film
:«m

 

 

 

1312:5341 a...» 3...... .A..a......wfi.2é;udirhtjmuh

.3

 

 

 

 

 

LIBRARY
Michigan State
University

 

 

This is to certify that the
dissertation entitled

QTL ANALYSIS OF GENETIC RESISTANCE TO WHITE
MOLD (SCLEROTINIA SCLEROTIORUM) IN COMMON
BEAN (PHASEOLUS VULGARIS)

presented by

MARCIO ENDER

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Plant Breeding and Genetics
Crop and Soil Sciences

 

 

Major Professor’s Signature /

% Q/ @-
/

07/02/2003

 

Date

MSU is an Affirmative Action/Equal Opportunity Institution

. .n-o-C-Int-l-C-O-I-C-O-I-l-O-n-I-v—h- .-

PLACE IN RETURN Box to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJClRC/DatoDuo.p65-p.15

QTL ANALYSIS OF GENETIC RESISTANCE TO WHITE MOLD
(SCLEROTINIA SCLEROTIORUM) IN COMMON BEAN
(PHASEOLUS VULGARIS)

BY

Marcio Ender

A DISSERTATION

Submitted to
Michigan State University
In partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Crop and Soil Sciences

2003

ABSTRACT

QTL ANALYSIS OF GENETIC RESISTANCE TO WHITE MOLD (SCLEROTINIA
SCLEROTIORUM) IN COMMON BEAN (PHASEOLUS VULGARIS)

By

Marcio Ender

White mold, caused by the necrotrophic fungus Sclerotinia sclerotiorum
(Lib.) De Bary, is a serious disease of common bean (Phaseolus vulgaris L.). The
use of resistant varieties is a preferred disease control strategy for white mold but
has been a difficult management strategy to implement as no major genes for
resistance have been identified in common bean. The objectives of this study
were to: i) to identify AFLP and RAPD markers linked to quantitative trait loci
(QTL) associated with resistance to S. sclerotiorum in a Middle American
mapping population of common bean; ii) map the putative QTL to the bean
integrated map; iii) identify QTL associated with important agronomic traits and
their relationship with resistance to white mold; iv) test the potential of marker
assisted selection using previously identified markers linked to QTL for
resistance to white mold in bean, v) introgress genetic variability from landrace
and wild genotypes into adapted genetic background of cultivated common bean.

The Bunsi/Raven RlL population was evaluated for AFLP and RAPD
markers linked to QTL for resistance to white mold in the field and greenhouse.
Multi-trait bulking segregant analysis including disease severity, yield and days to

flowering was used to detect markers linked to QTL for resistance to white mold.

QTL for disease severity index (DSI) using composite interval mapping were
detected on linkage groups 82, 85, B7 and 88. The QTL for DSI on 82 and B7
supported results from previous studies, suggesting that QTL for DSI are located
in those genomic regions and can be used in marker assisted selection.
Important QTL were detected for agronomic traits: days to maturity (R2=28.5%)
on 82, days to flowering (R2=35.7) on 87, lodging (R2=29.8) on 87, seed size
(R2=20.1%) on 82 and (R2=25.7%) on 87 and yield (R2=9.7%) on 32, and
(R2=12.7%) on 88, and (82:21.5%) on 87.

Marker assisted selection (MAS) was performed in a Bunsi/Midland
recombinant inbred line (RIL) population using previously identified markers
linked to QTL for resistance to white mold, in the cultivar Bunsi. The two groups
of RlLs (selected and control) resulting from MAS were significantly different for
DSI (p=0.03), supporting the potential use of MAS to improve resistance to white
mold in common bean.

The inbred backcross method was used to introgress into adapted bean
varieties putative resistance to white mold from exotic germplasm including
landraces and wild bean types. Two populations were selected, based on visually
rated agronomic traits, from seven populations of inbred backcross lines. The two
selected populations demonstrated putative resistance to white mold in
greenhouse using the straw test. The sources of resistance identified in this
study varied in defense response proteins and will be used to further enhance

resistance for white mold in common bean.

DEDICATION

To Andrea, Anita, Afonso, Geneci and Georgius.

iv

ACKNOWLEDGMENTS

I would like to express a very special thanks to my major advisor, Dr.
James D. Kelly, for giving me the opportunity to pursue the Ph.D. degree in his
bean breeding program. His friendship, unsurpassed knowledge regarding
common bean and experience in plant breeding made me fortunate for having
such an example to follow as a mentor. I wish to thank the committee members
Dr. Amy lezzoni, Dr. David Douches and Dr. Ray Hammerschmidt for their
contributions during the development of my research project. I thank the faculty
members of Crop and Soil Sciences Department and Plant Breeding and
Genetics program, especially Dr. Dechun Wang for all support during the QTL
analysis.

I want to thank the bean group, Jerry and Norm for all assistance in the
white mold trials, Mark, Veronica, Pat, Halima, Shitaye, Andrea, Emmalea,
Belinda, Gerardine, Celeste, Esther, Judy, Maeli, Eveline and Qi for their
friendship, help in the field and in the lab and interesting discussions.

Thanks to my friends in the Brazilian and Latin American Communities for
being my family in Michigan and all the support.

I am very thankful to Coordenacao de Aperfeicoamento de Pessoal de
Nivel Superior (CAPES), Ministry of Education of Brazil for the financial support
and to Santa Catarina State University for the opportunityto pursue the Ph.D.
Degree.

I thank my family and friends and especially my wonderful wife Andrea for

all the love, friendship and support.

TABLE OF CONTENTS

LIST OF FIGURES ............................................................................................... ix
LIST OF APPENDICES ....................................................................................... xi
INTRODUCTION ................................................................................................... 1
References ........................................................................................................ 8
Chapter I
IDENTIFICATION OF QTL ASSOCIATED TO WHITE MOLD RESISTANCE IN
COMMON BEAN ................................................................................................. 10
Introduction ...................................................................................................... 10
Materials and Methods .................................................................................... 21
Results and Discussion ................................................................................... 30
References ...................................................................................................... 58
Chapter II
MARKER-ASSISTED SELECTION FOR WHITE MOLD RESISTANCE ............ 65
IN COMMON BEAN. ........................................................................................... 65
Introduction ...................................................................................................... 65
Materials and Methods .................................................................................... 72
Results and Discussion ................................................................................... 76
References ...................................................................................................... 88
Chapter "I
USE OF |N8RED BACKCROSS METHOD TO INTRODUCE RESISTANCE TO
WHITE MOLD FROM EXOTIC GERMPLASM INTO COMMON BEAN .............. 91
Introduction ...................................................................................................... 91
Materials and Methods .................................................................................. 101
Results and Discussion ................................................................................. 111
References .................................................................................................... 1 31
GENERAL DISCUSSION AND FUTURE RESEARCH ..................................... 136
General Discussion ....................................................................................... 136
Future Research ............................................................................................ 142
References .................................................................................................... 146

vi

Table 1.1.

Table 1.2.

Table 1.3.

Table 1.4.

Table 1.5.

Table 1.6.

Table 1.7.

Table 1.8.

Table 1.9 .

LIST OF TABLES

DNA bulking strategy1 used for detecting QTL for white mold
resistance in common bean based on disease severity (DSI), yield
and days to flowering, in 2001. .......................................................... 28

Means, range and heritability (h2) of disease severity index (DSI) and
agronomic traits in 98 F4 derived (RIL) from the cross Bunsi/Raven,
grown in Montcalm, MI, in 2001 and 2002. ........................................ 33

Monthly precipitation from June to September in 2001 and 2002 in
Montcalm, Ml. .................................................................................... 34

Markers linked to QTL for resistance to white detected through single
marker analysis in a 98 derived RIL population from a Bunsi/Raven
cross, grown in Montcalm, MI, 2001. ................................................. 35

Markers linked to QTL for resistance to white detected through single
marker analysis in a 98 derived RIL population from a Bunsi/Raven
cross, grown in Montcalm, MI, 2002 .................................................. 37

Markers linked to QTL for resistance to white detected through single
marker analysis in a 98 derived RIL population from a Bunsi/Raven
cross, grown in Montcalm, MI, 2001 and 2002 .................................. 38

Markers identified on linkage group 82, most closely linked to QTL for
resistance to white mold and agronomic traits, including phenotypic
variability associated with the QTL and additive effect of the presence
of the marker. .................................................................................... 39

Markers identified on linkage group 85, most closely linked to the QTL
for resistance to white mold and agronomic traits, including
phenotypic variability associated with the QTL and additive effect of
the presence of the marker. ............................................................... 40

Markers identified on linkage group 88, most closely linked to the
QTL for resistance to white mold and agronomic traits, including
phenotypic variability associated with the QTL and additive effect of
the presence of the marker. ............................................................... 41

Table 1.10. Markers identified on linkage group 87, most closely linked to the

QTL for resistance to white mold and agronomic traits, including
phenotypic variability associated with the QTL and additive effect of
the presence of the marker. ............................................................... 48

vii

Table 1.11. Phenotypic correlations between Disease Severity index and
phenological and agronomic traits in a 98 F4 derived RIL population
from a 8unsi/Raven cross grown in Montcalm in 2001 and 2002. ..51

Table 2.1. Disease severity index (DSI), disease incidence (DI), resistance to
oxalate (ROX), of the two MAS groups of 8unsi/Midland RlLs grown
under white mold pressure at Montcalm, MI, in 2001 and 2002. ......... 78

Table 2.2. Combined yield, seed size, days to flowering, days to maturity, plant
architecture and lodging of the two MAS groups of RlLs (selected and
control), grown under white mold pressure in Montcalm, MI, in 2001
and 2002. ............................................................................................ 80

Table 2.3. Phenotypic correlation1 of disease severity index (DSI), disease
incidence (DI), resistance to oxalate (ROX), yield, seed size, days to
flowering, days to maturity, plant architecture and lodging for the 20
8unsi/Midland RlLs included in the two groups (selected2 and
control3), grown under white mold pressure, in Montcalm, MI, in 2001
and 2002. ............................................................................................ 81

Table 2. 4. Disease severity index (DSI), disease incidence (DI), yield (Y), seed
size, days to flowering, days to maturity, lodging, and architecture (A)
of MAS groups of 8unsi/Raven RlLs, grown under white mold pressure
at Montcalm, MI, in 2001 and 2002. .................................................... 85

Table 3.1. Number of inbred backcross lines developed in crosses between
unadapted Pl accessions with putative physiological resistance to S.
sclerotiorum and three commercial dry bean cultivars. ..................... 113

Table 3.2. Lodging, desirability coefficient, maturity, seed size and reaction to
straw test in BC2 l8L derived from the cross between Tacana and PI
318695, grown in Saginaw, MI, in 2002. ........................................... 115

Table 3.3. Lodging, desirability coefficient, maturity, seed size and reaction to

straw test in BC2 I8L derived from the cross between Tacana and PI
313850, grown in Saginaw, MI, in 2002. ........................................... 117

viii

Figure 1.1.

Figure 1.2.
Figure 1.3.
Figure 1.4.
Figure 1.5.

Figure 3.1.

Figure 3.2.

Figure 3.3.

Figure 3.4.

Figure 3.5.

LIST OF FIGURES

Linkage map with 118 markers from 98 F4 derived recombinant
inbred lines originated from a cross between 8unsi and Raven.
Linkage groups 82 to 88 correspond to the integrated map (Freyre
et al., 1998) and linkage groups LG-1 to LG-5 are unassigned

linkage groups detected in current study .................................. 31
QTL for DSI 01 detected on Linkage group 82 ............................... 42
QTL for DSI 01 detected on Linkage group 85 ............................... 44
QTL for DSI 01 detected on Linkage group 88 ............................... 45
QTL for DSI 02 detected on Linkage group 87-a. .......................... 47

Reaction of adapted and exotic bean genotypes to inoculation with
S. sclerotiorum using the straw test. Data represent mean of 10
replicates and the bars indicate standard deviation for each mean.
P<0.001 for bean genotypes. The straw test was rated in a scale
from 1 to 9 where 1: no symptoms and 9: total plant collapse. ..1 19

Reaction of 89 Tacana/Pl 318695 inbred backcross lines (IBL) to
inoculation with S. sclerotiorum using the straw test. P < 0.001 for
l8L. The straw test was rated in a scale from 1 to 9 where 1: no
symptoms and 9: total plant collapse ............................................... 119

Reaction of 76 Tacana/Pl 313850 inbred backcross lines (l8L) to
inoculation with S. sclerotiorum using the straw test. P < 0.001 for
l8L. The straw test was rated in a scale from 1 to 9 where 1: no
symptoms and 9: total plant collapse .................................. 120

Diameter of lesion in detached leaves of five common bean
accessions at 48 hours after inoculation with S. sclerotiorum. Data
represent mean of 9 replicates, and the bars indicate standard
deviation. P=0.003 for bean genotypes. ...................................... 122

Peroxidase activity in detached leaves of five common bean
accessions at 48 hours after inoculation with PDA plug (control),
and PDA plug with S. sclerotiorum mycelia. Data represent mean of
9 replicates, and the bars indicate standard deviation. P=0.02 for
genotypes when inoculated and non significant for genotypes
without inoculation (control). ........................................................ 123

ix

Figure 3.6.

Figure 3.7.

Figure 3.8.

Figure 3.9.

