes et al. 1996c) was used as the donor parent in a single backcross to the variety
Steptoe. Steptoe is the most widely-grown six-row feed barley in the Pacific Northwest
United States. It has been the subject of intensive genome mapping efforts by the North
American Barley Genome Mapping Project (see Hayes et al. 1996b for a review).
Steptoe is susceptible to stripe rust.
Four RFLP markers bracketing the stripe rust resistance QTLs on chromosomes 4
(4H) and 7 (5H) described by Chen et al. (1994) and Hayes et al. (1996c) were screened
in a population of 66 backcross-one (BC1) generation plants. Only a subset of the RFLP
markers available at the time this experiment was conducted showed polymorphism
between Steptoe and BSR41. The target regions on chromosome 4 (4H) and
chromosome 7 (5H) were poorly populated with markers when this work was carried out,
and they remain sparsely populated on the current barley consensus map of Qi et al.
(1996).
ABG366 and Bmyl flank the resistance QTLs on chromosome 4 (4H). ABG366
was not mapped in the reference mapping population, Calicuchima-sib x Bowman, where
51
the resistance QTL peak was detected between Bmyl and ABG397, a distance of 28.1
cM. ABG366 and Bmyl are 32.7 cM apart on the Steptoe x Morex map (Mather 1995).
In the Steptoe x Morex cross, ABG366 is 3.7 cM proximal to ABG397 (Mather 1995).
WG530 and CD057 flank the resistance QTL on chromosome 7 (5H). WG530 was not
mapped in Calicuchima-sib x Bowman, where the resistance QTL peak was detected in
the Ale-CD057 interval, a distance of 20.3 cM. WG530 and CD057 are 31.4 cM apart
on the Steptoe x Morex map (Mather 1995).
As shown in Figure 3.1, eleven of the sixty six BC1 plants (plant numbers 6, 7,
20, 21, 22, 28, 40, 50, 55, 56, and 58) were selected as heterozygotes for the target
BSR 41 x STEPTOE
Fl x STEPTOE
BC 1
1
20. 21. 22
6. 7
28
40
55. 56....58
50
n= 66
It
DH 6
17
15
I
SELECTIONS 2
8
14
3
10
19
6
17
19
n=134
I
2
n= 10
ir
e4ir
4/
Figure 3.1 Derivation of doubled haploid (DH) germplasm from a markerassisted selection program for adult plant stripe rust resistance. BC1 refers to the first
backcross generation and the numbers identify the BC1 plants that were selected based
on their genotypes at marker loci flanking stripe rust resistance QTLs on chromosomes 4
(4H) and 7 (5H). DH refers to doubled haploid and the numbers indicate the number of
DH lines produced from each BC1 plant. Selections refers to the ten most resistant DH
lines advanced to extensive phenotyping. See text for additional details on line derivation
and genotyping procedures.
52
flanking markers. Doubled haploid (DH) lines were derived from these BC1 plants,
using the Hordeum bulbosum technique, as described by Chen and Hayes (1989). A total
of 134 DH lines were produced, with varying numbers of DHs per BC1 plant, as shown
in Figure 3.1. For example, six DHs were produced from BC1 plant no. 6, 17 DHs were
produced from BC1 plant no. 7, etc.
Genotyping
The 134 DH lines were genotyped for the four RFLP markers used for resistance
QTL introgression and an additional RFLP on chromosome 4 (4H) (ABG54); two
morphological markers on chromosome 7 (5H), mSrh (rachilla hair length) and mR (awn
texture); 106 AFLPs; and eight RAPDs. RFLPs were assayed as described by Chen et al.
(1994). The morphological markers were scored under a dissecting microscope. RAPDs
were assayed as described by Barua et al. (1993). The AFLP methodology was
essentially as described by Zabeau and Vos (1993) with the following modifications.
Template DNA was prepared the using restriction enzyme combination EcoRIIMsel
(Boehringer Mannheim/New England Biolabs). 2.512g genomic DNA was digested as
outlined by Zabeau and Vos (1993) and two specific double stranded adapters were
ligated to the fragment ends. Neither of the adapters was biotinylated and the selection
step using streptavidin coated magnetic beads was omitted. Adapter sequences were:
EcoRI
5' CTCGTAGACTGCGTACC
5' AATTGGTACGCAGTC
MseI
5' GACGATGAGTCCTGAG
5' TACTCAAGGACTCAT
53
The digested and ligated DNA was preamplified using an EcoRI-directed primer and an
MseI-directed primer. The primers did not have additional selective nucleotides at the 3'
end. The sequences of the primers were
E00
5' GACTGCGTACCAATTC
MOO
5' GATGAGTCCTGAGTAA
Preamplification was performed in a total volume of 25111 containing 75ng each of
primers E00 and MOO, 0.2mM of all four dNTPs (Pharmacia), 1 x PCR buffer (Perkin
Elmer Cetus), 1U Amplitaq DNA Polymerase LD ( Perkin Elmer Cetus) and 3Ong of the
digested and ligated DNA. The cycle profile used was as follows; denaturation for 30s at
94°C, annealing for 30s at 60°C, and extension for 60s at 72°C, for 30 cycles. After
preamplification, the product was diluted by the addition of 55111 of buffer (10mM Tris
pH8, 0.1mM EDTA). This mixture was used as a template for selective amplification.
Selective amplification was carried out using adapter-directed primers. Nine different
primer combinations were used, each combination consisted of one 'Eco' primer and one
'Mse' primer. All of the primers had three selective nucleotides at the 3' end. Primer
combinations and base extensions were:
Primer combination
EcoRI Msel
e32m34
AAC AAT
e35m42
ACA AGT
e36m33
ACC AAG
e36m36
ACC ACC
e36m50
ACC CAT
e38m31
ACT AAA
54
Primer combination
EcoRI Msel
e38m50
ACT CAT
e39m61
AGA CTG
e41m33
AGG AAG
In each case the 'Eco' primer was radiolabelled using 33P -ATP as described by
Vos et al. (1995). The amplification reactions were carried out in a total volume of 20111,
comprising 6.7ng labeled EcoRI primer, 25ng unlabelled EcoRI primer, 3Ong Msel
primer, 0.2 mM of all four dNTPs, 1 x PCR buffer (Perkin Elmer Cetus), 0.5U Amplitaq
DNA Polymerase (Perkin Elmer Cetus) and 2111 of template DNA.
Reactions were carried out using the cycle profile described by Vos et al. (1995).
All PCR reactions were performed using a Perkin Elmer 9600 thermocycler. Reactions
were stopped by the addition of an equal volume of formamide loading buffer (98%
formamide, 10mM EDTA pH8, Bromophenol Blue, Xylene Cyanol). The samples were
denatured at 90°C for 5 minutes. Then, 3.51.11 of each sample was loaded onto a 40cm, 6%
denaturing polyacrylamide gel (Easigel, Scotlab) which had been preheated by running at
80W for 30mins. The samples were then electrophoresed at a constant power of 80W for
lhr 45mins. Gels were transferred to Whatmann 3MM paper and dried for 2 hrs at 80°C
on a gel drier (Biorad). They were then exposed to autoradiographic film (X-OMAT S,
Kodak) to visualize the results. Results were scored manually. The AFLP loci were
named based on enzyme, primer sequence, and band size. For example, e38m311 refers
to band "1" revealed in this germplasm, using primer sequence 38 with EcoRI and primer
sequence 31 with Msel.
55
Phenotyping
The DH lines were evaluated in four field tests. In the first test, lines and parents
were grown in uni-replicate hill plots at Celaya, Mexico in the winter of 1995. To initiate a
field epidemic, spreader rows (formed from a mixture of 15 extremely susceptible
genotypes) were inoculated with a stripe rust isolate whose virulence pattern corresponds to
the race 24 Varunda- Mazurka type described by Dubin and Stubbs (1986). Stripe rust
severity was rated at DGS 59 (Feekes stage 10.5) as percent severity on a plot basis.
Percent severity was estimated visually as the percentage of the total plant canopy in each
plot that was covered with stripe rust pustules.
DH lines and parents were then assessed at three planting dates at Toluca, Mexico
in the summer of 1995. A single replicate was grown at each planting date, using three-m,
one row plots. Spreader rows, planted at 5.25 m intervals and consisting of a mixture of 15
susceptible genotypes, were inoculated twice with infected plants placed in the foliage and
with applications of spores suspended in oil. Infected plants and spores were collected
locally. The race composition of this inoculum was not determined. Genotypes inoculated
in this fashion will never escape rust infection (H. Vivar, personal communication).