B-1,3-Glucanase activity in detached leaves of five common bean
accessions 48 hours after inoculation with PDA plug (control), and
PDA plug with S. sclerotiorum mycelia. Data represent mean of 9
replicates, and the bars indicate standard deviation. P=0.01 for
genotypes when inoculated and P=0.03 for genotypes without
inoculation (control). .................................................................... 124

Chitinase activity in leaves of five common bean accessions 48
hours after inoculation with PDA plug (control), and PDA plug with
S. sclerotiorum mycelia. Data represent mean of 9 replicates, and
the bars indicate standard deviation. Differences were not

significant among genotypes at P=0.05, with or without inoculation.
..................................................................................................... 126

Peroxidase activity in detached leaves of five common bean
accessions 48 hours after inoculation with PDA plug (control), and
PDA plug with S. sclerotiorum mycelia. ....................................... 127

Glucanase activity in detached leaves of five common bean
accessions 48 hours after inoculation with PDA plug (control), and
PDA plug with S. sclerotiorum mycelia. ....................................... 128

Figure 3.10. Chitinase activity in detached leaves of common bean accessions

48 hours after inoculation with PDA plug (control), and PDA plug
with S. sclerotiorum mycelia. ....................................................... 129

LIST OF APPENDICES

Table A1. Disease severity index (DSI), disease incidence (DI), yield, seed size,
days to flowering, days to maturity, lodging and architecture of
8unsi/Raven RlLs grown under white mold pressure at Montcalm, MI,
in 2001. ............................................................................................. 149

Table A2. Disease severity index (DSI), disease incidence (DI), yield, seed size,
days to flowering, days to maturity, lodging and architecture of
8unsi/Raven RlLs grown under white mold pressure at Montcalm, MI,
in 2002. ............................................................................................. 153

Table A3. Primer combination and fragment size of AFLP markers tested in
8unsi/Raven RIL population .................................................. 157

Table 8.1. Lodging, desirability coefficient, maturity, seed size and reaction to
straw test in 89 8C2 l8L derived from the cross between Tacana and
PI318695, grown in Saginaw, MI, in 2002 ................................... 161

Table 8.2. Lodging, desirability coefficient, maturity, seed size and reaction to

straw test in 76 802 I8L derived from the cross between Tacana and
Pl313850, grown in Saginaw, MI, in 2002 ................................... 164

xi

INTRODUCTION

Bean is one of the most important food crops in many countries of Latin
America and Africa, where it provides an important source of calories, proteins,
dietary fibers, minerals, and vitamins. Among the major food legumes the
common bean (Phaseolus vulgaris L.) is the third most important worldwide, only
surpassed by soybean and peanut. Considering only grain legume food crops
that are harvested for dry seeds, the common bean is the most important. Annual
per capita consumption of dry bean is as high as 60 kg in Rwanda (ISAR, 2001),
and almost 17 kg in Brazil and Mexico (FAO, 2002). Consumption is increasing in
the US. and Canada (4 kg/person/year), because of diversity in ethnic foods and
a greater awareness of the health benefits of bean in the diet.

The common bean is affected by both abiotic and biotic production
constraints. Abiotic limitations would include low soil fertility, high temperature
and water stress, whereas the most important biotic problems would include
weeds, insect pests, and diseases caused by bacteria, virus, and fungal
pathogens (Singh, 1999). One of the most destructive diseases of common bean
is white mold caused by the necrotrophic pathogenic fungus Sclerotinia
sclerotiorum. Strategies for controlling white mold in common bean include use of
resistant varieties, cultural practices, fungicides, and biological control. Sclerotia
produced by S. sclerotiorum over winter and remain viable in the soil for many
years (Hart and Saettler, 1981).

Field reaction to S. sclerotiorum is affected by physiological resistance,

agronomic and phenological traits that can act as avoidance mechanisms and

 

 

are highly influenced by environmental conditions. Avoidance mechanisms can
limit the establishment of the fungus and disease development. Avoidance is
associated with traits such as plant architecture erectness and porous plant
canopy or phenological traits such as early or late flowering. Most of the
morphological traits in common bean are easily scored exhibit high heritability
and can be effectively selected in the field (Kolkman and Kelly, 2002). Therefore,
selection for avoidance traits to white mold is feasible and genetic variability
exists in most of the bean commercial classes, especially in small seeded types
such as black beans. However in those regions with high humidity during the
growing season, conditions can be highly favorable for fungus development and
avoidance mechanisms may not be adequate to limit the infection and
development of white mold disease. Since white mold is a complex disease, the
combination of different approaches to improve resistance that include avoidance
mechanisms with physiological resistance seems to be most promising strategy
to improve resistance to S. sclerotiorum in common bean.

Physiological resistance is controlled by biochemical factors that limit the
infection or spread of the pathogen in plant tissue. Possible basis for
physiological resistance could include plant defense related enzymes,
phytoalexin accumulation, or plant proteins able to inhibit important fungal
enzymes. The development of resistant cultivars has had limited success
because few sources of physiological resistance to white mold have been
identified or used in breeding programs. The detection of physiological resistance

is difficult since the reaction to the pathogen under variable field conditions, is

affected by plant avoidance mechanisms. Therefore, selection for physiological
resistance requires effective screening methods. Most of the screening methods
proposed to detect physiological resistance to S. sclerotiorum in common bean
are based on measuring the ability of the plant to limit fungal growth in controlled
inoculation situations. These methods include the limited-term inoculation
(Hunter et al., 1981 ), the excised-stem inoculation technique (Miklas et al.,
1992a), growing callus on medium containing pathogen culture filtrate (Miklas et
al., 1992b), the straw test (Petzoldt and Dickson, 1996), and the leaf-agar plug
assay (Steadman et al., 1997). An indirect method to measure tolerance to oxalic
acid (Kolkman and Kelly, 2000) was based on the fact that oxalic acid is the
primary pathogenicity factor of S. sclerotiorum during the infection process
(Godoy et al., 1990). The limitation of using these screening methods is the lack
of consistency between methods, and weak associations with field resistance.
Taking in consideration that inheritance of resistance to white mold is
complex, highly influenced by the environment, and scoring demands intensive
field work, marker assisted selection (MAS) appears to be a promising approach
to improve resistance to white mold in bean. MAS might be more important for
physiological resistance, since desirable agronomic traits can be efficiently
selected through phenotypic selection. The identification of molecular markers
tightly linked to quantitative trait loci (QTL) associated with physiological
resistance could assist in the development of new varieties with higher levels of
resistance to white mold. The detection of QTL involves several important

aspects:

1. Identification of sources of resistance to white mold that possess major QTL
for resistance that may be useful in plant breeding;

2. Development of adequate mapping populations. Resistance to white mold has
exhibited low heritability, and populations such as recombinant inbred lines are
more valuable because they can be replicated in different years and locations,
resulting in more accurate phenotypic data;

3. Methodologies are available to select candidate markers such as selective
genotyping, bulked segregant analysis, and statistical tools to analyze the
molecular data and phenotypic data to predict the location and contribution of
QTL.

4. When QTL are detected, markers tightly linked to the QTL have to be
converted into easy screening robust techniques such as SCAR markers.

MAS goes beyond the identification of markers linked to QTL controlling
important quantitative traits. Effective strategies of selection are needed to
implement the use of MAS in breeding programs. The generation of adequate
population sizes required for selection is an important aspect that needs to be
considered. Use of MAS requires large segregating populations to provide

sufficient genetic variability for the selection of other important traits.

The identification of markers linked to different QTL for a complexly
inherited trait such as resistance to white mold might be a unique opportunity to

use MAS and combine QTL to enhance resistance to white mold.

 

 

A new approach to overcome the lack of physiological resistance to white
mold in cultivated bean could be the identification and introgression of novel
sources of resistance from primitive landrace varieties and related wild types of
bean. Singh (1999) suggests that as much as 90% of the genetic variability
available in the primary gene pool and related species remains underutilized.
Thousands of bean accessions, including landraces and wild types, are stored in
germplasm collections. Therefore, there is a reservoir of potential sources of
novel genes to be identified and introduced in bean breeding programs. To
evaluate and utilize efficiently these collections, it is necessary to identify a
smaller subset or core collection that likely represents most of the genetic
variation available in the entire collection.

Compared to their wild progenitors, cultivated plants display contrasting
phenotypic differences that combined are known as domestication syndrome
traits that are the consequence of thousands of years of selection for adaptation
to cultivated environments (Koinange et al., 1996). Landraces and wild
genotypes cannot be tested directly in the field for resistance to white mold due
problems of adaptation and especially for the wild bean germplasm, that lack the
domestication syndrome traits. Putative sources of physiological resistance to
white mold from exotic germplasm are normally detected by greenhouse tests.
To verify the reaction to S. sclerotiorum in the field it is necessary to introgress
the novel genetic variability from the exotic germplasm into adapted cultivars.

The inbred backcross method is recognized as an effective method to transfer

more complex quantitative traits from diverse and unadapted germplasm into
cultivated and productive varieties.

The introgression of physiological resistance to white mold into adapted
varieties is the main objective in using exotic germplasm. To understand the
different mechanisms involved in plant defense, the evaluation of plant defense
related proteins present in exotic germplasm is an important approach to be
conducted. To gain insight into physiological resistance and quantitative host-
pathogen relationships in general, the activity of plant defense related proteins

could be assayed in genotypes with different levels of resistance to S.
sclerotiorum. 0-1,3-Glucanases, chitinases and peroxidases are some of these

plant defense related proteins that could be involved in general plant defense

mechanisms (Van Loon, 1999).
The objectives of the research were:

1. to identify AFLP and RAPD markers linked to QTL associated with resistance
to S. sclerotiorum in Middle American mapping population of common bean;

2. to map the putative QTL on the been integrated map;

3. to compare the mapped QTL in this study with previously identified QTL for
resistance to white mold;

4. to identify QTL associated with important agronomic traits and their
relationship with resistance to white mold.

5. to test the potential of marker assisted selection using previously identified

markers linked to QTL for resistance to white mold;

 

6. to introgress genetic variability from landrace and wild genotypes into adapted
genetic background through the inbred backcross method;

7. to test the reaction to S. sclerotiorum and determine the endogenous and
induced levels of the defense related proteins in the selected parental

varieties and wild and landraces accessions.

The development of a linkage map and identification of QTL associated
with resistance to white mold was conducted in a F4 derived RlL population from

the cross between two small seeded bean cultivars differing in reaction to white

 

mold, 8unsi and Raven. The putative QTL detected for resistance to white mold
and the associations with other agronomic traits are discussed in chapter one.

The potential use of marker assisted selection is discussed in chapter two.
AFLP and RAPD markers previously identified to be linked to QTL associated to
resistance to white mold were used in a population of 96 RlLs derived from a
different cross between 8unsi and Midland.

Landraces and wild bean genotypes, identified as putative sources of
resistance physiological resistance against S. sclerotiorum, were used to
introgress novel genetic variability into cultivated bean varieties by developing
inbred backcross lines. The greenhouse evaluation of progeny derived from
these crosses for white mold resistance, and the determination of the level of

endogenous and induced defense related proteins in exotic and cultivated

 

parents are discussed in chapter three. A general discussion of strategies for

 

future improvement of white mold resistance in common bean is presented in the

finalsecfion.

References

FAO, 2002. Food and Agriculture Organization of the United Nations yearbook:
production. Rome.

Godoy, G., J.R Steadman, M.8. Dickman, and R. Dam. 1990. Use of mutants to
demonstrate the role of oxalic acid in pathogenicity of Sclerotinia
sclerotiorum on Phaseolus vulgaris. Physiological and Molecular Plant
Pathology 37:179-191.

Hart, PH, and AW. Saettler. 1981. White mold of beans. Cooperative Extension
Service, Michigan State University, Extension Bulletin E-892.

Hunter, J.E., M.H. Dickson, and A. Cigna. 1981. Limited-term inoculation: a
method to screen bean plants for partial resistance to white mold. Plant
Disease 65:414-417.

ISAR, 2001. Bean program 2001: activity review. p 1-8.

Koinange, E. M.K., S.P. Singh, and P. Gepts. 1996. Genetic control of the
domestication syndrome in common bean. Crop Sci. 36:1037-1045.

Kolkman, J.M., and JD. Kelly. 2000. An indirect test using oxalate to determinate
physiological resistance to white mold in common bean. Crop Sci. 40:281-
285.

Kolkman, J.M. and Kelly, J .D. 2002. Agronomic traits affecting resistance to white
mold in common bean. Crop Sci. 42-693-699.

Miklas, P.N., K.F. Grafton, and 8D. Nelson. 1992a. Screening for partial
physiological resistance to white mold in dry bean using excised stems. J.
Amer. Soc. Hort. Sci. 117:321-327.

Miklas, P.N., K.F. Grafton, G.A. Secor, and PE. McClean. 1992b. Use of
pathogen filtrate to differentiate physiological resistance of dry bean to
white mold disease. Crop Sci. 32:310-312.

Petzoldt, R., and M.H. Dickson. 1996. Straw test for resistance to white mold in
beans. Annu. Rep. Bean lmprov. Coop. 39:142-143.

 

 

Singh, PS. 1999. Common bean improvement in the twenty-first century. S.P.
Singh (ed). Kluver Academic Publishers, Dordrecht, The Netherlands.

Steadman, J. R., K. Powers, and 8. Higgins. 1997. Screening common bean for
white mold resistance using detached leaves. Annu. Rep. Bean lmprov.
Coop. 40:140-141.

Van Loon, LC, 1999. Occurrence and properties of plant pathogenesis-related
proteins. In: S. K. Datta and S. Muthukrishnan (eds). Pathogenesis-
related proteins in plants. CRS Press: New York. p 1-19.

Chapter I

IDENTIFICATION OF QTL ASSOCIATED TO WHITE MOLD RESISTANCE IN
COMMON BEAN

Introduction

White mold, caused by the necrotrophic fungus Sclerotinia sclerotiorum
(Lib.) de Bary, is a serious disease of common bean (Phaseolus vulgaris) world
wide, and causes yield loss and reduced seed quality (Steadman, 1979).
Strategies for control of white mold in common bean include cultural practices
(crop rotation, residue management, irrigation timing), fungicide applications,
biological control, and use of resistant cultivars (Steadman, 1979; Ferraz et al.,
1999; Gerlagh et al., 1999; Lamey , 1996; Huang et al., 2000). Different
microorganisms including fungi and bacteria have been tested as biological
control alternatives. One of the most promising fungal biological control agents,
Coniothyn'um minitans, is a mycoparasite of sclerotia of S. sclerotiorum, and its
potential as a biological control has been reported (Gerlagh et al., 1999; Whipps
2001). The use of resistant varieties is a preferred disease control strategy for
white mold but has been a difficult management strategy to implement as no
major genes for resistance have been identified in common bean (Steadman et
al., 2000). A limited number of resistant bean varieties are available in certain
seed type classes, whereas commercial classes such as pinto and great northern

are highly susceptible to white mold. Sources of resistance have to be identified

10

in small seeded cultivars and need to be introduced in different bean seed type
market classes.

White mold severity in common bean is affected by interaction of
environmental conditions such as temperature and soil moisture, plant avoidance
mechanisms and physiological resistance. The avoidance mechanisms are
associated with plant morphological traits such as upright plant architecture and
porous plant canopy which affect the microclimate conditions within the plant
canopy that can limit the fungal establishment and development. Upright plants
and porous canopy retain less moisture inside the canopy, and this provides less
favorable conditions for development of white mold (Schwartz et al., 1978). Soil
moisture is a main factor controlling germination of sclerotia from S. sclerotiorum
(Abawi and Grogan, 1975; Steadman, 1983). The number of apothecia under the
porous canopy of the bean genotype Aurora was much lower than under the
dense canopy of the genotype GN Tara. As a result there was a reduction in the
number of ascospores available to infect senescent blossoms in the plant
(Schwartz and Steadman, 1978). Ascospores germination and infection of
senescent flowers, stems and pods is favored by high humidity and moisture
associated with dense plant canopies, caused by high plant populations, narrow
row widths, high fertility growth responses, and excess irrigation or rainfall
especially during and after flowering (Schwartz et al., 1987; Lamey, 1996).
Phenological traits such as early flowering and early maturity can be important
plant avoidance mechanisms, since the flowering period may occur before full

canopy closure or before excessive number of ascospores are present. Primary

ll

 

infection of S. sclerotiorum is from air-borne ascospores released from apothecia
produced by sclerotia that over-winter in the soil. Ascospores dispersed into plant
canopy require a nutrient source, such as flower blossoms, for germination
(Abawi et al., 1975; Haas and Bolwyn, 1972). Infection of healthy pods, leaves
and stems generally results from an infected flower that has fallen and lodged on
leaf axis and come into contact with other plant tissues (Hart and Saettler, 1981).
These studies indicate that disease severity of S. sclerotiorum could be
considerably reduced by developing common bean genotypes with an upright
porous type of plant architecture (Coyne, 1980).

Stiff-strawed bean genotypes with strong standing ability may have a
better chance of escaping initial white mold infection. Bush and upright plant type
were tested in various row widths and bush types lodged more than upright types
under narrow rows. Lodging resistance of stiff-strawed upright genotypes helped
lessen white mold infection. However, other traits such as canopy porosity were
also effective in minimizing disease development. For example, 8unsi, an
indeterminate type II bean cultivar with an open porous canopy, lodged as much
as a short bush type, but had low levels of white mold infection (Park, 1993).

Many studies have attempted to associate differences in growth habit with
levels of white mold infection in common bean. Beans are classified in four
discrete growth habits known as types I, II, III and IV (Singh, 1982). Type I is
strictly determinate, commonly referred to as a bush bean, with few nodes, highly
branched and a concentrated flower set. Type II, III, and IV are indeterminate but

differ in the length of vine extension, stem strength, and branch number and

12

 

angle. Type II growth habit, commonly referred to as an upright short vine, has
lodging resistance and a limited branching pattern maintained at an acute angle
to the dominant main stem. Type III growth habit, commonly referred to as a vine
type, has a more prostrate growth habit that lacks adequate stem strength to
remain erect during pod fill. Type IV growth habit is a climbing type and is suited
to production systems of intercropping with corn (Kelly, 2001).

Determinate navy bean genotypes such as Newport and Midland, were
highly susceptible to white mold (Kolkman and Kelly, 2002) when grown under
high humidity conditions. The open, porous canopy of larger-seeded Andean
determinate beans; however, have been considered a resistant phenotype in
semi-arid locations (Miklas et al., 2001; Park et al., 2001). Determinate plant
growth habit could be a disease avoidance mechanism based on studies with
near isogenic lines differing in growth habit. Near-isogenic determinate genotype
GN Nebraska 1 exhibited lower white mold infection than the indeterminate
genotype GN Nebraska 1 (Steadman et al., 1973; Coyne et al., 1977), whereas
determinate G.N. P 92 was more susceptible than the indeterminate near-
isogenic G.N. P 82. Thus, the bush growth habit per se does not decrease the
incidence of infection (Schwartz et al., 1978; Coyne et al., 1977; Coyne et al.,
1978). Indeterminate plant growth habit commonly has been associated with
increased disease incidence and severity. Critical determinants of disease
severity were the distribution of the leaf area, especially in the lower canopy
levels near the soil (Schwartz et al., 1978). An important aspect that has to be

considered is the location where the tests were performed. Studies conducted in

13

the western US, under semi-arid conditions and irrigation, support the
importance of plant architecture as an avoidance mechanism. Such mechanisms
are less effective in high humidity regions such as Michigan, where avoidance
has to be combined with physiological resistance to obtain an effective level of
field resistance. Avoidance mechanisms may be defeated under favorable
conditions for the pathogen, such as high precipitation, humidity and
temperatures around 15 to 20°C. An ideal bean resistant ideotype for the
Midwest was proposed by Kolkman and Kelly (2002) and should be
indeterminate, with type II growth habit, resistant to lodging, medium canopy
width and branching pattern, and medium height, days to flowering and maturity,
to ensure adequate vegetative growth and yield potential without creating a
favorable microclimate within the plant canopy for white mold development.
Growth habit was not a reliable predictor of the ability to escape white
mold infection under field conditions (Schwartz et al., 1987). Instead, the plant
canopy structure and canopy density associated with the respective growth
habits determine whether microclimatic conditions created within the plant
canopy are more or less favorable for S. sclerotiorum mold colonization and
infection (Schwartz et al, 1978, 1987). Park (1993) obtained similar results and
concluded that narrow canopy, higher flowering nodes and fewer branches in the
upright plant type might have contributed to avoidance of white mold infection.
Plants with narrow canopies allow better light penetration and air movement that
results in rapid drying of dew on leaves and reduced moisture on the soil surface,

which affects sclerotia germination, reducing the number of ascospores available

14

that contribute to disease development (Coyne et. al., 1974; Blad et al. 1978;
Schwartz et al., 1987).

Since most of the studies of relationship between plant architecture and
white mold were conducted in a limited number of genotypes, it is possible that
some conclusions are specific to those genotypes. White mold is a complex
disease and is highly affected by environmental conditions, physiological
resistance, as well as avoidance mechanisms. However, one can assume that
white mold avoidance mechanisms are related to a number of morphological
traits that influence the microclimate conditions that can be more or less
favorable for the fungus establishment and development. Most of the
morphological traits in common bean are highly heritable and are easily scored in
the field. Estimates of heritability for days to flowering and days to maturity in a
RIL population derived from a cross between two navy beans, 8unsi and
Newport, were 0.91 and 0.75 respectively (Kolkman and Kelly, 2002), making
early generation selection efficient.

Physiological resistance to white mold in common bean might be
associated with endogenous factors that limit the spread of the pathogen in plant
tissue. Such factors include increased activities of plant defense-related enzymes
(Miklas et al., 1993), phytoalexin accumulation (Sutton and Deverall, 1984), and
oxalate tolerance (Tu, 1985). The development of resistant cultivars has had
limited success because few sources of physiological resistance to white mold
have been found and exploited in bean breeding programs (Miklas et al., 1999).

Screening for physiological resistance is highly confounded by morphological

15

avoidance mechanisms in the field; therefore, incorporating physiological
resistance into bean requires reliable screening techniques that detect this type
of resistance. Many methods based on the fungus growth were proposed to
detect physiological resistance to white mold in common bean, including the
limited-term inoculation method (Hunter et al., 1981), the excised-stem
inoculation technique (Miklas et al., 1992a), growing callus on medium containing
pathogen filtrate (Miklas et al., 1992b), the straw test (Petzoldt and Dickson,
1996), and the leaf-agar plug assay (Steadman et al., 1997). An indirect assay to
estimate physiological resistance, based on resistance to oxalic acid was
developed by Kolkman and Kelly (2000). Since oxalic acid is the primary
pathogenicity factor of S. sclerotiorum in the infection process (Godoy et al.,
1990), genotypes that resist damage due to oxalate may possess a form of
physiological resistance.

The tests used to detect physiological resistance to white mold in common
bean have produced variable results, are influenced by environmental conditions,
and their routine use in breeding programs has been limited. For example, Park '
et al., (2001) reported low heritability estimates for physiological resistance to
white mold, using the straw test that is the most used greenhouse screening
method, whereas Miklas et al. (1992; 2001) estimated high heritability for
resistance to white mold also using straw test. Heritability estimates for
physiological resistance using the oxalate assay were also low (0.19; Kolkman
and Kelly, 2000).

Low heritability estimates (0.05 to 0.30) of resistance to white mold under

16

field conditions were reported by Roberts et al., (1982), whereas Miklas et al.
(2001) reported high heritability (0.65) for resistance to white mold using a scale
from 1 to 9. However, moderate heritability (0.42) for disease severity was
obtained by Kolkman and Kelly (2002) under field conditions. Considering the
high heritability of avoidance mechanisms compared to low values for
physiological resistance, selection based solely on disease severity may improve
white mold resistance in the field via avoidance mechanisms, but may limit the
opportunity to improve the level of physiological resistance (Kolkman and Kelly,
2002). Undesirable avoidance mechanisms, such as early or late flowering and
reduced plant growth, may be not only associated with low levels of white mold,
but also may affect negatively the yield potential (Kolkman and Kelly, 2002).
Avoidance mechanisms associated with low yield are not useful in plant breeding
for resistance to white mold, because they limit the ability of breeders to select
for high yield that is always one of the most important objectives. Selection for
yield under consistent high white mold pressure could be one of the best
approaches to select resistant genotypes with high yield potential.

Combining physiological resistance with morphological avoidance
mechanisms is currently the most viable breeding strategy to assure adequate
levels of field tolerance to white mold (Miklas et al., 2001). Considering that
white mold is a complexly-inherited trait, with low to moderate heritability and
highly influenced by environmental factors that demands intensive field work,
marker assisted selection (MAS) seems to be a promising approach to improve

resistance to white mold in beans. Therefore, the identification of new and novel

17

 

 

genetic sources of resistance and detection of quantitative trait loci (QTL)
associated with physiological resistance in current and new sources could be a
very important breeding strategy.

The combination of disease resistance with agronomically desirable traits
has been accomplished by performing a multivariate bulking strategy for QTL
detection. The use of multiple traits that are related to the trait of interest was
effective in identifying QTL that may not be detected through screening extreme
phenotypes (Ronin et al., 1998). Detection of QTL associated with different
sources of white mold resistance could provide the opportunity to efficiently
combine physiological resistance with avoidance mechanisms. The major
potential benefit of multivariate bulking is to identify QTL conditioning resistance
in an agronomically acceptable plant type, by avoiding the detection of QTL
associated with undesirable avoidance mechanisms such as extreme early
flowering or maturity or short dwarf plants that restraint disease development but
limit yield potential (Kolkman and Kelly, 2003). Detection of QTL associated with
different sources of resistance to white mold could provide the opportunity to
combine unique sources of physiological resistance with avoidance mechanisms
in common bean, which would not be possible with direct phenotypic selection
under white mold pressure in the field.

Mapping of QTL for white mold resistance was undertaken by Park et al.
(2001) using both field experiments and the straw test to evaluate resistance to
S. sclerotiorum. The QTL analysis was conducted on a population of RILs

derived from the cross between PC-50 and XAN-159, genotypes differing in

18

 

 

reaction to white mold. PC-50 is resistant genotype to white mold and is derived
from a single plant selection in the landrace ‘Pompadour Checa’ from the
Dominican Republic (Park et al., 2001). QTL controlling physiological resistance,
measured through the straw test, were detected through composite interval
mapping on linkage groups 82, 83, 84, 87, 88 and 811 of the integrated bean
linkage map, and QTL for field resistance were detected on B4, 87 and 88. Six of
the seven candidate QTL for field resistance were found in the same locations as

QTL for physiological resistance. Miklas et al., (2001) detected through interval

 

mapping analysis, a QTL on linkage group 87 that explained 38% of the
phenotypic variation for disease using the straw test, in a RIL population derived
from a cross between G122 and A55. G122 is an upright determinate been with
physiological resistance to white mold. The same QTL detected on 87 were also
identified for field resistance along with an additional QTL on 81. The QTL on 81
was mapped near the fin gene that is responsible for determinate growth habit in
the Andean gene pool. The QTL detected on 87 derived from the Andean gene
pool genotypes PC-50 and G122, both mapped near the phaseolin seed protein
(Phs) gene, suggesting that an important QTL associated with resistance to white
mold must reside in this genomic region.

Molecular markers (RAPD and AFLP) linked to white mold resistance
were identified by Kolkman and Kelly (2003) using bulked segregant analysis, in
a RIL population obtained from a cross between the small seeded cultivars

8unsi and Newport from the Middle American gene pool. QTL for resistance

 

were detected on four linkage groups, 82, 83, B7 and 88. One major QTL for

19

disease incidence identified on linkage group 82, was located near RAPD
markers 80201800 and 015.1800. The region around 80201800 and
015.1800 on the integrated linkage map contains several identified genes that
may be important in disease defense responses (Kelly et al., 2003). The largest
linkage group associated with white mold resistance was 87. Markers in this
linkage group were significantly associated with disease severity index, days to
flowering, lodging, and yield and seed size. The QTL identified by Kolkman and
Kelly (2003), were found in regions of the genome associated with either plant
architecture or general plant defense response genes, such as PvPR-2 (Walter
et al., 1990), and Pgip (Toubart et al., 1992) on linkage group 82, the PvPR-1
gene on 83, and the seed lectins (Brambl and Gada, 1985) on linkage group 87.
The QTL on linkage B7, detected in a Middle American genotype by Kolkman
and Kelly (2003), did not map near the Phs, suggesting the QTL for resistance to
white mold was different from that detected in the Andean genotypes by Miklas et
al., (2001) and Park et al., (2001),

This study was conducted with the following objectives: 1. to identify AF LP
and RAPD markers linked to QTL associated with resistance to S. sclerotiorum in
a Middle American RIL mapping population of common bean; 2. to map all
putative QTL on the bean integrated map; 3. to compare the mapped QTL in this
study with previously identified QTL for resistance to white mold; 4. to identify
QTL associated with important agronomic traits and their relationship with

resistance to white mold.