Stripe rust was rated as percent severity on a plot basis. At the time of rating,
genotypes at the three planting dates were at growth stages DGS 75, DGS 59, and DGS 49,
respectively. Multiple planting dates were used in an attempt to determine the effect of
maturity on the expression of stripe rust resistance. In the summer of 1996 the ten most
resistant and ten most susceptible lines (as measured by average performance in the
previous four tests), and the two parents, were grown at Toluca, Mexico. The phenotyping
and rating procedures were the same as those employed in 1995.
56
Data Analysis
Of the 134 DH lines that were produced, there was sufficient seed to include 96 in
all four 1995 phenotyping experiments. Subsequent analyses were based on these 96
genotypes. Each of the four experiments grown in 1995 were considered replicates for the
purposes of calculating an estimate of the heritability of stripe rust severity. This
heritability estimate was calculated as:
H2
cy2g/(0.2g
62e /r)) where 62g is the variance among DH lines, 62e is the error
variance, and r = 4, the number of environments sampled.
Alleles at the 120 marker loci were scored as 0 (Steptoe) or 1 (BSR41) and were
considered independent variables. Stripe rust severity data were considered dependent
variables and were used for estimating genotype: phenotype associations via simple and
multiple regression. These associations were determined for each of the 1995
environments, and from the mean of the four 1995 environments. Individual markers
that were significant (p < .05) determinants of trait expression were included in multiple
regression models.
Determining the joint effects of multiple loci could be biased if linked loci are
included in a multi-locus model. Therefore, only the most significant single locus from
each group of linked loci was included in the multi-locus model. Procedures for
grouping linked loci are described in the next paragraph. Multiple regression models
were evaluated using Sawa's Bayesian Information Criterion (BIC), as available in the
Statistical Analysis System (SAS, 1988).
The sign of the slope was used to identify the value of stripe rust reaction alleles
contributed by each parent. Negative and positive effects indicate that BSR41
57
contributed resistant and susceptibility QTL alleles, respectively. The R2 value was used
as a measure of the total phenotypic variance accounted for by each marker and by the
joint analysis of multiple markers. A genotypic coefficient of variation was calculated as
R2p/H2, where Rep is the proportion of phenotypic variance accounted for by a marker or
set of markers, and H2 is the heritability. Putative map positions of the AFLP and RAPD
loci were established as follows. Homologous AFLP products, identified by fragment
sizes, have been shown to map to the same regions in the barley genome in a study of a
number of barley crosses (Waugh et al., 1997). Some of the primer combinations used in
the current study had also been used in extending the Dicktoo x Morex map (Hayes et al.,
1996a) and inclusion of Dicktoo and Morex on the AFLP gels for the current study
enabled the identification of 12 markers also segregating in the Dicktoo x Morex
population. The chromosomal locations of these markers were therefore established by
reference to the Dicktoo x Morex map.
In order to establish tentative linkage group assignments for AFLP and RAPD
markers that could not be directly related to mapped loci in other mapping populations,
we employed multivariate analysis of all of the marker data. Similarities between markers
in a backcross population are equivalent to 1-r, where r = the recombination value
(Ramsay and Thomas, 1992).
For an unselected backcross of 66 individuals, markers showing similarities >0.74
and >0.82 are significantly linked at the 0.05 and 0.01 probability levels respectively.
Groups showing similarities of >0.82 were therefore formed and given a tentative
chromosomal assignment if they contained markers mapped in other populations.
Because the population assayed with molecular markers is based on 11 selected BC1
58
genotypes, we cannot be completely sure of the chromosome location of these groups.
However, the putative locations provided an objective basis for identifying a single locus
from each group of loci for the multilocus regression models, as described in the previous
paragraph.
Results
The phenotypic distribution of stripe rust severity in the DH lines, averaged over
the four environments sampled in 1995, did not show discrete classes (Figure 3.2).
15
Steptoe
81.6+1-4.5
BSR41
12.1+/-1.5
12
i)
9
0
a)
E
6
z
3
0
I
1
I
10
I
I 1 I- 111111111
1 1 I 1 1 I I 1111
11
N
1
20
I
30
40
50
60
Disease severity (% )
70
90
100
Figure 3.2 Average stripe rust disease severity (%) in doubled haploid lines
derived from one cycle of marker-assisted backcrossing using BSR41
and Steptoe as the donor and recurrent parents, respectively.
59
Similar distributions were observed for each of the individual environments (data not
shown). The standard errors for disease severity for BSR41 and Steptoe were + 1.5 and +
4.5%, respectively. The heritability of stripe rust severity, calculated using environments
as replicates, was 0.95.
A total of 120 data points were generated on the 96 genotypes. Of these, four
were RFLPs, two were morphological markers, one-hundred and six were AFLPs and
eight were RAPDs. The individual markers that were significant determinants of stripe
rust severity are shown in Table 3.1. These include markers used for introgression of
stripe rust resistance QTLs on chromosome 4 (4H) (Bmyl) and chromosome 7 (5H)
(CD057). The chromosome 4 markers ABG54 and ABG366 were not significantly
associated with stripe rust severity. WG530, which is proximal to CD057, did not have a
significant association with stripe rust severity, while mSrh, which is distal to CD057,
did have a significant association with stripe rust severity. Effects for these flanking loci
were negative, indicating that BSR41 contributed the resistance alleles. Nine of the
AFLPs and one of the RAPDs also showed significant associations with stripe rust
severity. Of these, three were positive effects, indicating that Steptoe contributed the
favorable allele.
Of the markers that were significant determinants of stripe rust severity in single
locus regressions, five were significant in the multi-locus model for the average of the
four 1995 environments (Table 3.2). These included the CD057 and Bmyl markers used
for introgression of resistance QTLs and three AFLP markers: e36m36a, e38m311, and
e36m50i. Based on the relationship with the Dicktoo x Morex map, the e36m36a marker
is on chromosome 3 (3H). The chromosome positions of the e36m50i, and e38m311
60
markers can only be inferred by the multivariate analysis. The e38m311 marker was
grouped with the e39m61k locus, which is equivalent to the e39m61s locus on the
Dicktoo x Morex map (Hayes et al. 1996a). The e39m61s locus is 48.8 cM distal to the
mSrh locus on the Dicktoo x Morex map. Therefore, with reference to the current study,
the e38m311 locus would be expected to be unlinked with CD057.
Table 3.1 Chromosome location, slope, p-value, and R2 for markers significantly
associated with stripe rust severity in single locus regressions.
Chromosome locations in italics are putative. Negative and positive slopes
indicate that BSR41 contributed resistance and susceptibility QTL alleles,
respectively.
Marker
Chromosome
location
Slope
p-value
e36m36a
e41m33j
e41m33p
e36m33j
e36m36h
OPD8a
3
3
3
3
3
3
-17.5
-13.1
-13.5
-12.0
-12.6
-12.9
0.000
0.001
0.001
0.009
0.002
0.005
10.4
10.8
8.3
9.7
9.6
Bmyl
e32m24q
4
4
-11.1
-9.4
0.006
0.042
7.9
5.2
CD057
7
7
?
?
?
-15.6
-13.9
0.000
0.000
0.011
0.041
0.063
15.6
12.1
MSrh
e36m36d
e36m50i
e38m311
13.6
12.1
11.9
19.1
8.1
5.4
4.5
On the Dicktoo x Morex map, the e39m61 s locus is 16.5 cM from the mR locus. In the
current study, there was no consistent pattern of association between alleles at the mR and
e38m311loci, indicating a lack of linkage. The e36m50i locus was grouped with the
chromosome 3 (3H) loci in the multivariate analysis.
61
In the average of the four environments, the CD057, e36m36a, Bmyl, e38m311,
and e36m50i loci accounted for 54% of the phenotypic, and 57% of the genotypic,
variation in stripe rust severity. As shown in Table 3.2, CD057 was the only locus
significant in all four environments. The e36m33d locus effect was unique to the Celaya
environment. The e36m36a and e38m311loci were common to all of the Toluca
environments. The e36m50i and Bmyl loci were significant in the second and third
Toluca environments.
Table 3.2 Markers with significant effects in multi-locus regression models of
stripe rust severity in individual environments and in the average of
four environments. Negative and positive slopes indicate that BSR41
contributed resistance and susceptibility QTL alleles, respectively.