20

Materials and Methods

Plant Material

This study was conducted in a RIL population of 98 F47 derived from the
cross between 8unsi and Raven. Both genotypes are classified as small-seeded
race Mesoamerica from the Middle American gene pool (Singh, 1999). 8unsi
(a.k.a Ex-Rico 23; Tu and Beversdorf, 1985) is an indeterminate navy bean, with
physiological resistance and a porous plant canopy (Schwartz et al., 1987; Miklas
et al., 19920; Kolkman and Kelly, 2002). Raven is an indeterminate black bean
that despite an upright architecture lacks adequate levels of physiological
resistance to S. sclerotiorum and is highly susceptible to white mold (Kelly et al.,
1994). Ninety-eight F2 plants were advanced from F2 to F4 through single seed '
descent method in the greenhouse. Seeds of individual F4 plants were bulked,
and advanced in a greenhouse during spring in 2000. Seeds harvested from
three F45 plants were bulked and F45 lines were increased in the field, in
Saginaw, MI, during summer 2000. No selection for agronomic or phenological
traits was made during the population development. The 98 F47 and F43 lines
and the parents, 8unsi and Raven, were evaluated for reaction to white mold in

field trials in Entrican, in Montcalm County, MI, in 2001 and 2002, respectively.

Field trials and Traits
The F4 derived lines were evaluated for resistance to white mold in
naturally infested field plots in Montcalm County, MI, during 2001 and 2002.

Supplemental irrigation was provided to enhance the development of white mold.

21

he soil type at the Montcalm Research Farm site is a combination of Eutric
Elossoboralfs (coarse-loamy, mixed) and Alfic Fragiorthods (coarse-loamy,
xed, frigid). Experimental design was a 10 x 10 lattice with four replications.
Ir row plots were 6 m in length, with 0.5 m row spacing. The two center rows
each plot were planted with the RIL, while the two border rows were planted
h the highly susceptible cultivar, Midland. Standard agronomic practices for
le, fertilization, and insect and weed control were applied to ensure adequate
nt growth and development. Plots were irrigated during the growing season
:ing in the third week after planting with 12 mm of water at least once a week,
)ending on rainfall, in order to promote uniform disease pressure across the
I. During 2001 plots were irrigated 8 times for a total of 127 mm. In 2002 the
lots were irrigated with approximately 180 mm distributed in 11 irrigations.
Plots were rated for disease prior to harvest, when the plants had reached
ihysiological maturity. Thirty plants per plot were rated from 0 to 4, where 0 =
isease present, 1 = 1 to 25 %; 2 = 26 to 50 %; 3 = 51 to 75 %; and 4 = 76%
IO % of the plant with white mold symptoms in order to estimate disease
rity index (DSI) and disease incidence (Dl) (Hall and Phillips, 1996). Disease
arity Index was calculated for each plot on a percentage basis, using the

iving formula:

DSI (°/o) = 2: (rating of each plant) 5 100

4 x (number of plants rated)

Disease incidence as a percentage was calculated as the number of

:3 out of the thirty individuals infected with white mold. The disease was also

22

 

 

rated in 2001, based on a scale from 1 to 9 as described by Miklas et al., (2001).
The RlLs were also evaluated for agronomic and phenological traits, including:
yield, seed size, days to flowering, days to maturity, canopy height, architecture
and lodging. Days to flowering were characterized by the number of days
following planting, when 50% of the plants in a plot have at least one open
flower. Four weeks after flowering the plots were evaluated for architecture, using
a 1 to 5 scale, where 1 = fully upright, 3 = bush, and 5 = prostrate. Lodging was
determined at the same time that architecture was rated, based on a 1 to 5 scale,
where 1 = no lodging, 3 = moderate lodging, and 5 = excessive lodging. Days to
maturity were calculated as the number of days following planting, until 90% of
the pods were physiologically mature and the drying down process was initiated.
Canopy height was measured at plant maturity on each individual plot, from six
measurements per plot. All plots were harvested at maturity, when plots were
individually pulled and threshed. Seed yield and weight of 100 seeds were
adjusted to 18 % moisture content.

In addition, the 100 RlLs and the parents were screened in the
greenhouse based on resistance to oxalate (ROX), an indirect assay to
determine physiological resistance to white mold (Kolkman and Kelly, 2000).
Twenty-one-day old seedlings (2nd trifoliate emerging) were cut at the base of the
stem and placed in a 20 mM oxalate solution (pH 4.0). The seedlings were rated
for wilting symptoms after 12 to 15 h of exposure to the oxalate solution. A 1 to 6
scale was used to measure wilting, where 1 = no wilting symptoms visible, 2 = 1

leaf with wilting symptoms (the two unifoliate leaves were rated together as one

23

 

 

leaf, and the 3 leaflets of a trifoliate leaf were rated together as one leaf), 3 = 2
leaves with wilting symptoms, 4 = 3 or more leaves with wilting symptoms, 5 =
petioles collapsing, 6 = main stem (total plant) collapsing. Wilting symptoms

ranged from curled leaf tip, to total loss of turgidity in the entire leaf.

AFLP and RAPD Analysis

Plant tissue for DNA extraction was collected from parental genotypes and
individual recombinant inbred lines from 10 F47 plants. DNA was extracted from
the plant tissue using a mini-prep procedure (Edwards et al, 1991; Haley et al.,
1994)

AFLP molecular markers were detected by performing the DNA digestion
with restriction enzymes EcoRI and Msel (V05, 1995). Bulked segregant analysis
(BSA) approach using AFLP markers with 256 primer pair combinations (EcoRI +
ANN and Msel + CNN) were first tested on the parents and resistant and
susceptible bulks to detect useful associated markers. Candidate markers were
first tested on the RlLs included in the bulks and subsequently on the
recombinant population. In addition to the candidate markers, all polymorphic
bands were scored and included in the construction of the linkage map. Chi-
square test for goodness-of—fit was used to test the deviations of the expected

segregation ratio of 9:7 (for F4 derived lines) or 1:1 (RILs).
AFLP procedure
Double digestion, adapter ligation, pre-amplification, and selective

amplification were carried out as previously proposed by Vos et al. (1995) with

24

 

 

the following modifications described by Hazen et al. ( 2002). Pre-amplification of
2 ul of restriction ligation product was combined with 25 ng of Msel and EcoRI
adapter, 0.5 mM dNTP, 1X PCR buffer (10mM Tris-HCL, pH 7.2 50mM KCI, and
0.1% Triton X-100), 0.5 U Taq polymerase, 1.5 mM MgCl2, total volume 20 pi
and amplified in a 96-well PTO-100 Programmable Thermal Controller (MJ
Research, Inc) programmed with the following profile [94°C 2 minutes — 26
cycles (94°C 1 min, 56°C 1 min, 72°C 1 min) - 72°C 5 min]. The preamplification
PCR product was added to 100 pl of sterile water and 1 pl of diluted product was
added to 19 ul of selective amplification cocktail (25 ng EcoRl primer, 30 ng Msel
primer, 0.4 mM dNTPs, 1X PCR buffer, 0.4 U Taq polymerase, 1.5 mM MgCl2)
and amplified via PCR [94°C 2 min - 12 cycles with annealing temperatures
decreasing by 0.7°C/cycle (94°C 30 sec, 65°C 30 sec, 72°C 1 min) — 23 cycles

(94°C 30 sec, 56°C 30 sec, 72°C 1 min) - 72°C 2 min].

Polyacrylamide gel electrophoresis
The selective amplification product was combined with 8 ul of formamide
loading buffer (98% formamide, 10mm EDTA, pH 8.0, 1.0 mg/ml bromophenol
blue; and 1.0 mg/ml xylene cyanol) and electrophoresed for 2hrs 30min at 80W
on a 6% polyacrylamide gel cast on a 38 x 50 cm Sequi-Gen GT sequencing cell
(BioRad Laboratories Inc, Hercules, CA, USA) to separate the AFLP fragments.
Fragments were visualized using a silver staining procedure (Promega #

Q4132). Gels were scored for polymorphism, based on presence or absence of

25

the AFLP, estimating band size in reference to a 100 and a 10 base pair DNA

ladden

RAPD procedure

The parents were also screened with 11 RAPD markers (Williams et al.,
1990) mapped by Kolkman and Kelly (2003) on linkage groups 82, B3, 87 and
88. In these linkage groups QTL associated with resistance to white mold were
identified. Polymerase Chain Reaction was conducted in a 96-well PTO-100
Programmable Thermal Controller (MJ Research, Inc) programmed for 3 cycles
of 1 min at 94°C, 1 min at 35°C, and 2 min at 72°C; 34 cycles of min at 94°C, 1
min at 40°C, and 2 min at 72°C with the final step extended by 1 s for each of the
34 cycles, and a final extension cycle of 5 min at 72°C (Haley et al., 1994).
Agarose gel electrophoresis (1.4%) was used to separate RAPD fragments. The
band size was estimated in reference to a 100 bp DNA ladder.

The AFLP and RAPD markers were coded in the following way. The
letters E and M of the AFLP marker names indicate the restriction enzymes Eco
R l and Mse 1, respectively. The three letters following E indicate the Eco RI
(ANN) selective nucleotides, while the second group of three letters indicate the
Mse l (CNN) selective nucleotides used in this study. The number following the
enzyme/primer combination represents the size of the polymorphic band. RAPD
markers were identified by the name of the Operon primer, followed by the size

of the polymorphic fragment.

26

Bulked Segregant Analysis

Bulked segregant analysis (BSA; Michelmore et al., 1991) based on
multiple trait DNA pooling procedure as proposed by Kolkman and Kelly (2003),
was conducted to create DNA bulks and identify candidate markers significant
markers for the 8unsi/Raven RIL population. Low levels of DSI and high yield
were combined in the resistant bulk to white mold and high levels of DSI and low
yield were combined in the susceptible bulk, within a range from 40 to 46 days to
flowering (Table 1.1). Days to flowering was chosen to minimize disease

avoidance based on early or late flowering (Kolkman and Kelly, 2002).

Statistical Analysis

The 124 molecular markers (118 AFLP + 6 RAPD) and two phenotypic
markers for disease resistance were analyzed and divided in linkage groups by
Joinmap software (Stam, 1993). The linkage groups were first anchored through
common markers previously mapped on the integrated map. All 28 AFLP primer
combinations used in this study were then tested for polymorphism against
8AT93 and Jalo EEP558, parents of mapping population used to create the
integrated map. Those polymorphic markers were then tested in the BAT93/Jalo
EEP558 (BJ) RIL population (Freyre et. al., 1998) and mapped to the integrated

bean map using MAPMAKER/EXP (Lander et al., 1987).

27

2881.62 Ba $62.82 a 82.8% 65 e .225 9.2.3 292-232 _.

 

 

 

 

 

we. 1:. u mam v.2. 9. .. 9. 8:258 a man 54 8 i ow @2826: e mac

8 3. :8 r 85 BB 8.8V 22> 33 484 88... A 229; 22> :2:

8.3. can i 28 NB :8 A .8 :2... 28 28 v 3°C .8 2,3

cams. $ch 58.2 mmcmm :95. cows. $ch :8...
c2238.”. 2.38 2238.03 2.35 2.22.0.2

 

.58 2 .8526: 2 may Em 22> .59 2.98

$822 co 8me ammo :oEEoo E 8:92me 208 BE; .8 ._._.O @5823 Lot. new: 302mb"... 95:5 <20 .2 033

28

Significant markers associated with resistance to white mold and other
traits were identified through single marker analysis (SMA) and composite
interval mapping (CIM) analyzed with software QTL Cartographer (Basten et al.,
2001) using a window size of 5 cM. Additive effects and coefficient of
determination (R2) were calculated through QTL cartographer. Threshold LOD
scores (90 and 95%) for individual traits were determined through a permutation
test, with 1000 permutations (Churchill and Doerge, 1994). The effect of multiple
QTL located in more than one linkage groups was analyzed using multiple
regression (SAS, 1995). The Pearson correlation coefficients (r) were calculated
by PROC CORR (SAS, 1995). The RIL population was analyzed across the two
years as a RCBD, using PROC GLM (SAS, 1995). Estimates of heritability for all
traits were calculated on entry mean basis, where h2 = 092/ 092/ ry + ogezl y +
092, (Fehr, 1987). Confidence interval estimates were calculated on entry mean

basis (Knapp et al., 1985).

29

Results and Discussion

Linkage Map

Twenty-eight AFLP primer combinations were identified using selective
genotyping and bulked segregant analysis. In total, 124 molecular markers (118
AFLP and 6 RAPD) and two phenotypic markers were evaluated in the
8unsi/Raven (BR) population of 98 F4 derived RILs. Linkage map construction,
using Joinmap (Stam, 1993), placed 118 markers on twelve linkage groups for a
total of 227.5 cM (Figure 1.1). Linkage groups were compared against the
integrated map (Freyre et al., 1998) and seven linkage groups were anchored to
the integrated map. 87-3 and B7-b were associated on different regions of
linkage group 87 of the integrated map. B4-a and 84-h were anchored to linkage
group 84 of the integrated map.

Mapping of F2 or F3 populations with dominant markers is not ideal, since
heterozygous individuals cannot be identified. Mapping RIL populations with
dominant markers is much more efficient, since heterozygous genotypes are in
very low frequencies in the population (Knapp et al., 1995). DNA fingerprinting
an F4-derived population with dominant markers based on a subset of individuals
from a more advanced generation, such as the F47 generation, was adequate to

detect QTL and generate a linkage map (Figure 1.1).
QTL associated with resistance to white mold

The two parental genotypes, 8unsi and Raven react differently to white

mold and exhibit variation in other agronomic traits in both environments.

30

 

 

 

 

 

 

 

82
0.0 AFLP49
12.5 012-1600
15.5 AFLP51
18,4 AFLP10
AFLP41
19.0 2 AFLP81
20.2 AFLP91
AFLP35
AFLP92
2"0 AFLP105
015-1800
AFLP58
AFLP36
AFLP63
2‘16 AFLP27
AFLP119
AFLP33
AFLP16
22.6 AFLP23c
8C20-1800
26.4 AFLP30
B8
0.0 AFLP102
1.4 AFLP29
6 5 JAFLP77c
' IAFLP120
10,8 \_/AFLP32
_. AFLP74
11° / AFLP76
14.3 AFLP57

 

 

 

 

 

 

84-a
00 AFLP25
an AFLP56
AFLP80
33 AFLP47
AFLP93
AFLP99
*5 AFLP101c
AFLP59
52 AFLP114
74 AFLP108
LG-‘I
00 AFLP95
63 AFLPIZ
113 AFLP21
173 AFLP06

 

 

 

00
L5

15
4B

68

00

28

 

B4-b

LG-2

AFLP20

 

 

AFLP19

 

 

AFLP07

 

AFLP34
AFLP94

AFLP73
AFLP04

AFLP61

 

 

 

 

 

 

 

 

 

BS
00 AFLP79
3.6 AFLP18
7 AFLP89
AFLP28
AFLP72
AFLP75
5'2 AFLPOS
AFLP107
AFLP31
6'9 AFLP96
AFLP26
7,5 AFLP44
AFLP43
AFLP112
8.6 AFLP54
‘ AFLP15
10.1 AFLP46
11.4 A AFLP71
16.9 —\ AFLP66
19.9/ \AFLP97
21.2 _ AFLP52
24.1 “ AFLP45
24,6 /‘ AFLP69
25.7 AFLP98
36.7 AFLP17
LG-3
0.0 AFLP106
10.0 AFLP50
13.5 /~\ AFLP62
jAFLP64
2‘2 lAFLP09
23,1 8C20-1200

87-a

 

 

 

 

 

 

 

 

 

 

0.0 \Q/ AFLP60
0.6 .2 AFLP109
1,1/ \AFLP113
7.0\_/ AFLP116
6 /~\ AFLP22
15.8 AFLP117
20.0 AFLP87
22.8 AFLP48
24.2 AFLP11
25.8 AFLP100
27.9 AFLP86
36.2 AFLP121c
LG-4
0,0 AFLP65
2 2 AFLP83
. AFLP84
AFLP90
2.7 AFLP110
AFLP39
3,3 AFLP78
7 5 AFLPSS
' AFLP1116
8.9 AFLP67

 

 

 

 

 

 

 

 

87-b
0.0 AFLP118
8.0 \ / H12-1050
_ AFLP68
/” AFLP40
8.6 AFLP103
AFLPOl
— AFLP03
12.8 / \ P9-1750
17.1 AFLP'IO
25.7 AFLP37
LG-s
0.0 AFLP53
2.9 AFLP14
6.4 AFLP24

 

 

 

 

Figure 1.1. Linkage map with 118 markers from 98 F4 derived recombinant
inbred lines originated from a cross between 8unsi and Raven.
Linkage groups 82 to 88 correspond to the integrated map (Freyre
et al., 1998) and linkage groups LG-1 to LG-5 are unassigned
linkage groups detected in current study.

 

.3]

 

Disease pressure was adequate in both years but the environmental
variation resulted in different levels of DSI and DI between the two years (Table
1.2). Data for resistance to white mold followed a normal distribution. The
agronomic traits, days to flower and yield in individual environments were not
always normally distributed. Since combined environments were normally
distributed, data was not transformed to correct skewness.

Markers associated with QTL for resistance to white mold and for
agronomic traits were detected through single marker analysis and composite
interval mapping. The QTL identified with CIM analyses in most of the cases
were supported by SMA.

As expected, markers were consistent between DSI and DI
measurements since both traits were highly correlated (r> 0.90; P<0.01).
Therefore, only the results for DSI will be discussed, but the considerations are
valid for both traits. Different QTL associated with resistance to white mold were
identified across the two environments. The different QTL detected can be
explained in part by differences in environmental conditions between years 2001
and 2002. In August and September 2001, the cumulative precipitation was 256
mm compared to 105 mm in 2002, whereas the normal precipitation for that
period would be 177 mm (Table 1.3). The average maximum temperature in
September 2002 was 26°C compared to 21°C in September of 2001. High
temperatures combined with reduced precipitation affected disease development

in 2002, despite the supplemental irrigation applied to the experiment.

32

Table 1.2. Means, range and heritability (h2) of disease severity index (DSI) and
agronomic traits in 98 F4 derived (RIL) from the cross 8unsi/Raven,
grown in Montcalm, MI, in 2001 and 2002.

 

 

 

 

Parental Recombinant inbred lines
Means
Trait Year 8unsi Raven Mean Range h2 (90% Cl)
DSI (%) 2001 29.9 37.2 44.9 23.1 — 72.6
2002 14.6 43.6 27.2 8.4 — 51.2

Combined 22.3 40.4 36.1 19.7 — 52.6 0.41 (0.18-0.58)
Yield 2001 4046 4719 4136 2130 — 5571
(kg/ha) 2002 3968 4013 4495 3127— 5627

Combined 4013 4371 4315 3015 — 5380 0.56(0.38-0.68)
Days to 2001 40.6 45.9 43.8 36.5 — 48.4
flower 2002 45.5 47.0 45.3 42.0 - 52.5

Combined 43.1 46.5 44.6 39.5 — 50.5 0.79(0.71-0.85)
Days to 2001 104.9 103.9 104.9 103.2 - 111.7
Maturity 2002 109.5 107.1 105.8 91.8 — 116.5

Combined 107.2 105.5 105.4 96.3 — 113.1 0.82(0.75-087)
Lodging 2001 3.6 2.4 3.3 2.0 - 4.9

2002 3.5 2.9 3.1 1.6 - 4.1

Combined 3.6 2.7 3.2 1.8 - 4.5 0.76(0.67-0.83)
Seed size 2001 22.7 21.9 22.2 18.9 — 26.1
(g.100seed“) 2002 22.8 19.0 21.0 16.7 — 26.2

Combined 22.8 20.5 21.6 18.3 - 25.7 0.90(0.85-0.93)

 

33

Table 1.3. Monthly precipitation from June to September in 2001 and 2002 in

 

 

 

Montcalm, Ml.
Precipitation (mm)
June July August September
Observed 2001 71.9 26.9 154.9 101.6
Deviation 2001 -15.2 -36.6 57.4 22.4
Observed 2002 58.9 71.1 81.0 23.9
Deviation 2002 -28.2 7.6 -16.5 -55.4
Normal1 87.1 63.5 97.5 79.2

 

‘ Calculated over 30 year period (1951-1980).

34

Five linkage groups with QTL for resistance to white mold were detected
in 2001 through SMA and three of these genomic regions were confirmed by CIM
(Table 1.4 and Table 1.7, 1.8 and 1.9). The first linkage group where a QTL for
DSI was identified was 82, located near marker 012.1600. The QTL accounted
for 9.9% (R2) of the phenotypic variability for DSI. Another QTL detected on
linkage 82, was located close to the RAPD marker, 015.1800 and explained
9.2% of the DSI (Table 1.7 and Figure 1.2). The QTL for DSI mapped close to
marker 015.1800 previously detected by Kolkman and Kelly (2003) in a different
RIL population, supporting the existence of a QTL for resistance to white mold in
that region on 82. The three AFLP markers AFLP35, AFLP92, AFLP105 mapped
to the same position as the marker 015.1800 on 82 (Figure 1.1; Appendix A —
Table A3).

The region around the 015.1800 marker on the integrated linkage map
possesses several genes identified in general host defense mechanisms that
may play a role in disease resistance. Plant defense-related genes may be
triggered as a general resistance response to S. sclerotiorum infection. PvPR-2,
a low molecular weight acidic protein induced during fungal elicitation (Walter et
al., 1990), and Pgip, a polygalacturonase-inhibiting protein (Toubart et al., 1992),
are all located in this region on 82. Differences in PvPR gene arrangements were
detected between anthracnose (Colletotrichum. Iindemuthianum) resistant and
susceptible bean genotypes indicating that polymorphism between PvPR as well
as other defense response-related genes may contribute to quantitative

resistance (Walter et al., 1990).

35

Table 1.4. Markers linked to QTL for resistance to white detected through single
marker analysis in a 98 derived RIL population from a 8unsi/Raven
cross, grown in Montcalm, MI, 2001.

 

Linkage Group Marker L0D1 P-value
B5 AFLP66 1 .09 0.027
AFLP97 0.95 0.039
AFLP52 1.37 0.013
AFLP98 1.42 0.012
LG-2 AFLP20 1 .66 0.006
AFLP19 1.47 0.010
32 AFLP10 0.92 0.042
AFLP91 1.27 0.017
AFLP35 1.58 0.008
AFLP92 1 .58 0.008
AFLP105 1.76 0.005
015.1800 1.78 0.005
AFLP58 1.44 0.011
AFLP36 1.42 0.011
AFLP63 1.34 0.014
AFLP27 1.34 0.014
AFLP119 1.34 0.014
AFLP33 1.19 0.021
AFLP16 1.06 0.029
AFLP23C 1 .02 0.032
BC20.1800 0.99 0.035
B7-a AFLP121C 1.15 0.023
38 AFLP102 0.94 0.040
AFLP29 1.11 0.026
AFLP77C 1.28 0.016
AFLP120 1.41 0.012

 

l Log-odds ratio

36

Table 1.5. Markers linked to QTL for resistance to white detected through single

marker analysis in a 98 derived RIL population from a 8unsi/Raven

cross, grown in Montcalm, MI, 2002

 

Linkage Group Marker LOD1 P-value
B7-b AFLP118 2.04 0.002
H12.1050 2.92 <0.001

AFLP68 3.54 <0.001

AF LP40 3.58 <0.001

AFLP103 3.58 <0.001

AFLP01 3.58 <0.001

AFLP03 3.43 <0.001
P9.1750 2.09 0.002
AFLP70 2.61 0.001

85 AFLP98 0.94 0.040
LG-2 AF LP07 1.21 0.020
32 AFLP49 1.03 0.032

 

1 Log-odds ratio

37

Table 1. 6. Markers linked to QTL for resistance to white detected through single
marker analysis in a 98 derived RIL population from a 8unsi/Raven
cross, grown in Montcalm, MI, 2001 and 2002

 

Linkage Group Marker L0D1 P-value
37-b AFLP118 1.07 0.028
H12.1050 1.50 0.009
AFLP68 1.77 0.005
AF LP40 1 .67 0.006
AFLP103 1.67 0.006
AFLP01 1.67 0.006
AFLP03 1.68 0.006
P9.1750 0.83 0.054
AFLP70 132 0.015
35 AFLP66 1 .07 0.028
AFLP97 0.90 0.045
AFLP52 1.38 0.013
AFLP98 1.75 0.005
LG-2 AFLP20 1 .14 0.024
AFLP19 1.37 0.013
AFLP07 1.26 0.017
32 AF LP35 0.99 0.035
AF LP92 0.99 0.035
AFLP105 1.13 0.024
015.1800 1.26 0.018
38 AFLP102 1.04 0.031
AFLP29 0.92 0.042
AFLP32 0.87 0.048
AFLP57 0.92 0.043
LG-5 AFLP24 0.95 0.039

 

1 Log-odds ratio

38

Table 1.7. Markers identified on linkage group 82, most closely linked to QTL for
resistance to white mold and agronomic traits, including phenotypic
variability associated with the QTL and additive effect of the presence
of the marker.

Linkage Group 82

 

Trait Marker Position of Environment R2 Additive
QTL (cM) (%) Effect1
DSI (%) 012.1600 8.0 Combined 10.1 3.15 (8)2
2001 9.9 3.83
2002 ns
015.1800 21.0 Combined 8.7 -2.66(B)
2001 9.2 -3.85
2002 ns
Yield (kg/ha) AFLP49 10.1 Combined 9.7 -171(8)
2001 ns
2002 ns
Days to flower AFLP63 21.6 Combined 7.6 0.57(R)
2001 6.9 0.65
2002 ns
Days to maturity AFLP16 22.6 Combined 28.5 2.06(R)
2001 28.3 1.58
2002 25.3 2.61
Lodging AFLP49 0.01 Combined 8.5 017(8)
2001 7.1 0.19
2002 ns
Seed size 015.1800 21.0 Combined 20.1 0.78(B)
(g.100seed") 2001 10.3 0.51
2002 18.8 0.83

 

1 Estimated additive effect of substituting one allele of Raven by one allele of 8unsi.
2 The letter in parenthesis indicates the parent where the marker is present, B=8unsi;

R=Raven.

39

Table 1.8. Markers identified on linkage group 85, most closely linked to the QTL
for resistance to white mold and agronomic traits, including phenotypic
variability associated with the QTL and additive effect of the presence

of the marker.