Environment
Marker
Slope
p-Value
CD057
e36m36a
Bmyl
-17.7
-14.3
-12.4
e38m311
e36m501
13.9
10.4
0.0001
0.0001
0.0008
0.0032
0.0206
0.54
Celaya
CD057
-14.6
DGS 59
e36m33d
e36m50i
0.0100
0.0400
0.0300
0.29
0.0001
0.0001
0.0129
0.41
0.0005
0.0005
0.0001
0.0012
0.0170
0.52
0.0001
0.0051
0.0001
0.0017
0.0649
0.60
Average
14.1
16.3
Toluca 1
CD057
DGS 75
e36m36a
-22.2
-19.4
e38m311
15.0
CD057
e38m311
-17.1
-17.4
-19.8
20.9
e36m50i
14.6
Toluca 2
DGS 59
Toluca 3
DGS 49
e36m36a
Bmyl
CD057
-20.91
e36m36a
Bmyl
-10.26
-15.90
e38m311
e36m50i
14.36
8.24
62
Table 3.3 Average stripe rust severities in 1995 and 1996 for the ten most resistant
and susceptible doubled haploid lines derived from one cycle of marker
assisted selection. Allelic structure for each line is shown at marker loci
significant in multi-locus regressions. Alleles from BSR41 are coded as
"1", and this genotype is expected to contribute resistance alleles at QTL
linked to e36m36a, Bmyl, and CD057. Alleles from Steptoe are coded as
"0" and this genotype is expected to contribute resistance alleles at QTL
linked to e36m50i and e38m311. Marker allele genotypes contrary to these
expectations are shown in italics.
Line number
Mean stripe
rust severity
Mean stripe
rust severity
1995
1996
e36m36a
CD057
Bmyl
e36m50i
e38m311
10 most resistant lines
SR33
SR35
SR43
SR47
SR66
SR80
SR116
SR123
SR125
SR127
13
8
1
1
1
0
0
18
18
1
0
1
0
23
29
1
0
1
1
1
1
14
22
22
27
0
0
0
18
22
18
14
7
7
18
17
20
13
0
0
1
1
1
1
0
0
1
1
0
0
1
1
1
1
1
1
0
0
0
0
1
1
1
0
0
1
0
1
0
0
0
0
0
0
0
0
10 most susceptible lines
SR5
SR8
SR34
SR46
77
90
0
73
63
75
91
0
0
0
1
1
0
1
0
0
0
0
88
0
0
1
SR50
SR72
SR85
SR93
SR97
SR120
0
0
1
1
83
85
87
93
95
0
0
0
1
0
65
90
65
78
70
72
72
90
0
0
1
1
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
82
92
0
0
0
0
0
12
8
1
1
1
1
1
1
Parents
Steptoe
BSR41
Marker genotypes for loci significant in the average multi-locus regression model
of the ten most resistant and ten most susceptible genotypes, together with their average
63
stripe rust severity ratings in the 1995 and 1996 field trials, illustrate the phenotype:
genotype associations (Table 3.3). The ten most resistant genotypes traced to six different
backcross plants (Figure 3.1). Three of the four resistant lines selected for accelerated
assessment as potential varieties (SR123, SR125, and SR127) traced to BC1 plant no. 58
(Figure 3.1). The first three of these selections have the favorable marker genotypes at
all loci. SR127 has the susceptibility marker allele at the chromosome 4 (4H) marker
locus. There were five deviations from the predicted favorable allele genotype among the
most resistant lines and 20 deviations from the predicted unfavorable genotype in the
susceptible lines.
7
93
127
6
85
123
43
5
_8 34
116
125
12
72
35
80
47
66
33
97
4
50
5
46
.n 3
2
1
Donor genome (%)
Figure 3.3 Percentage of donor parent genome in doubled haploid lines derived
from one cycle of marker assisted backcrossing as measured by 120
markers. The numbers in standard and italic font are line numbers
(see Table 3.3) of the ten most resistant and susceptible genotypes,
respectively.
64
Based on 120 data points, the percentage of donor parent genome in the
population ranged from seven to 56 percent, with an average of 32.6% (Figure 3.3).
Considering the ten most resistant genotypes, the percentage of donor genome ranged
from 19 to 46%, with an average of 34.8%. The percentage of donor genome in the ten
most susceptible lines ranged from 8 - 43%, with an average of 24.6%. The four resistant
genotypes selected for accelerated regional assessment as potential varieties were
selected before the AFLP data were available. Selection was based on phenotypic
resemblance to the recurrent parent (Steptoe). The percentage of donor parent genome in
these genotypes ranged from 32 - 39%.
Discussion
The continuous distribution of disease severity phenotypes (Figure 3.2) is
probably due to the quantitative expression of the genetic determinants of resistance
rather than to experimental error in phenotyping. The heritability of 0.95 is an
approximation since environments were used as replicates. However, this high value
indicates a high degree of consistency, or repeatability, in the measurement of the stripe
rust severity phenotype.
The single locus regressions underscore the importance of resistance QTLs on
chromosomes 7 (5H) and 4 (4H) that were introgressed from BSR41 into the Steptoe
background. The significance of mSrh, and lack of significance of WG530, indicates that
the resistance QTL on chromosome 7 (5H) is most likely distal to CD057. Multiple QTL
peaks were observed in the original mapping population (Chen et al. 1994). On
chromosome 4 (4H), the higher R2 value for Bmyl indicates the resistance locus is closer
65
to this marker than to the proximal markers ABG366 and ABG54. The significance and
sign of the chromosome 4 (4H) and 7 (5H) markers confirms that in these regions of the
genome there are loci determining adult plant reaction to stripe rust.
We recognize that experiment-wise error rates complicate single locus regression
procedures with large marker data sets and that a selected population is not as appropriate
for QTL detection as a defined mapping population. However, the presence of loci
significant in both the single locus and multi-locus models indicates that other regions of
the genome were potentially important in determining reaction to stripe rust in the
Steptoe background. There is an important resistance QTL, tentatively mapped to
chromosome 3, where BSR41 contributes the favorable allele. Steptoe contributed
resistance QTL alleles at markers that could not be assigned genome positions. We
hypothesize that favorable (resistance) alleles were fixed at these additional resistance
loci in the original mapping population (Calicuchima-sib x Bowman). This underscores
the lack of predictability that may be encountered when QTLs detected in a mapping
population are transferred to other genetic backgrounds.
Determining the joint effects of multiple resistance loci could be biased if linked
marker loci are included in a multi-locus model. In our data, some linkage relationships
are documented, as in the case of the marker loci used to introgress resistance QTLs on
chromosomes 4 (4H) and 7 (5H). In other cases, we used multivariate analysis to
establish tentative linkage groups. In this way, five AFLP and one RAPD locus (Table
3.1) group together and were tentatively assigned to chromosome 3 (3H). One of these
loci (e36m36a) was significant in the multi-locus model of average stripe rust severity.
BSR41 contributed the resistance allele at this locus.
66
In the case of the e38m311 and e36m50i loci, we could not assign map positions to
these AFLPs that showed associations with stripe rust resistance. These associations are
particularly intriguing because the susceptible parent (Steptoe) contributed resistance
alleles. For the purposes of simultaneous locus discovery and advance of breeding
material through backcrossing, as proposed by Tanksley and Nelson (1996), the ideal
marker would have a defined map position. AFLPs are an excellent type of marker for
rapidly generating large amounts of polymorphism data (Becker et al. 1995; Waugh et al.
1997). AFLP products of the same size assayed in different genotypes may represent the
same locus. However, this needs to be demonstrated on a case-by-case basis.
Five markers accounted for 54% of the variation in phenotypic expression of
stripe rust severity, averaged over the four environments. Using the heritability estimate
of 95%, these markers accounted for 57% of the genetic variation in trait expression. The
fact that the marker loci do not account for a higher proportion of phenotypic and
genotypic variance is probably attributable to recombination between marker loci and the
target QTLs, and the effects of additional loci that contribute to the expression of
resistance.
The individual effects of these additional loci cannot be detected at the level of
resolution afforded by the experiment. This raises the question of how much variance
can be accounted for by QTLs, because the magnitude of the phenotypic, or genotypic,
R2 may be used to determine the likelihood that all loci that are important determinants of
trait expression have been detected.
By way of perspective, the mV locus determines fertility of lateral florets in
barley. Two-row barley genotypes, with sterile lateral florets, typically have higher
67
kernel weights than six-row barley genotypes. When kernel weight was mapped as a
QTL in the doubled haploid progeny of Calicuchima-sib x Bowman a two-row x sixrow cross, the mV locus accounted for 77% of the variation in phenotypic trait expression
(Hayes et al. 1996c). Therefore, having detected 54% and 57% of the phenotypic and
genotypic variance, respectively, in the expression of adult plant stripe rust resistance, we
believe that we have located the principal genes determining stripe rust resistance in this
germplasm. Two are the QTLs that were introgressed, and these are located on
chromosomes 4 (4H) and 7 (5H). One is likely on chromosome 3 (3H), and the map
positions of the remaining loci cannot be determined at this time.
The multi-locus analyses of the individual environment data underscore the
importance of using multiple measures of phenotype and, potentially, the importance of
measuring disease reaction at different growth stages. The heritability estimate of stripe
rust severity was 0.95. This indicates a consistency of response across the four
environments. Therefore, the average of the four environments is one appropriate
measure of the stripe rust reaction phenotype.
However, individual environment data may also be useful. For example, the
Celaya environment was useful in revealing the significance of the resistance QTL linked
to e36m33d. This locus was not significant in any of the Toluca environments, or in the
multi-locus model based on average severity. Markers accounted for the lowest
percentage of phenotypic variance at Celaya, and we attribute this to the use of hill plots.
Iyamabo and Hayes (1995), in a comparison of hill and row plots for QTL detection,
reported that hill plots were best suited to detection of characters determined by largeeffect QTLs.
68
In terms of growth stage, the chromosome 4 (4H) marker Bmyl was significant
only at growth stages DGS 49 and DGS 59. If assessments were based only at DGS 75,
the effect of the chromosome 4 (4H) resistance QTL would not have been detected. The
chromosome 4 (4H) effect may be due to maturity or it may reflect a resistance locus
important at earlier stages of plant growth. The Sh locus is 2.6 cM from Bmyl (Laurie et
al. 1995) and a heading date QTL was detected at the same position in the Calicuchima-
sib x Bowman population (Hayes et al. 1996c). On the other hand, in a controlled
environment seedling test of the Calicuchima-sib x Bowman mapping population, a stripe
rust resistance QTL was detected proximal to Bmyl (Hayes et al. 1996c).
The patterns of association of genotype and phenotype in the ten most resistant
genotypes are additional evidence for the importance of resistance QTLs linked to
markers on chromosomes 3 (3H), 4 (4H), and 7 (5H) and to the two unmapped loci
(Table 3.3). The ten most resistant selections traced to different BC 1 plants, but three of
the selected genotypes traced to a single backcross plant (Figure 3.1). The first three of
these selections have the favorable marker genotypes at all loci (Table 3). The fourth
selection may represent a crossover event between the resistance locus and the Bmyl
marker locus on chromosome 4 (4H) , or it may lack the chromosome 4 (4H) QTL. This
implies that in marker-assisted backcross projects it may be advisable to advance material
derived from multiple backcross plants. This would be even more important when
multiple cycles of marker assisted backcrossing are used without the benefit of
phenotypic assessment for the target phenotype.
The higher degree of correspondence between marker genotype and phenotype in
the resistant vs. the susceptible lines suggests epistasis, as one would expect similar
69
frequencies of crossovers between marker and resistance loci in the two groups. That is,
the resistant phenotype resulted only when appropriate resistance alleles were configured
at multiple loci, while the susceptible phenotype resulted from the presence of only one
or a few susceptibility alleles. SR97, for example, was susceptible to stripe rust but had
resistant marker alleles at four out of five loci. This suggests that if a complex
phenotype, such as adult plant resistance, is the consequence of a complex multi-locus
pathway, it would be relatively easy to disrupt the pathway with susceptible alleles at
various points in the pathway. The resistant phenotype would result only with resistant
alleles at all, or most, points in the pathway.
AFLP markers were useful for identifying additional resistance loci and they
provided information on the genetic structure of the BC1-derived DH population. The
population average of 32.6% donor parent genome is, as would be expected, higher than
25%, as the BC1 plants were selected (Figure 3.3). As shown in this Figure, percentages
of recurrent parent genome ranged from seven to fifty-six percent. The resistant
genotypes had a higher average percentage of donor parent genome (34.8%) than the
susceptible genotypes (24.6%), but the difference was not that great, considering that
portions of the genome on at least three chromosomes were introgressed into the resistant
lines. Four DH lines were selected, based on their phenotypic resemblance to the
recurrent parent, for accelerated assessment as potential varieties before the AFLP data
were available. The percentage of donor parent genome in these lines ranged from 32 39%. If resources are available, larger populations of BC1-derived lines could be used.
In this case, the percentage of recurrent parent genome would be a more useful selection
criterion.
70
In summary, marker-assisted mapping and transfer of stripe rust resistance QTLs
allowed us make use of limited resources to rapidly develop barley germplasm with
potential for commercial production. Markers that were targets for transfer were
significantly associated with resistance in a genetic background different from the
reference mapping population. The use of high throughput markers - primarily AFLPs
allowed us to detect additional resistance QTLs, including QTLs where the susceptible
parent contributed resistance alleles. Our findings may be useful in view of the many
ongoing efforts in a number of crop species to introgress QTLs. When QTLs mapped in
a reference population are introgressed into new genetic backgrounds, the anticipated
selection responses may not be achieved.
Precautions can be taken during the introgression process to minimize the loss of
QTLs through double crossovers between flanking markers and to guard against the
consequences of imprecise positioning of the QTLs in the original mapping population.
However, the configuration of alleles in the breeding population at loci where alleles
were fixed in the mapping population may not be known. We hypothesize that this was
the case with the new resistance QTLs resistance alleles we detected in the Steptoe
background. An intriguing, but unanswered, question is the relationship between disease
resistance QTLs and race-specific resistance genes. At this point we know nothing
regarding the kinds of genes that are detected as adult plant stripe rust resistance QTLs.
Syntenic relationships in the Triticeae should allow for useful comparative mapping and
extension of findings from one genus to another. For example, the stripe rust resistance
QTL on chromosome 7 (5H) may be homoeologous with one of the durable stripe rust
resistance loci in wheat described by Law and Worland (1997).
71
References
Bakshi JS, Luthra JK (1971) Inheritance of resistance to stripe rust (Puccinia striiformis
west.) in barley. In: Nilan RA (ed.) Proc Symp Barley Genet II. Washington State
University Press, Pullman, Washington, pp 478-483
Barua UM, Chalmers KJ, Hackett CA, Thomas WTB, Powell W, Waugh R (1993)
Molecular mapping of genes determining height, time to heading, and growth habit in
barley (Hordeum vulgare). Heredity 71:177-184
Becker J, Vos P, Kuiper M, Salamini F, Heun M (1995) Combined mapping of AFLP
and RFLP markers in barley. Mol Gen Genet 249:65-73
Chen F, Prehn D, Hayes PM, Mulrooney D, CoreyA, Vivar H (1994) Mapping genes for
resistance to barley stripe rust (Puccinia striiformis f. sp. hordei). Theor Appl Genet
88:215-219
Chen F, Hayes PM (1989) A comparison of Hordeum bulbosum - mediated haploid
production efficiency in barley using in vitro floret and tiller culture. Theor Appl
Genet 77:701-704
Dubin HJ, Stubbs RW (1986) Epidemic spread of barley stripe rust in South America.
Plant Disease 70:141-144
Hayes PM, Chen FQ, Corey A, Pan A, Chen THH, Baird E, Powell W, Thomas W,
Waugh R, Bedo Z, Karsai I, Blake T, Oberthur L (1996a) The Dicktoo x Morex
population: a model for dissecting components of winterhardiness in barley. In: Li
PH, Chen TH(ed.) Plant Cold Hardiness. Plenum Press, New York, USA
Hayes PM, Chen FQ, Kleinhofs A, Kilian A, Mather D (1996b) Barley genome mapping
and its applications. In: Jauhar PP (ed) Methods of Genome Analysis in Plants: CRC
Press. Boca Raton, USA
Hayes PM, Prehn D, Vivar H, Blake T, Comeau A, Henry I, Johnston M, Jones B,
Steffenson B (1996c) Multiple disease resistance loci and their relationship to
agronomic and quality loci in a spring barley population. J.QTL
http ://probe.nalusda.gov:8000/otherdocs/j QTL /index.htm
Iyamabo OE, Hayes PM (1995) Effects of plot type on detection of quantitative trait
locus effects in barley (Hordeum vulgare L.). Plant Breeding 114:55-60
Laurie DA, Pratchett N, Bezant JH, Snape JW (1995) RFLP mapping of five major genes
and eight quantitative trait loci controlling flowering time in a winter x spring barley
cross. Genome 38:575-585
72
Law CN, Worland AJ (1997) The control of adult-plant resistance to yellow rust by the
translocated chromosome 5SB-7BS of bread wheat. Plant Breeding 116:59-63
Lehmann CO, Nover I, Scholz F (1975) The Gatersleben barley collection and its
evaluation. In Gaul H (ed) Proc Symp Barley Genet III. Verlag Karl Thiemig,
Munich, pp 64-69
Line RF (1993) Durability of resistance to Puccinia striiformis in North American wheat
cultivars. p. 332. In Durability of Disease Resistance. Jacobs TH,. Parlevliet JE (eds)
Kluwer Academic Publishers 375 pp
Marshall D, Sutton RL (1995) Epidemiology of stripe rust, virulence of Puccinia
striiformis f. sp. hordei, and yield loss in barley. Plant Disease 70:732-737
Mather D (1995) Barley Steptoe x Morex and Harrington x TR306 basemaps.
gopher://greeng enes. cit. cornell. edu:70/1ftp%3Agnome. [email protected]/
genetics/data/basemaps/
Parlevliet JE, Van Ommeren A (1975) Partial resistance of barley to leaf rust, Puccinia
hordei. II. Relationship between field trials, micro plot test and latent period.