 

Linkage Group 85

 

 

Trait Marker Position of Environment R2 Additive
QTL (cM) (%) Effect1
DSI (%) AFLP98 Combined ns
27.7 2001 10.7 316(8) 2
2002 ns
Yield (kg/ha) AFLP98 25.7 Combined 7.5 -122(B)
Seed (g.1003eed") AFLP96 6.9 Combined 9.3 -0.49(R)
2001 8.3 -0.46
2002 ns

 

1 Estimated additive effect of substituting one allele of Raven by one allele of 8unsi.
2 The letter in parenthesis indicates the parent where the marker is present, B=8unsi;

R=Raven.

40

Table 1.9 . Markers identified on linkage group 88, most closely linked to the
QTL for resistance to white mold and agronomic traits, including
phenotypic variability associated with the QTL and additive effect of the

presence of the marker.
Linkage Group 88

 

 

Trait Marker Position of Environment R2 Additive
QTL (cM) (%) Effect‘
DSI (%) AFLP29 Combined ns
1.4 2001 9.2 293(8) 2
2002 ns
Yield (kg/ha) AFLP29 1.4 Combined 12.7 -158(8)
2001 13.8 -228
2002 ns
Lodging AFLP102 0.1 Combined 6.4 015(8)
2001 8.6 0.21
2002 11.0 0.19

 

1 Estimated additive effect of substituting one allele of Raven by one allele of 8unsi.
2 The letter in parenthesis indicates the parent where the marker is present, B=Bunsi;

R=Raven.

41

0,0

12,5

15,5
18,4
19,0
20,2

21,0\

21,6/

22,6

26,4

 

 

 

ll
\

\

IJJI

 

 

 

AFLP49

O12-1600

AFLP51

AFLP10
AFLP41 AFLP81

AFLP35 AFLP92
AFLP105 O15-1800
AFLP58 AFLP36
AFLP63 AFLP27
AFLP119 AFLP33
AFLP16 AFLPZBC
BC20-1800

AFLP30

F8
F17
I-S
-9
FA
F8
-6
0L

 

 

 

"'"I ThresholdP=0.1
--I-- ThresholdP=0.05
+ 08/01

 

LOISO

 

 

o....--.-o-oao¢-.u-.- ...-ouncov...---..-....o-a...no.5... oo....-o..........o--..-u-o
pa—.-o--.—o-.-u—u—c—u-.--—--o--.-u-o-o--o-n—.—-—o-.—--.-o-u—-—.-.-n-o—-—.--—o-oq

 

 

 

Figure 1.2. QTL for DSI 01 detected on Linkage group 82.

 

Pgip could also be a very important defense mechanism since
polygalacturonases are produced by S. sclerotiorum during the infection process.
Chalcone synthase, (ChS; Ryder et al., 1987), located near 015.1800, is an
enzyme required for isoflavonoid phytoalexins biosynthesis, and may be
important in general resistance to S. sclerotiorum. QTL associated with
resistance to white mold (Kolkman and Kelly, 2003), common bacterial blight
(C88) (Nodari et al., 1993), and root rot (Schneider et al., 2001) were identified in
this region of linkage group 82.

The second linkage group with a QTL associated with resistance to white
mold was mapped to linkage group 85. AFLP98 was the closest marker linked to
this QTL that contributed up to 10.7% of phenotypic variation (Table 1.8 and
Figure 1.3).

The third linkage group where QTL for resistance to white mold was
detected in 2001 was 88. AFLP 29 was the closest marker associated with this
QTL that accounted for 9.2% of the phenotypic variability for disease severity

(Table 1.9 and Figure 1.4).

43

B5

11,4

IIIIIII

AFLP79
AFLP18
AFLP89 AFLP28
AFLP72 AFLP75
AFLP05 AFLP107
/IAFLP31 AFLP96
/ AFLP26 AFLP44
AFLP43
\ AFLP112 AFLP54
\ AFLP15
\AFLP46
AFLP71

 

AFLP66

 

16,9

AFLP97

 

19,9

AFLP52

 

21,2

24,1 \
24,6 ’
25,7 /

 

/ AFLP45
\ AFLP69
\ AFLP98

 

AFLP17

 

36,7

LOISO

 

 

 

.-._.-.-.-.-.-._._._._.-._._._.-._.-.-.-.-.-.-.-._._.-.-.-.-.-._.-.-.-._.-.-.-.-.-.4

 

 

 

 

m-l- ThresholdP=0.1
--I-- ThresholdP=0.05
+ DSI01

 

Figure 1.3. QTL for DSI 01 detected on Linkage group 85

44

 

38

 

0,0 AFLP102

 

1,4 AFLP29

 

 

 

--l-- ThresholdP=0.05

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6,5 {AFLP77c AFLP120 D 3
<2 1 + DSI01
3 E
10.8 AFLP32 g
12.0 IAFLP74 AFLP76 E
4- 2
14.3 AFLP57 2

Figure 1.4. QTL for DSI 01 detected on Linkage group 88

45

In previous studies, a QTL for resistance to white mold was also found on
88, but mapped to a different region on the linkage group (Kolkman and Kelly,
2003). Another QTL on 88 that accounted for 26% of phenotypic variability for
disease reaction in the field was detected by Miklas and Delorme (2003) in snap
bean. This QTL detected on B8 was mapped between markers AH05.1000 and
D1468 on the integrated map and probably is different from that detected in
8unsi/Raven RIL population, since the QTL were mapped more than 30 cM
apan.
The genomic region that was associated with DSI in 2002 was mapped on
linkage group B7-b, anchored to 87 on the integrated map (Freyre et al., 1998)
by marker H12.1050. AFLP03 was the closest marker to this QTL that
accounted for 14.7% of the phenotypic variability of DSI (Table 1.5, Table 1.10
and Figure 1.5). In this region of B7, Kolkman and Kelly (2003) identified a
major QTL for resistance to white mold along with QTL for other agronomic
traits, including days to flowering and maturity, architecture, lodging, seed size
and yield. One of the most important markers detected by Kolkman and Kelly
(2003) on B7 was EaacMcn130 that corresponded to marker AFLP01 in this
study. This QTL accounted for 15.8% of the phenotypic variability for DSI. Seed
lectins (phytohaemoglutanins) genes also mapped to this region of 87 and may
have a putative role in plant defense (Shewry and Lucas, 1997). Miklas et al.,
(2001) and Park et al., (2001) also detected QTL for resistance to white mold of

87, but those QTL correspond to 87-a in this study.

46

B7-a

 

 

0,0 AFLP118

8.0 \ / H12-1050

I l

780/80

AFLP103 AFLP01

 

\{AFLPGB AFLP40

 

 

 

 

 

 

AFLP03
--I-- ThresholdP=0.05
12,8 P9-1750 + 03’0”-
17.1 AFLP70
25,7 AFLP37

 

 

 

 

 

 

Figure 1.5. QTL for DSI 02 detected on Linkage group B7-a.

47

Table 1.10. Markers identified on linkage group 87, most closely linked to the

QTL for resistance to white mold and agronomic traits, including

phenotypic variability associated with the QTL and additive effect of the

presence of the marker.

 

 

 

Trait Marker Position of Environment R2 Additive
QTL (cM) (%) Effect‘
Linkage Group B7—b
DSI (%) AFLP03 8.6 Combined ns
2001 ns
2002 14.7 -4.17(B)2
P9.1750 14.8 Combined ns
2001 ns
2002 14.2 -4.01(8)
Days to flower AF LP03 8.6 Combined 35.7 -1.21(B)
2001 24.5 -1.21
2002 38.3 -1.24
Days to Maturity AFLP70 17.1 Combined 9.8 -1.21(R)
2001 8.2 -0.84
2002 8.5 -1.51
Lodging AFLP03 8.6 Combined ns
2001 ns
2002 9.6 -0.17(B)
Seed size AFLP03 8.6 Combined 9.4 0.49(8)
(g.100seed'1) 2001 7.6 0.44
2002 11.4 0.59
Linkage Group B7-a
Yield (kg/ha) AFLP121c3 33.9 Combined 21.5 -158
2001 ns
2002 11.3 -148
Days to flower AFLP113 1.1 Combined 8.4 -0.60(B)
AFLP22 13.6 2001 14.8 -0.96(R)
Lodging AFLP117 15.8 Combined 29.8 033(8)
2001 30.2 0.39
2002 19.9 0.26
Seed size AFLP22 9.6 Combined 25.7 -0.83(R)
(g.100seed") 2001 27.3 -0.87
2002 19.6 -0.81

 

1 Estimated additive effect of substituting one allele of Raven by one allele of Bunsi.
2 The letter in parenthesis indicates the parent where the marker is present, B=Bunsi;

R=Raven.

3 Codominant marker.

48

The major QTL identified in this study would appear to be different from the QTL
detected on B7 in Andean genotypes G122 and PC-50 (Miklas et al.. 2001; Park
et al., 2001). QTL for DSI were detected on 87-a and LG2 using SMA, but the
QTL were not confirmed by CIM. The lack of consistency between methods does
not mean necessarily that these QTL are false positives but were more easily
detected by SMA than by CIM. The power of QTL detection is different between
methods and is influenced by population size and also by environmental
conditions, resulting in the lack of expression and detection of a QTL at a
significant level of probability.

The QTL for resistance to white mold identified in the 8unsi/Raven
population on 82 and 87 supported the results of previous studies (Kolkman and
Kelly, 2003). The fact that different QTL were identified in different years suggest
that depending on the environmental conditions, different genes may play a more
important role for resistance to white mold, including genes associated with
general defense or those associated with avoidance traits.

Resistance to oxalate was tested in the 98 RlLs and three QTL were detected
through SMA at 5% level of probability, but were not confirmed by CIM. The QTL
were detected near markers AFLP69 on 85, AFLP100 on B7-a, and AFLP42 that
was not linked to any of the12 linkage groups identified in this study. On 85
AFLP 69 was placed near AFLP 98 where a QTL for DSI was mapped. Additional
support for QTL for DSI on 85 comes from the detection of QTL for ROX in the

same genomic region.

49

Resistance to white mold and agronomic traits

Understanding the associations between DSI and phenological/agronomic
traits should contribute to the development of more efficient strategies to select
for resistance to white mold.

DSI was significantly correlated with yield in both years, however the
correlation coefficient was lower in 2002 (r = - 0.45") compared to 2001 (r = -
0.77") (Table 1.11). The lower correlation coefficient could be a consequence of
slower disease development in 2002. White mold did not develop to the same
level observed in 2001 due the high temperatures and low precipitation that
occurred in September 2002 (Table 1.3). Lodging was correlated with DSI in
both years. In 2001 high lodging probably was a consequence of the disease,
since the disease developed early after flowering and the highly susceptible lines
lodged due to disease. In contrast in 2002 more prostrate plants contributed to
high DSI by promoting a more favorable microclimate for the disease
development under the decumbent plant canopy. Architecture was not
associated with DSI in 2001, whereas more upright lines had less white mold
(r=0.42**) in 2002 (Table 1.11). Plant architecture likely played an important role
in avoiding disease as a result of the higher temperatures and lower precipitation
in 2002. The environmental conditions were more favorable for white mold
development in 2001 and plant structure was not detected as an important factor
in disease development. Under more favorable conditions for S. sclerotiorum,

avoidance mechanisms may not be sufficient to restrict disease development

50

Table 1.11. Phenotypic correlations between Disease Severity index and

phenological and agronomic traits in a 98 F4 derived RIL population

from a 8unsi/Raven cross grown in Montcalm in 2001 and 2002.

 

 

DSI 011 DSI 02 DSI Combined
DSI 01 1 0.33“ 0.79“
DSI 02 0.33“ 1 0.84"
Yield 01 -0.77** -0.23* -0.60**
Yield 02 -0.31** -0.45** -0.47**
Combined Yield -0.69** -0.38** -0.64**
Days to flower 01 ns 0.42“ 0.29"
Days to flower 02 ns 0.43“ 0.30“
Combined flower ns 0.46** 0.33"
Days to maturity01 ns 0.27** ns
Days to maturity02 ns 0.26“ ns
Combined maturity ns 0.28" ns
Lodging 01 0.51“ 0.23* 0.44“
Lodging 02 0.27“ 0.65“ 0.58"
Combined Lodging 0.45" 0.46" 0.56"
Architecture 01 ns 0.29“ 0.27"
Architecture 02 ns 0.42" 0.36"
Combined
Architecture ns 0.38“ 0.35“
100 seed weight 01 -0.21* ns ns
100 seed weight 02 ns -0.26** -0.24*
Combined seed ns ns -0.22*
Height 02 -0.22* -0.31** -0.33**
Resistance to
oxalate ns ns ns

 

1 01=2001; 02=2002.

51

and physiological resistance would play an increasingly important role in field
resistance. The identification of QTL on 82 instead of on 87 suggests that
physiological resistance was more important in 2001 compared to 2002. More
upright RlLs that were very susceptible to white mold in 2001 developed lower
levels of disease in 2002.

The linkage groups 82 and 87 were the most important for white mold
resistance in 2001 and 2002, respectively. QTL associated with phenological and
agronomic traits were detected on 82 and 87. Significant QTL for days to
flowering (AFLP 63; R2 = 7.6%), days to maturity (AFLP16; R2 = 28.5%), and
lodging (AFLP 49; R2 = 8.5%) across environments were detected on 82 (Table
1.7). The association of the marker AFLP03 with a QTL for DSI in 2002 in the
same region of B7-b where QTL were detected for days to flowering (R2 = 35.7),
and seed size (R2 = 9.4) across the environments and for lodging (R2 = 9.6) in
2002. Another QTL close to marker AFLP70 (R2 = 9.8), associated with days to
maturity was detected in the same linkage group. On 87-a, QTL associated with
yield, days to flower, lodging and seed size were identified. A QTL explaining
21.5% of phenotypic variation for yield across the two years was detected near
marker AFLP121c on 87-a. Marker AFLP116 from 87-a mapped to the Phs gene
region on 87 in BJ population and might explain the effect on traits such as seed
size and yield.