Euphytica 24:293-303
Powell W, Baird E, Booth A, Lawrence P, MacAulay M, Bonar N, Young G, Thomas
WTB, McNicol JW, Waugh R (1996) Single and multi-locus molecular assays for
barley breeding and research. In: Barley Genetics VII Proc of Intl Symp, Saskatoon,
Canada
Qi X, Stam P, Lindhout P (1996) Comparison and integration of four barley genetic
maps. Genome 39:379-394
Ramsay G, Thomas WTB (1992) Two-dimensional representations of linkage groups
using similarity matrices. Abstracts of the 11th International Chromosome
Conference, Edinburgh, UK, 4 -4 8 August, 1992
Statistical Analysis Institute (1988) SAS/STAT User's Guide, 6.03. SAS Institute, Cary,
NC, USA
Tanksley SD, Nelson JC (1996) Advanced backcross QTLs analysis: a method for the
simultaneous discovery and transfer of valuable QTLs from unadapted germplasm
into elite breeding lines. Theor Appl Genet 92:191-203
73
Thomas WTB, Powell W, Waugh R, Chalmers KJ, Barua UM, Jack P, Lea V, Forster
BP, Swanston JS, Ellis RP, Hanson PR, Lance RCM (1995) Detection of quantitative
trait loci for agronomic, yield, grain and disease characters in spring barley (Hordeum
vulgare L.). Theor Appl Genet 91:1037-1047
von Wettstein-Knowles P (1992) Cloned and mapped genes:current status. In Shewry PR
(ed) Barley:genetics, biochemistry, molecular biology, and biotechnology. CAB Int,
Wallingford UK, pp 73-98
Vos P, Hogers R, Bleeker M, Reijans M, van de Lee T, Homes M, Frijters A, Pot J,
Peleman J, Kuiper M, Zabeau M (1995) AFLP: a new technique for DNA
fingerprinting. Nucl Acids Res 23: 4407-4414
Waugh, R, Bonar N, Baird E, Hayes P, Graner A, Thomas WTB, and Powell W (1997)
Homology of AFLP products in three mapping populations of barley. Mol. Gen.
Genet. 255:311-321.
Young ND, Tanksley SD (1989) RFLP analysis of the size of chromosomal segments
retained around the Tm-2 locus of tomato during backcross breeding. Theor Appl
Genet 77:353-359
Zabeau M, Vos P (1993) European Patent Application publication no. EP 0534858
74
CHAPTER 4
CONCLUSION
The first objective of this research was to map genes responsible for resistance to
multiple diseases - stripe rust (Puccinia striiformis f.sp hordei)), leaf rust (Puccinia
hordei), and Barley Yellow Dwarf Virus (BYDV) - in the doubled haploid (DH) progeny
of Shyri x Galena. We generated eight hundred and nine polymorphism using RFLP,
SSR, AFLP and RGAP markers. Our strategy was to use RFLPs and SSRs as anchor .
markers and to use AFLPs and RGAPs to fill the gaps. A subset of the total markers was
used to construct the base map, which consists of 99 markers with an average interval
distance of 10-15 cM. This map covers approximately 92 % of the barley genome.
Locus ordering and distance were in agreement with published maps. SSR
markers are a powerful tool for constructing a skeleton map, due to their high informative
content and the ease of genotyping. However, the catalog of SSRs does not yet provide
full genome coverage. RFLPs markers were of crucial importance for extending our
linkage maps into some regions of the genome.
The high throughput AFLP and RGA markers are useful for rapidly generating
abundant markers to fill gaps. In our hands, AFLPs were not as useful as we had hoped.
Although data scoring was not an issue - band patterns on autoradiographs were clear and
consistent the resulting AFLPs data, when added to established linkage groups of SSR
and RFLP markers, usually led to significant map expansion. AFLPs maybe useful in
cases where marker saturation is required, and the resulting AFLP products will be
cloned and converted to a more stable format. RGAP markers, on the other hand, were
75
generally high quality data points. The RGAP marker technology is attractive from a
number of standpoints, and research with this class of markers should be continued.
Using the 99-point base map, we determined the number, genome location, and
effects of loci determining resistance to the three diseases. QTLs were identified on
chromosomes 2, 3, 5 and 6 for stripe rust resistance; on chromosome 1 for leaf rust
resistance; and chromosomes 1, 3, 4, and 5 for BYDV resistance. Leaf rust resistance
showed qualitative inheritance and subsequently mapped as single locus. Two-locus
epistasis was important in determining resistance to stripe rust, but epistasis was not
significant in the case of leaf rust or BYDV resistance.
Shyri clearly has a different major different QTL for stripe resistance than
Calicuchima-sib (Chen et al. 1994) and CMB643 (Hayes et al. 1998). The smaller-effect
stripe rust resistance QTL on chromosome 3 was coincident with QTLs detected in
CMB643 (Hayes et al. 1998), Kold (Hayes et al.1998), and Steptoe derivatives (Toojinda et
al. 1997). The smaller-effect QTLs on chromosomes 4 and 6 were only detected in
Calicuchima-sib (Chen et al. 1994) and Shyri respectively. We have observed significant
two locus interactions between small-effect QTLs and large effect QTLs in multiple
mapping populations: Calicuchima-sib x Bowman (Chen et al.1994), Gobernadora x
CMB643 (Hayes et al. 1998) , Kold x Colter (Hayes et al. 1998), and now in Shyri x
Galena. These significant interactions indicate the importance of epistasis in determining
stripe rust resistance. Epistasis of loci determining stripe rust resistance was also reported
by Chen and Line (1998).
The chromosome 1 QTL was the single largest determinant for leaf rust resistance
in Shyri. When the qualitative data were treated as a single locus effects, we mapped the
76
Rphxs locus to approximately the same location as the Rph3 (Jin et al. 1993) and Rphxc
(Hayes et al. 1996) loci. We identified three major QTLs, and one secondary QTL, for
serotype non-specific resistance to BYDV. Resistance was determined by additive effects
at these loci. The contribution of two major QTLs on chromosome 1 and 5 was consistent
with all trait measurements. The secondary QTL on chromosome 3 was coincident with a
QTL determining plant height. At the level of resolution afforded by this mapping
population, we cannot determine if this coincidence is due to linkage or pleiotropy.
However, the genome location of this QTL is different from Yd2 locus. Therefore, BYDV
resistance in this germplasm is clearly not due to the Yd2 locus. The chromosome 4 QTL
was detected with the dwarfness score and plant height reduction, but it was not significant
with the tillering score data.
We demonstrated linkage relationships of RGAPs with genes conferring
quantitative and qualitative resistance to two fungal pathogens: Puccinia striiformis fsp.
hordei and Puccinia hordei, and quantitative resistance to a viral pathogen, BYDV. The
coincidence of RGAPs with QTL peaks, or loci, and a qualitative locus determining disease
resistance demonstrates the power of the RGAP technique for mapping qualitative and
quantitative disease resistance genes.
The clustering of RGAPs, and their linkage relationships with qualitative and
quantitative resistance genes, is evidence for the existence of resistance gene clusters in
barley and may be based on commonalities of qualitative and quantitative resistance genes.
The availability of abundant RGAP markers should facilitate isolation and characterization
of both qualitative and quantitative resistance. This will provide tools for answering
fundamental questions regarding the genetic basis of the two classes of resistance.
77
We found that quantitative stripe rust resistance was more complex than the single
large-effect QTLs that are detected in reference mapping populations. When chromosome 4
and 7 stripe rust resistance QTLs mapped in Cali-sib x Bowman (Chen et al., 1994) were
introgressed into an unrelated genetic background (Steptoe), the resistance QTLs that were
introgressed were significant determinants of resistance, as were previously undetected
QTLs. Therefore, small-effect QTLs should be considered in resistance QTL introgression
and pyramiding experiments.
This experiment demonstrated the efficiency of marker-assisted disease resistance
breeding in barley. Molecular markers also increase the efficiency of backcrossing, by
allowing for selection of genotypes with maximum percentage of the recurrent parent
genome. In our case, the experimental population was not large enough to make effective
use of such information.
With this series of experiments, we have demonstrated the power of applied
genomics to locate and use disease resistance genes. We are optimistic that the RGAP
technique, with some modifications, will allows us to increase the efficiency of resistance
gene mapping. It should also enable us to systematically pyramid multiple loci,
determining resistance to multiple diseases, in agronomically useful germplasm.
78
BIBLIOGRAPHY
Alemayehu F, Parlevliet JE (1996) Variation for resistance to Puccinia hordei in
Ethiopian barley landraces. Euphytica 90:365-370.
Bakshi JS, Luthra JK (1971) Inheritance of resistance to stripe rust (Puccinia
Striiformis west.) in barley. In: Nilan RA (ed.) Proc Symp Barley Genet II.
Washington State University Press, Pullman, Washington, pp 478-483
Barua UM, Chalmers KJ, Hackett CA, Thomas WTB, Powell W, Waugh R
(1993) Molecular mapping of genes determining height, time to heading, and
growth habit in barley (Hordeum vulgare). Heredity 71:177-184
Barua UM, Chalmers KJ, Thomas WTB, Hackett CA, Lea V, Jack P, Forster BP,
Waugh R, Powell W (1993) Molecular mapping of genes determining height,
time to heading, and growth habit in barley (Hordeum vulgare). Genome 36:
1080-1087.
Bassam BJ, Bentley S (1994) DNA fingerprinting using arbitrary primer
technology (APT): a tool or a torment. Australasian Biotechnology 4: 232-236
Becker J, Heun M (1995) Barley microsatellites: allele variation and mapping.
Plant Mol Biol 27:835-845.
Becker J, Vos P, Kuiper M, Salamini F, Huen M (1995) Combined mapping of
AFLP and RFLP markers in barley. Mol. Gen. Genet. 249:65-73.
Bennetzen JL, Freeling M (1997) The unified grass genome: synergy in synteny.
Genome Research 7: 301-306.
Bezant J, Laurie D, Pratchett N, Chojecki J, Kearsey M (1997) Mapping QTL
controlling yield and yield components in a spring barley (Hordeum vulgare
L.) cross using marker regression. Molecular Breeding 3: 29-38.
Borst P, Greaves DR (1987) Programmed gene rearrangements altering
expression. Science 235:658-667.
Bowles DJ (1990) Defense-related proteins in higher plants. Annu. Rev. Biochem.
59: 873-907.
Browning JA, Simons MD, Torres E (1977) Managing host genes:epidemiologic
and genetic concepts. In Horsfall JG, Cowling EB (eds). Plant Disease:an
advanced treatise. Vol.1 . Academic Press New York.
79
Buschges R, Hollricher K, Panstruga R, Simons G, Wolter M, Frijters A, van
Daelen R, van der Lee T, Diergaarde P, Groenendijk J, Topsch S, Vos P,
Salamini F, Schulze-Lefert P (1997) The barley Mlo gene: A novel control
element of plant pathogen resistance. Cell 88:695-705.
Chen F, Hayes PM (1989) A comparison of Hordeum bulbosum mediated
haploid production efficiency in barley using in vitro floret and tiller culture.
Theor Appl Genet 77:701-704.
Chen F, Prehn D, Hayes PM, Mulrooney D, CoreyA, Vivar H (1994) Mapping
genes for resistance to barley stripe rust (Puccinia striiformis f. sp. hordei).
Theor Appl Genet 88:215-219.
Chen XM, Line RF, Leung H (1998) Genome scanning for resistance-gene
analogs in rice, barley, and wheat by high-resolution electrophoresis. Theor.
Appl. Genet 97:345-355.
Chen XM, Line RF, Leung 1-1 (1995) Virulence and polymorphic DNA
relationships of Puccinia striiformis f.sp. hordei to other rusts.
Phytopathology 85:1335-1342.
Collins NC, Paltridge NG, Ford CM, Symons RH (1996) The Yd2 gene for barley
yellow dwarf virus resistance maps close to the centromere on the long arm of
barley chromosome 3. Theor. Appl. Genet 92:858-864.
Concibido VC, Young ND, Lange DA, Denny RL, Danesh D, Orf JH (1996)
Targeted comparative genome analysis and qualitative mapping ofa major
partial-resistance gene to the soybean cyst nematode. Theor. Appl.Genet.
93:234-241.
Conner RL, Whelan EDP, MacDonald MD (1989) Identification of sources of
resistance to common root rot in wheat-alien amphiploid and chromosome
substitution lines. Crop Science 29:916-919.
Crute IR (1985) The genetic basic of relationships between microbial parasites
and their host. Pages 62 in: Mechanisms of resistance in plant diseases. Kluwer
Academic Press.
D'Acry CJ (1995) Symptomatology and host range of barley yellow dwarf. Pages
9-28 in: Barley Yellow Dwarf: 40 Years of Progress. D'Acry CJ and Burnett
PA, eds. The American Phytopathologal Society, St. Paul, MN.
Dahleen LS (1997) Mapped clone sequences detecting differences between 28
North American Barley cultivars. Crop Science 37:952-957.
80
Dahleen LS, Hoffman DL, Dohrmann J, Gruber R, Franckowiak J (1997) Use of a
subset of double-haploid lines for RAPD interval mapping in barley. Genome
40: 626-632.
Dangl JL (1995) Piece de resistance: novel classes of plant disease resistance
genes. Cell 80:363-366.
Dean C and Schmidt R (1995) Plant genome: a current molecular description.
Annual Review of Plant Physiology and Plant Molecular Biology 46: 395-418.
Dingerdissen AL, Geiger HH, Lee M, Schechert A, Welz HG (1996) Interval
mapping of genes for quantitative resistance of maize to Setoria turcica, cause
of norther leaf blight, in a tropical environment. Molecular
Breeding 2:143-156.
Dubin HJ, Stubbs RW (1986) Epidemic spread of barley stripe rust in South
America. Plant Dis70:141-144.
El-Kharbotly A, Leonards-Schippers C, Huigen DJ, Jacobsen E, Pereira A (1994)
Segregation analysis and RFLP mapping of the RI and R3 alleles conferring
race-specific resistance to Phytohthora infestans in progeny of dihaploid
potato plants. Mol.Gen.Genet. 242:749-754.
Ellis JG, Lawswnce GJ, Peacock WJ, and Pryor AJ (1998) Approaches to cloning
plant genes conferring resistance to fungal pathogens. Annual Review of
Phytopathology 26:245-263.
Ferreira ME, Rimmer SR, Williams PH, Osborn TC (1995) Mapping loci
controlling Brassica napus resistance to Leptosphaeria maculans under
different screening conditions. Phytopathology 85:213-217.
Feuerstein U, Brown AHD, Burdon JJ (1990) Linkage of rust resistance genes
from wild barley (Hordeum spontaneum) with isozyme markers. Plant Breed
104:318-324.
Garner A, Tekauz A (1996) RFLP mapping in barley of a dominant gene
conferring resistance to scald (Rhynchosporium secalis). Theor. Appl. Genet.
93:421-425.
Geiger HH, Heun M (1989) Genetics of quantitative resistance to fungal diseases.
Annu. Rev. Phytopathol. 27:317-341.
Giease H, Holm-Jensen AG, Jensen HP, Jensen J (1993) Localization of the
Laevigatum powdery mildew resistance gene to barley chromosome 2 by the
use of RFLP markers.Theor.Appl.Genet. 85:897-900.
81
Giease H, Jorgensen J, Jensen HP, Jensen J (1981) linakge relationships of ten
powdery mildew resistance genes on barley chromosome 5. Hereditas 95:4350.
Giese H, Holm-Jensen AG, Mathiassen H, Kjaer B, Rasmussen SK, Bay H,
Jensen J (1994) Distribution of RAPD markers on a linkage map of barley.
Hereditas Landskrona 120: 267-273.