The QTL detected in 8unsi/Raven population for DSI accounted
individually for up to 15% of the phenotypic variability. In other crops similar

results were reported for QTL for resistance to white mold and associations with

52

 

SCI
PIE

all

I’ll

II.

agronomic traits. In soybean, three QTL, explaining 8, 9 and 10% of the
variability for DSI were detected in a 152 F3-derived line population in four
environments (Kim and Diers, 2000). Two of these QTL were also significantly
associated with disease avoidance mechanisms such as plant height, lodging,
and date of flowering. The other QTL was not related to agronomic traits and
may be involved in physiological resistance to white mold. Another study with
soybean was conducted in five RIL populations using the laboratory detached
leaf method to measure lesion area on leaves inoculated with mycelium plugs
(Arahana et al., 2001). Seven, out of 28 putative QTL for resistance to S.
sclerotiorum, were identified in multiple populations. Some of the QTL regions
mapped near resistance genes and resistance gene analogs. The level of
phenotypic variation explained by individual QTL ranged from 4 to 10%.

QTL for branching pattern, seed weight, oil content and flowering date
were also found to be associated with resistance to white mold in sunflower
(Mestries et al., 1998). Colocalization of QTL affecting resistance to both S.
sclerotiorum and Diaphon‘e helianthi in sunflower was found by Bert et al.,
(2002), suggesting a common component in the mechanism of resistance to
these two pathogens.

The 8unsi/Raven RIL population was known to be segregating for two
disease resistance genes, 00-1 and bc—3. 00-1 confers resistance to race 73 of
C. Iindemuthianum and bc-3 conditions resistance to bean common mosaic
necrotic virus (BCMNV) strain NL-3 in common bean. The 98 RlLs were tested

for the reaction to race 73 of C. Iindemuthianum and BCMNV strain NL-3 and

53

both traits were considered phenotypic markers in 8unsi/Raven RIL population.
Each phenotypic marker was linked with one molecular marker that was not
included in the twelve linkage groups. The Co-1 gene was linked to AFLP 42 at
7.20M and gene bc-3 was mapped 16.0 cM from marker AFLP85. These
phenotypic markers were not studied further as 00-1 is located on B1 and bc-3 is
located on 86 (Kelly et al., 2003), genomic regions where QTL for white mold

were not detected in this study.

Factors affecting Mapping and QTL detection

The use of bulked segregant analysis (BSA) to detect candidate markers
using only field data from 2001 may have resulted in the clustering of markers on
three linkage groups, 82, 84 and 85. BSA resulted in detection of many
polymorphic bands that mapped in regions associated with resistance to white
mold. This hypothesis is supported by the fact that QTL for DSI in 2001 were
detected on two linkage groups where markers clustered, 82 and 85.

A second reason for the clustering could be the use of restriction enzymes
EcoRI and Msel to digest the DNA in the AFLP analysis. EcoRl is non-sensitive
to cytosine methylation compared to the alternative restriction enzyme Pstl that is
notably sensitive to methylation. In soybean, AFLP markers generated using
EcoRI/Msel deviated from a random distribution when compared to markers
generated by Pstl/Msel (Young et al., 1999). The presence of one EcoRI/Msel
cluster per linkage group and the infrequent presence of markers derived from

Pstl suggested that those clusters could be mapped to heterochromatic regions

54

close to centromers. However, in the rice genome the clustering of markers in
only a few chromosomes could in part be explained due the small size of the rice
genome (Harushima et al., 1998).

In the 8unsi/Raven population the detection of two linkage groups 84-a
and B4-b that are mapped to B4 on the integrated map did not support the
hypothesis of one cluster per chromosome in the centromere region. Further
studies could be conducted in 8unsi/Raven population by including AFLP
markers using Pstl/Msel restriction enzymes.

The detection of QTL for complexly-inherited traits is highly dependent on
the environment where the trait is measured. Environmental conditions or even
microclimate variations may affect the expression of QTL for resistance to white
mold in common bean. The detection and location of the QTL is also affected by
the population size used in the study (Miklas et al., 2001).

The population used in this study was developed from a cross between
8unsi and Raven that are adapted to Michigan environmental conditions. A
population derived from parents with a narrow genetic background is considered
more useful for detection of QTL with minor effects. If a population from a wide
cross is tested, QTL are more likely identified, but may be related with broad
differences in gene pools, instead of minor differences associated with an
agronomically important complexly-inherited trait. Based on the narrow genetic
background of the 8unsi/Raven RIL population used in this study, a broad
coverage of the genome was not expected. Genetic maps to characterize a crop

species are constructed between more diverse parents, for instance, from

55

different gene pools in common bean. The map generated in 8unsi/Raven
population, covered 227.5 cM of the genome compared to 1226 cM of the
integrated map from BJ population, indicating that 8unsi and Raven have a
similar genetic background, despite differences for resistance to white mold. Only
genetic differences can result in polymorphic bands that are used to establish a
linkage map. In the 8unsi/Raven population, the similar genetic background
increases the probability of detecting those genomic regions associated with the
trait of interest. Both parents display an indeterminate growth habit, which should
be considered a useful aspect to detect minor QTL, because growth habit is
eliminated as a factor that might interfere with resistance to white mold.
Differences in growth habit have been shown to be associated with different
levels of white mold in common bean. Determinate navy bean genotypes such as
Newport and Midland, are very susceptible to white mold (Kolkman and Kelly,
2000), whereas the open, porous canopy of larger-seeded Andean determinate
beans have been considered a resistant phenotype (Miklas et al., 2001; Park et
al., 2001) in semi-arid regions. 8unsi and Raven exhibit contrasting architecture,
and the effect of upright plant type on resistance to white mold was evaluated in
different years with different environmental conditions. Further studies to detect
more minor QTL associated with physiological resistance should be carried out in
populations derived from parents with similar architecture and growth habit.
Different genomic regions detected in this study in different years can be
used to improve resistance to white mold in common bean. The QTL detected on

85 may represent a new source of physiological resistance for white mold. The

56

identification of these different QTL is crucial because it allows one to enhance
resistance by selecting more than one QTL from different linkage group

regardless of the type of resistance mechanism.

57

References

Abawi, GS, and RC. Grogan. 1975. Source of primary inoculum and effects of
temperature and moisture on infection of beans by Sclerotinia
sclerotiorum. Phytopathology 65:300-309.

Abawi, G.S., F.J. Polach, and WT Molin. 1975. Infection of bean by
ascospores of Whetzelinia sclertoiorum. Phytopathology 65:673-678.

Arahana, V.S., G.L. Graef, J.E. Specht, J.R. Steadman, and KM. Eskridge.
2001. Identification of QTL for resistance to Sclerotinia sclerotiorum in
soybean. Crop Sci. 41 :180-188.

Basten, CJ, BS Weir, Z-B Zeng. 2001. QTL Cartographer, Version 1.15. Dept. of
Statistics, Norht Caroline State University, Raleigh, NC.

Bert, P.F., l. Jouan, D. Tourvieille de Labrouhe, F. Serre, P. Nicolas, and F. Vear.
2002. Comparative genetic analysis of quantitative traits in sunflower
(Helianthus annuus L.) 1. QTL involved in resistance to Sclerotinia
sclerotiorum and Diaporthe helianthi. Theor. Appl. Genet. 105:985-993.

Blad, 8.L.; J.R. Steadman, and A. Weiss. 1978. Canopy structure and irrigation
influence white mold disease and microclimate of dry edible beans.
Phytopathology 68:1431-1437.

Brambl, R., and W. Gada. 1985. Plant seed lectins disrupt growth of germinating
fungal spores. Physiol. Plant 64:402-408.

Churchill, GA, and R.W. Doerge. 1994. Empirical threshold values for
quantitative trait mapping. Genetics 138:963-971.

Coyne, D.P., J.R. Steadman, and PM Anderson. 1974. Effect of modified plant
architecture of Great Northern dry bean varieties (Phaseolus vulgaris) on
white mold severity and components of yield. Plant Disease Reporter
58:379-382.

Coyne, D.P., J.R. Steadman, and HF. Schwartz. 1977. Reaction of Phaseolus
dry bean germoplasm to Sclerotinia sclerotiorum. Plant Disease Reporter
61 :226—230.

Coyne, D.P., J.R. Steadman, and HF. Schwartz. 1978. Effect of genetic blends
of dry beans (Phaseolus vulgaris) of different plant architecture on
apothecia production of Sclerotinia sclerotiorum and white mold infection.
Euphytica 27:225-231.

58

Coyne, DP. 1980. Modification of plant architecture and crop yield by breeding.
HortScience 15:244-247.

Edwards, KC, 0. Johnstone and C. Thompson. 1991. A simple and rapid
method for the preparation of plant genomic DNA for PCR analysis. Nuc.
Acids Res. 19:1349.

Fehr, W. 1987. Heritability. pp. 95-105. In: Principles of cultivar development,
theory and technique, v.1. Iowa State University, Ames, U.S.A.

Ferraz, L.C.L., A.C. Cafe Filho, L.C.B. Nasser. 1999. Effects of soil moisture,
organic matter and grass mulching on the carpogenic germination of
sclerotia and infection of bean by Sclerotinia sclerotiorum. Plant Pathology
48:77-82.

Freyre, R., P.W. Skroch, V. Gelfroy, AF. Adam-Blondon, A Shirmohamadali,
W.C. Johnson, V. Llaca, R.0. Nodari, P.A. Pereira, S.M. Tsai, J. Tohme,
M. Dron, J. Nienhuis, C.E. Vallejos. P. Gepts. 1998. Towards and
integrated linkage map of common bean. 4. Development of a core
linkage map and alignment of RFLP maps. Theor. Appl. Genet. 97:847-
856.

Gerlagh, M., H.M. Goossen-van de Geijn, N.J. Fokkema, and RFC. Vereijken.
1999. Long-term biosanitation by application of Coniothryrium minitans on
Sclerotinia sclerotiorum-infected crops. Phytopathology 89:141-147.

Godoy, G., J.R Steadman, M.8. Dickman, and R. Dam. 1990. Use of mutants to
demonstrate the role of oxalic acid in pathogenicity of Sclerotinia
sclerotiorum on Phaseolus vulgaris. Physiological and Molecular Plant
Pathology 37:179-191.

Haas, J.H., and B. Bolwyn. 1972. Ecology and epidemiology of Sclerotinia wilt
of white beans in Ontario. Can. J. Plant Sci. 52: 525-533.

Haley, 8.0., PM Miklas, L. Afanador, and JD. Kelly. 1994. Random amplified
polymorphic DNA marker variability between and within gene pools of
common bean. J. Am. Soc. Hort. Sci. 119:122-125.

Hall, R., and LG. Phillips. 1996. Evaluation of parameters to assess resistance
of white bean to white mold. Annu. Rep. Bean lmprov. Coop. 39:306-307.

Hart, PH, and AW. Saettler. 1981. White mold of beans. Cooperative Extension
Service, Michigan State University, Extension Bulletin E-892.

59

Harushima, Y., M. Yano, A. Shomura. 1998. A high-density rice genetic linkage
map with 2275 markers using a single F2 population. Genetics 148:479-
494.

Hazen, SP, P Leroy, R Ward. 2002. AFLP in Triticum aestivum L.: patterns of
genetic diversity and genome distribution. Euphytica 125:89-102.

Huang, H0, and RS. Erickson. 2000. Biocontrol of apothecial production of
Sclerotinia sclerotiorum in pulse and oiseed crops. Annual Report of the
Bean Improvement Cooperative 43:90-91.

Hunter, J.E., M.H. Dickson, and A. Cigna. 1981. Limited-term inoculation: a
method to screen bean plants for partial resistance to white mold. Plant
Disease 65:414-417.

Kelly, J. D. 2001. Remarking bean plant architecture for efficient production.
Advances in Agronomy. 71 :109-144.

Kelly, J.D., G.L. Hosfield, G.V. Varner, M.A. Uebersax, S.D. Haley, and J. Taylor.
1994. Registration of Raven black bean. Crop Sci. 34:1406-1407.

Kelly, J.D., P. Gepts. P.N. Miklas, and DP. Coyne. 2003. Tagging and mapping
of genes and QTL and molecular marker-assisted selection for traits of
economic importance in bean and cowpea. Field Crops Res. 82:135-154.

Kim, HS, and B.W. Diers. 2000. Inheritance of partial resistance to Sclerotinia
stem rot in soybean. Crop Sci. 40:55-61.

Knapp. S.J., W.W. Stroup, and WM. Ross. 1985. Exact confidence intervals for
heritability on a progeny mean basis. Crop Sci. 36:1037-1045.

Kolkman, J.M., and JD. Kelly. 1999. Inheritance to white mold in common bean.
Annual Report of the Improvement Cooperative 42:47-48.

Kolkman J.M. 2000. Genetic dissection of resistance to white mold (Sclerotinia
sclerotiorum) in common bean (Phaseolus vulgaris) Thesis (Ph.D.)
Michigan State University. Plant Breeding and Genetics Program, Dept. of
Crop and Soil Sciences.

Kolkman, J.M., and JD. Kelly. 2000. An indirect test using oxalate to determinate
physiological resistance to white mold in common bean. Crop Sci. 40:281-
285. ‘

Kolkman, J.M. and Kelly, JD. 2002. Agronomic traits affecting resistance to white
mold in common bean. Crop Sci. 42-693—699.

60

Kolkman, J.M. and Kelly, JD. 2003. QTL conferring resistance and avoidance to
white mold in common bean. Crop Sci. 43:539-548.

Lamey, A. 1996. Disease control update. Northarvest Bean Grower springz22.

Lander, E.S., P. Green, J. Abrahamson, A. Barlow, M. J. Daly, S.E. Lincoln, and
L. Newburg. 1987. MAPMAKER: An interactive computer package for
constructing primary genetic linkage maps of experimental and natural
populations. Genomics 1 : 174-181.

Mestries, E., L. Gentzbittel, D. Tourvieille de Labrouhe, P. Nicolas, and F. Vear.
1998. Analyses of quantitative trait loci associated with resistance to
Sclerotinia sclerotiorum in sunflowers (Helianthus annuus L.) using
molecular markers. MoI. Breed. 4:215—226.