Graner A, Bauer E (1993) RFLP mapping of the ym4 virus resistance gene in
barley. Theor.Appl.Genet. 86:689-693.
Graner A, Jahoor A, Schondelmaier J, Siedler H, Pillen K, Fischbeck G, Wensel
G, Herrmann G (1991) Construction of an RFLP map of barley. Theor.Appl.
Genet. 83:250-256.
Han F, Kleinhofs A, Kilian A, Ullrich SE (1997) Cloning and mapping of a
putative barley NADPH-dependent HC-toxin reductase. Molecular Plant
Microbe Interactions 10: 234-239.
Han F, Romagosa I, Ullrich SE, Jones BL, Hayes PM, Wesenberg DM (1997)
Moleular marker-assisted selection for malting quality traits in barley.
Molecular Breeding 3: 427-437.
Harder DE, Harber S (1992) Barley yellow dwarf virus. Pages 379 in: Oats
Science and Technology. Marshall HG, Sorrells ME, eds. American Society
of Agronomy, Inc, Madison, WI.
Hayes PM, Chen FQ, Corey A, Pan A, Chen THH, Baird E, Powell W, Thomas
W, Waugh R, Bedo Z, Karsai I, Blake T, Oberthur L (1996a) The Dicktoo x
Morex population: a model for dissecting components of winterhardiness in
barley. In: Li PH, Chen TH(ed.) Plant Cold Hardiness. Plenum Press, New
York, USA
Hayes PM, Chen FQ, Kleinhofs A, Kilian A, Mather D (1996) Barley genome
mapping and its applications. In:Jauhar PP (ed) Methods of genome analysis in
plants. CRC Press. Boca Raton, USA
Hayes PM, Liu BH, Knapps SJ, Chen F, Jones B, Blake T, Franckowiak J,
Rasmusson D, Sorrells M, Ullrich SE, Wesenberg D, Kleinhofs A (1993)
Quantitative trait locus effects and environmental interaction in a sample of
North American barley germplasm. Theor. Appl. Genet. 87: 392-401.
Hayes PM, Prehn D, Vivar H, Blake T, Comeau A, Henry I, Johnston M, Jones B,
Steffenson B (1996) Multiple disease resistance loci and their relationship to
agronomic and quality loci in a spring-barley population. J.QTL
http://probe.nalusda.gov:8000/otherdocs/jQTL/index.htm
82
Heun M, Kennedy AE, Anderson JA, Lapitan NLV, Sorrels ME, Tanks ley SD
(1991) Construction of a restriction fragment length polymorphism map for
barley (Hordeum vulgare). Genome 34:437-447.
Hoffbeck MD, Openshaw SJ, Geadelmann JL, Peterson RH, Stuthman DD (1995)
Backcrossing and intermating in an exotic X adapted cross of maize. Crop
Science 35:1359-1364.
Holloway J, Knapp SJ (1994) Gmendel 3.0 Users Guide. [email protected]
Hooker AL (1985) Corn and sorghum rust. In The cereal rusts. Vol.2. Diseases,
Distribution, Epidemiology, and Control (Roelfs AP and Bushnell WR, eds)
pp 207-233.
Horvath DP, Dahleen LS, Stebbing JA, Penner G (1995) A co-domonant PCRbased marker for assisted selection of durable stem rust resistance in barley.
Crop Science 35: 1445-1450.
Iyamabo OE, Hayes PM (1995) Effects of plot type on detection of quantitative
trait locus effects in barley ( Hordeum vulgare L.). Plant Breeding 114:55-60
Jin Y, Cui GH, Steffenson BJ, Franckowiak JD (1996) New leaf rust resistance in
barley and their allelic and linkage relationships with other Rph genes.
Phytopathology 86:887-890.
Jin Y, Statler GD, Franckowiak JD, Steffenson BJ (1993) Linkage between leaf
rust resistance genes and morphological markers in barley. Phytopathology
83:230-233.
Johnson R (1981) Durable resistance: definition of, genetic control, and
attainment. Phytopathology 71:567-568.
Jorgensen JH (1992) Discovery, characterization and Exploitation of MTh
powdery mildew resistance in barley. Euphytica 63:141-152.
Kanazin V, Marex LF, Shoemaker RC (1996) Resistance gene analogs are
conserved and clustered in soybean. Proc. Natl. Acad. Sci. USA 93:1174611750.
Kilian A, Chen J, Han F, Steffenson B, Kleinhofs A (1997) Towards map-based
cloning of the barley stem rust resistance gene Rpgl and Rpg4 using rice as an
intergenomic cloning vehicle. Plant Molecular Biology 35:187-195.
Kleinhofs A (1991) The NABGMP mapping progress report, spring 1992. Barley
Genetics Newsletter 21: 38-48.
83
Kleinhofs A, Kilian A, Saghai Maroof M, Biyashev R, Hayes PM, Chen F,
Lapitan A, Fenwick A, Blake T, Kanazin V, Ananiev E, Dahleen L, Kudrna D,
Bollinger J, Knapp SJ, Liu B, Sorrel ls M, Heun M, Franckowiak J, Hoffman
D, Skadsen R, Steffension B (1993) A molecular, isozyme, and morphological
map of the barley (Hordeum vulgare) genome. Theor. Appl. Genet. 86:705712.
Lamb CJ (1994) Plant disease resistance genes in signal perception and
transduction. Cell 76:419-422.
Laurie DA, Pratchett N, Bezant JH, Snape JW (1995) RFLP mapping of five
major genes and eight quantitative trait loci controlling flowering time in a
winter x spring barley cross. Genome 38:575-585.
Law CN, Worland AJ (1997) The control of adult-plant resistance to yellow rust
by the translocated chromosome 5SB-7BS of bread wheat. Plant Breeding
116:59-63.
Lefebvre V, Palloix A (1996) Both epistatic and additive effects of QTLs are
involved in polygenic induced resistance to disease: a case study, the
interaction pepper-Phytophthora capsici Leonian. Theor. Appl.Genet. 93:503511.
Lehmann CO, Nover I, Scholz F (1975) The Gatersleben barley collection and its
evaluation. In Gaul H (ed) Proc Symp Barley Genet III. Verlag Karl Thiemig,
Munich, pp 64-69.
Leister D, Ballvora A, Salamini F, Gebhardt C (1996) A PCR-based approach for
isolsting pathogen resistance genes from potato with potential for wide
application in plants. Nature Genetics 14:421-429.
Leister D, Kurth J, Laurie DA, Yano M, Sasaki T, Devos K, Graner A, SchulzeLefert P (1998) Rapid reorganization of resistance gene homologues in cereal
genomes. Proc. Natl. Acad. Sci. USA 95:370-375.
Line RF (1993) Durability of resistance to Puccinia striiformis in North American
wheat cultivars. p. 332. In Durability of Disease Resistance. Jacobs TH.,
Parlevliet JE (eds) Kluwer Academic Publishers 375 pp.
Liu ZW, Biyashev RM, Sagai-Maroof MA (1996) Development of simple
sequence repeat DNA markers and their integration into a barley linkage map.
Theor. Appl.Genet. 93: 869-876.
Liu ZW, Biyashev RM, Saghai Maroof MA (1996) Development of simple
sequence repeat markers and their integration into a barley linkage map. Theor.
Appl. Genet. 93:869-876.
84
Maisonneuve B, Bellec Y, Anderson P, Michelmore RW (1994) Rapid mapping
of two genes for resistance to downy mildew from Lactuca serriola to existing
clusters of resistance genes. Theor.Appl.Genet. 89:96-104.
Marek LF, Shoemaker RC (1997) BAC contig development by fingerprint
analysis in soybean. Genome 40: 420-427.
Maroof MAS, Biyashev RM, Yang GP, Zhang Q, Allard RW (1994)
Extraordinarily polymorphic microsatellite DNA in barley: species diversity,
chromosomal locations, and population dynamics. Proc. Natl. Acad. Sci. USA
91: 5466-5470.
Marshall D, Sutton RL (1995) Epidemiology of stripe rust, virulence of Puccinia
striiformis f. sp. hordei, and yield loss in barley. Plant Disease 70:732-737.
Martin GB (1996) Molecular cloning of plant disease resistance genes. In:
Plantmicrobe interactions vol. 1 . Martin GB, Herrera-Hestrella L, Rosales LS,
Rivera- Bustamante R, Neuenschwander U, Lawton K, Ryals J, Schardl
CL, Kronstad JW, Thomashow LS, Weller DM, Philips DA, Streit WR, Prome
JC, Demont N, Stacey G (eds).Chapman & Hall; New York; USA.
Mather D (1995) Barley Steptoe x Morex and Harrington x TR306 basemaps.
gopher://greengenes.cit.cornell.edu:70/1ftp %3Agnome.agrenv.mcgill.ca
@pub/genetics/data/basemaps/
Mather DE, Tinker NA, LaBerge DE, Edney M, Jones BL, Rossnagel BG, Legge
WG, Briggs KG, Irvine RB, Falk DE, Kasha KJ (1997) Region of the genome
that affect grain and malting quality in a North American two-row barley
cross. Crop Science 37: 544-554.
Mazur BJ, Tingey SV (1995) Genetic mapping and introgression of genes of
agronomic importance. Current Opinion in Biotechnology 6: 175-182.
Melchers LE, Parker JH (1992) Rust resistance in winter wheat varieties. US Dep
Agric Bull 1046
Melchinger AE, Utz HF, Schon CC (1998) Quantitative trait locus (QTL) mapping
using testers and independent population samples in maize reveals low power of
QTL detection and large bias in estimates of QTL effects. Genetics 149:383403.
Michelmore RW (1995) Molecular approaches to manipulation of disease
resistance genes. Ann. Rev. Phytopath 33:393-428.
85
Mitchell SE, Kresovich S, Jester CA, Hernandez CJ, Szewc-McFadden AK
(1997) Application of multiplex PCR and fluorescence-based, semi-automated
allele sizing technology for genotyping plant genetic resources. Crop
Science 37: 617-624.
Moharramipour S, Tsumuki H, Sato K, Yoshida H (1997) Mapping resistance to
cereal aphids in barley. Theor. Appl. Genet. 94: 592-596.
Morgante M, Rafalski A, Bidle P, Tingey S, Oliveri AM (1994) Genetic mapping
and variability of seven soybean simple sequence repeat loci. Genome 37:763769.
Noli E, Salvi S, Tuberosa R (1997) Comparative analysis of genetic relationships
in barley based on RFLP and RAPD markers. Genome 40: 607-616.
Ordon F, Friedt W (1993) Mode of inheritance and genetic diversity of BaMMV
resistance of exotic barley germplasms carrying genes different from ym4.
Theor. Appl. Genet. 86:229-233.
Ori N, Eshed Y, Paran I, Presting G, Aviv D, Tanksley S, Zamir D, Fluhr R
(1997) The I2C family from the wilt disease resistance locus 12 belongs to the
nucleotide binding, leucine-rich repeat superfamily of plant resistance genes.
Plant Cell 9:521-532.
Paran I, Kesseli R, Michelmore R (1991) Identification of restriction fragment
length polymorphism and random amplified polymorphic DNA markers linked
to downy mildew resistance genes in lettuce, using near-isogenic lines.
Genome 34:1021-1027.
Parlevliet JE (1978) Future evidence of polygenic inheritance of partial resistance
in barley to leaf rust, Puccinia hordei. Euphytica 27:369-379.
Parlevliet JE, Van Ommeren A (1975) Partial resistance of barley to leaf rust,
Puccinia hordei. II. Relationship between field trials, micro plot test and latent
period. Euphytica 24:293-303.
Powell W, Baird E, Booth A, Lawrence P, MacAulay M, Bonar N, Young G,
Thomas WTB, McNicol JW, Waugh R (1996) Single and multi-locus
molecular assays for barley breeding and research. In: Barley Genetics VII
Proc. of Intl. Symp., Saskatoon, Canada.
Powell W, Thomas WTB, Baird E, Lawrence P, Booth A, Harrower B, McNicol
JW, Waugh R (1997) Analysis of quantitative traits in barley by the use of
amplified fragment length polymorphisms. Heredity 79: 48-59.
86
Pryor T (1987) The origin and structure of fungal disease resistance genes in
plants. Trends in Genetics 3:157-161.
Pryor T, Ellis J (1993) The complexity of fungal resistance genes in plants.
Advances in Plant Pathology 10:281-304.
Qi X (1998) Identification and mapping of genes for partial resistance to Puccinia
hordei Otth in barley. Thesis Wageningen Agricultural University.
Qi X, Stam P, Lindhout P (1996) Comparison and integration of four barley
genetic maps. Genome 39:379-394.
Qi X, Stam P, Lindhout P (1998) Use of locus-specific AFLP markers to
construct a high-density molecular map in barley. Theor. Appl. Genet. 96: 376384.
Ramsay G, Thomas WTB (1992) Two-dimensional representations of linkage
groups using similarity matrices. Abstracts of the 11th International
Chromosome Conference, Edinburgh, UK, 4 -4 8 August, 1992.
Reuvein R, Dudai N, Putievsky E, Elmer WH, Wick RL (1997) Evaluation and
identification of basil germplasm for resistance to Fusarium oxysporum f.sp.
Basilicum. Plant Disease 81:1077-1081.
Roder MS, Plaschke J, Konig SU, Borner A, Sorrells ME, Tanksley SD, Ganal
MW (1995) Abundance, vaiability and chromosomal location of
microsatellites in wheat. Molecular and General Genetics 246: 327-333.
Russell J, Fuller J, Young G, Thomas B, Taramino G, Macaulay M, Waugh R,
Powell W (1997) Discriminating between barley genotypes using
microsatellite markers. Genome 40:442-450.
Russell JR, Fuller JD, Macaulay M, Hatz BG, Jahoor A, Powell W, Waugh R
(1997) Direct comparison of levels of genetic variation among barley
accessions detected by RFLPs, AFLPs, SSRs, RAPDs. Theor. Appl. Genet.
95:714-722.
Saghai-Maroof MA, Zhang Q, Biyashev R (1995) Comparison of restriction
fragment length polymorphisms in wild and cultivated barley. Genome 38:
298-306.
Saxena KMS, Hooher AL (1968) On the structure of a gene for disease resistance
in maize. Proc. Natl. Acad. Sci. USA 61:1300-1305.
87
Sharma H, Ohm H, Goulart L, Lister R, Appels R, Benlhabib 0 (1995)
Introgression and characterization of barley yellow dwarf virus resistance from
Thinopyrum intermedium into wheat. Genome 38:406-413.
Singh RP (1993) Genetic association of gene Bdv 1 for tolerance to barley yellow
dwarf virus with genes Lr34 and Yr18 for adult plant resistance to rusts in
bread wheat. Plant Dis 77:1103-1106.
Statistical Analysis Institute (1988) SAS/STAT User's Guide, 6.03. SAS Institute,
Cary, NC, USA
Steffenson BJ, Hayes PM, Kleinhofs A (1996) Genetics of seedling and adult
plant resistance to net blotch (Pyrenophora teres f. teres) and spot blotch
(Cochiobolus sativus) in barley. Theor. Appl.Genet. 92: 552-558.
Tanksley SD (1993) Mapping polygenes. Annu. Rev. Genet 27:205-233.
Tanksley SD, Ganal MW, Martin GB (1995) Chromosome landing: a paradigm
for map-based gene cloning in plants with large genomes. Trends Genet. 11:
63-68.
Tanksley SD, Nelson JC (1996) Advanced backcross QTLs analysis: a method for
the simultaneous discovery and transfer of valuable QTLs from unadapted
germplasm into elite breeding lines. Theor Appl Genet 92:191-203.
Tanksley SD, Young ND, Paterson AH, Bonierbale MW (1989) RFLP mapping
in plant breeding: new tools for an old science. Bio/Technology 7:275-264.
Thomas WTB, Powell W, Waugh R, Chalmers KJ, Barua UM, Jack P, Lea V,
Forster BP, Swanston JS, Ellis RP, Hanson PR, Lance RCM (1995) Detection
of quantitative trait loci for agronomic, yield, grain and disease characters in
spring barley (Hordeum vulgare L.). Theor Appl Genet 91:1037-1047
Thresh JM (1998) In memory of James Edward Vanderplank 1909-1997. Plant
Pathology 47: 114-115.
Tinker NA, Mather D (1995) Methods for QTL analysis with progeny replicated
in multiple environments. J.QTL:http://probe.nalusda.gov:8000/otherdocs/jqt1/
Toojinda T, Baird E, Booth A, Broers L, Hayes PM, Powell W, Thomas W, Vivar
H, Young G (1997) Introgression of quantitative trait loci (QTLs) determing
stripe rust resistance in barley: an example of marker-assisted line
development. Theor Appl Genet 96:123-131.
Touzet P, Winkler RG, Helentjaris T (1995) Combined genetic and physiological
analysis of a locus contributing to quantitative variation. Theor. Appl. Genet.
91:200-205.