Michelmore, R.W., I. Paran, and RV. Kesseli. 1991. Identification of markers
linked to disease resistance genes by bulked segregant analysis: a rapid
method to detect markers in specific genomic regions using segregating
populations. Proc. Natl. Acad. Sci. USA 88:9828-9832.

Miklas, P.N., K.F. Grafton. and 80. Nelson. 1992a. Screening for partial
physiological resistance to white mold in dry bean using excised stems. J.
Amer. Soc. Hort. Sci. 117:321-327.

Miklas, P.N., K.F. Grafton, G.A. Secor, and PE. Mcclean. 1992b. Use of
pathogen filtrate to differentiate physiological resistance of dry bean to
white mold disease. Crop Sci. 32:310-312.

Miklas, P.N., and K.F. Grafton. 1992c. Inheritance of partial resistance to white
mold in inbred populations of dry bean. Crop Sci. 32:943-948.

Miklas, P.N., K.F. Grafton, and P. E. McClean. 1993. Estimating phenylalanine
ammonia-lyase activity in common beans inoculated with Sclerotinia
sclerotiorum. HortScience 28:937-938.

Miklas, P.N., R. Delorme, R. Hannan, and M.H. Dickson. 1999. Using a
subsample of the core collection to identify new sources of resistance to
white mold in common bean. Crop Sci. 39:569-573.

Miklas P.N., R. Delorme, W.C. Johnson, and P. Gepts. 2001. QTL conditioning
physiological resistance and avoidance to white mold in dry bean. Crop
Sci. 41:309-315.

Miklas, P.N., and R. Delorme. 2003. Identification of QTL conditioning resistance
to white mold in snap bean. J. Amer. Soc. Hort. Sci. 128:564-570.

61

Nodari, R.0., S.M. Tsai, P. Guzman, R.L. Gilbertson, and P. Gepts. 1993.
Towards an integrated linkage map of common bean. lll. Mapping genetic
factors controlling host-bacteria interactions. Genetics 134:341-350.

Park, SJ. 1993. Response of bush and upright plant type selections to white
mold and seed yield of common beans grown in various row widths in
southern Ontario. Can. J. Plant Sci. 73:265-272.

Park 80., DP. Coyne, J.R. Steadman, and PW. Shroch. 2001. Mapping of QTL
for resistance to white mold disease in common bean. Crop Sci. 41:1253—
1262.

Petzoldt, R., and M.H. Dickson. 1996. Straw test for resistance to white mold in
beans. Annu. Rep. Bean Improv. Coop. 39:142-143.

Ronin , Y.l., A.B. Korol, and J.|. Weller. 1998. Selective genotyping to detect
quantitative trait loci affecting multiple traits: interval mapping analysis.
Theor. Appl. Genet. 97:1 169-1 178.

Roberts, M.E., M.H. Dickson, and J.E. Hunter. 1982. Heritability of white mold
resistance. Annu. Rep. Bean Improv. Coop. 25:104.

Ryder, T.B., S.A. Hedrick, J.N. Bell, X. Liang, S.D. Clouse, and OJ. Lamb. 1987.
Organization and differential activation of a gene family encoding the plant
defense enzyme chalcone synthase in Phaseolus vulgaris. Mol. Gen.
Genet. 210:219-233.

SAS Institute. 1995. The SAS system for Windows. Release 6.12. SAS Inst.,
Cary, NC.

Schneider, K.A., K.F. Grafton, and JD. Kelly. 2001. QTL analysis of resistance to
Fusarium root rot in bean. Crop Sci. 41 :535-542.

Schwartz, H.F., J.R. Steadman, and DP. Coyne. 1978. Influence of Phaseolus
vulgaris blossoming characteristics and canopy structure upon reaction to
Sclerotinia sclerotiorum. Phytopathology 68:465-470.

Schwartz, H.F., D.H. Casciano, J.A. Asenga, and DR. Wood. 1987. Field
measurement of white mold effects upon dry beans with genetic
resistance or upright plant architecture. Crop Sci. 27:699-702.

Shewry, P.R., J.A. Lucas. 1997. Plant proteins that confer resistance to pests
and pathogens. Adv. Bot. Res. 26:135-192.

Singh, SP. 1982. A key for identification of different growth habits of Phaseolus

62

vulgaris L. Annu. Rep. Bean Improv. Coop. 25:92-95.

Singh, PS. 1999. Integrated Genetic Improvement. lnzCommon bean
improvement in the twenty-first century. S.P. Singh (ed). Kluver Academic
Publishers, Dordrecht, The Netherlands.

Stam, P. 1993. Construction of integrated genetic linkage maps by means of new
computer package: Joinmap. Plant J. 32739-744.

Steadman, J.R., D.P. Coyne, and GE. Cook. 1973. Reduction of severity of
white mold disease on great northern beans by wider row spacing and
determinate plant growth habit. Plant Disease Reporter 57:1070-1071.

Steadman. JR. 1983. White mold- a serious yield-limiting disease of bean. Plant
Disease 67:346-50.

Steadman, JR. 1979. Control of plant diseases caused by Sclerotinia species.
Phytopathology 69:904-907.

Steadman, J. R., K. Powers, and 8. Higgins. 1997. Screening common bean for
white mold resistance using detached leaves. Annu. Rep. Bean Improv.
Coop. 40:140-141.

Steadman, J.R., R. Jung, M.S. Adams, K. Powers, and B Higgins. 2000. Random
amplified polymorphic DNA (RAPD) distinguishes three species of
Sclerotinia sclerotiorum but not pathogenic variability in S. sclerotiorum
isolates from diverse host and geographic origin. Annu. Rep. Bean
Improv. Coop. 43:158-159.

Sutton, 00, and B.J. Deverall. 1984. Phytoalexin accumulation during infection
of bean and soybean by ascospores and mycelium of Sclerotinia
sclerotiorum. Plant Pathology 33:377-383.

Tar’an, 8., TE. Michaels, and KP. Pauls. 2001. Mapping genetic factors
affecting the reaction to Xanthomonas axonopodis pv phaseoli in
Phaseolus vulgaris L. under field conditions. Genome 44:1046-1056.

Toubart, P., A Desiderio, G. Salvi, F. Cervone, L. Daroda, and G. De Lorenzo.
1992. Cloning and characterization of the gene encoding the
endopolygalacturonase-inhibiting protein (Pgip) of Phaseolus vulgaris L.
Plant J. 2:367-373.

Tu, JG 1985. Tolerance of white bean (Phaseolus vulgaris) to white mold

(Sclerotinia sclerotiorum) associated with tolerance to oxalic acid.
Physiological Plant Pathology 26:11 1-1 17.

63

Tu, J.C., and W.D. Beversdorf. 1982. Tolerance to white mold (Sclerotinia
sclerotiorum (Lib.) De Bary) in Ex Rico 23, a cultivar of white bean
(Phaseolus vulgaris L.). Can. J. Plant Sci. 62:65-69.

Vos, P., R. Hogers, M. Bleeker, M. Rijans, T. van der Lee, M. Hornes, A. Frijters,
J. Pot, J. Peleman, M. Kuiper, and M. Zabeau. 1995. AFLP: A new
technique for DNA fingerprinting. Nuc. Acid Res. 23:4407-4414.

Walter, M.H., J. Liu, 0. Grand, C.J. Lamb, and D. Hess. 1990. Bean-
pathogenesis-related proteins deduced from elicitor—induced transcripts
are members of a ubiquitous new class of conserved PR proteins
including pollen allergens. Mol. Gen. Genet. 222:353-360.

Whipps, J.M. 2001. Developments in the use of Coniothyrum minitans for the
biocontrol of Sclerotinia sclerotiorum. The XI International Sclerotinia
Workshop. York, UK.

Williams, J.G.K., A.R. Kubelik, K.J. Livak, J.A. Rafalski, and S.V. Tingey. 1990.
DNA polymorphisms amplified by arbitrary primers are useful as genetic
markers. Nuc. Acids Res. 18:6531-6535.

Young, W.P., J.M. Schupp, and P. Keim. 1999. DNA methylation and AFLP

marker distribution in the soybean genome. Theor. Appl. Genet. 99:785-
790.

64

Chapter II
MARKER-ASSISTED SELECTION FOR WHITE MOLD RESISTANCE
IN COMMON BEAN.

Introduction

Complexly inherited traits such as resistance to white mold (Sclerotinia
sclerotiorum) seem to have the most potential for marker assisted selection
(MAS), because environmental effects confound direct selection. Since the cost
to evaluate these complex traits with low to moderate heritability is high, MAS
would be more likely to show the greatest gain. However, for many reasons
including complex inheritance, low heritability, and environmental effects,
markers associated with QTL have been difficult to find, account for small
amount of actual genetic variability, and when available have exhibited limited
usefulness across a range of genetic backgrounds or environments (Kelly and
Miklas, 1999).

Application of MAS in plant breeding was analyzed by Staub and Serquen
(1996) where they considered that the potential utility of MAS is limited by: 1) the
number of molecular marker loci required to detect all significant associations; 2)
population sizes required to detect QTL for traits with low heritability; 3) the
sampling errors associated with the weighting of indices when combining
molecular marker loci; and 4) phenotypic information and cost per unit of
information gained.

Greater genetic gain can be made when flanking markers tightly linked to

QTL are used as compared to single markers loosely linked to a QTL. The trade

65

off is that it requires the characterization of twice as many markers as compared
to selection using markers.

To improve the effectiveness of MAS, markers closely linked to the QTL
need to be detected. Initially, the QTL needs to be precisely located, which is
dependent on the population size, environmental conditions where the population
is being tested and accuracy in the rating process. Another important aspect is to
have available an adequately saturated linkage map at least in these areas were
the QTL are mapped. The effectiveness of any MAS procedure will depend on
the accuracy of the phenotypic classification of the complex trait of interest and
the degree of linkage between marker(s) and traits of interest.

Various strategies for plant breeding were tested by Lande and Thompson
(1990) using computer simulations to characterize MAS. Gain from selection of
quantitative traits based on estimated additive effects could be greater for MAS
than for phenotypic selection. The relative worth of MAS was greatest for
characters with low heritability when additive genetic variance was associated
with the marker. Later Gimelfarb and Lande (1994) demonstrated that MAS could
also be effective for non-additive characters when compared to phenotypic
selection.

Another limitation of using molecular markers to assist in selection of
quantitative traits is the level of phenotypic variability explained by the QTL. The
detection of QTL or combination of QTL that account for a significant amount of
phenotypic variation is always desired, otherwise the MAS may not be cost-

effective to be used in regular basis by breeders. Markers close to genomic

66

regions with QTL linked to complexly inherited traits that explain only a small
fraction of the phenotypic variability may not be considered useful by plant
breeders in a selection process.

Detection of QTL for complexly inherited traits requires larger populations
than those commonly used in simple genetic studies. Simulation studies
including different population sizes showed that with a population of 100
individuals, only from 3 to 6 of 10 theoretical simulated QTL controlling 63% of
the phenotypic variability were detected (Young, 1999), suggesting that in order
to obtain more reliable and comprehensive data about quantitative traits,
populations of at least 500 individuals are needed. In terms of practical
application, large populations are not only more difficult to develop but are also
more difficult to evaluate for complex traits. In addition, complexly inherited traits
are highly influenced by the environment and require increased replication that
results in very large field experiments. The increased number of years and
locations necessary to effectively evaluate a population for a quantitative trait
represents very high costs to plant breeding programs.

The effectiveness of large populations can quickly shrink if many data
points are missing. QTL mapping presumes accurate phenotypic scoring
methods, something that can be difficult to optimize in traits that are difficult to
score. Just a few individuals incorrectly scored can totally confound QTL
detection and mapping. High quality data from both field and the laboratory
combined with the latest biometric approaches should reduce the problem of

false positives in the identification of QTL (Young, 1999).

67

Methodologies of identification of QTL involve different aspects, including
the similarity of the genetic background of the parents used to develop the
mapping population, approaches to select candidate markers such as selective
genotyping (Lander and Botstein, 1989), bulked segregant analysis (BSA)
(Michelmore et al., 1991 ), and statistical methods to analyze the molecular data
and phenotypic data in order to predict QTL. More powerful software to detect
QTL was initiated with the development of Mapmaker-QTL (Lander et al., 1987)
and advances with computer programs based on algorithms like composite
interval mapping (CIM), analysis and permutation tests, such as QTL
cartographer (Basten et al., 2001; Churchill and Doerge, 1994). QTL
identification can be strongly affected by environmental factors. Methods, such
as CIM, allow the evaluation of QTL x E by combining field data from different
environments. Even when a promising QTL is identified, further work is required
to test QTL in a different genetic background or environments. Repetition over
different years and locations, tests in genetically unrelated populations and
analyses of near isogenic lines to isolate the effects of individual QTL, are
additional consideration in analyzing QTL.

Efforts to utilize MAS for the improvement of quantitative traits have been
limited, but may improve as better quantitative data is generated and denser
linkage maps become available from map co-integration across laboratories
(Freyre et al., 1998).

Resistance to white mold is affected by physiological responses, more

related with general plant defense mechanisms, than by avoidance mechanisms

68

that include phenological and morphological traits. Architectural traits are
normally easier to select in the field due the higher heritability. MAS probably
would not be cost-effective for agronomic traits, but could be very important for
the indirect selection of physiological resistance. However, agronomic traits could
be included with other traits in the screening process of MAS. Often, qualitative
traits are easier, faster, and more cost effective to select for directly rather than
indirectly. In the case of agronomic traits such as plant growth habit, height,
flowering, and maturity, indirect selection will likely not improve efficiency over
regular field selection.

Non-specific resistance for white mold is not easily identified in the bean
progeny when compared to major genes for race specific resistance to other
bean diseases, because it is a quantitative trait and highly influenced by the
environment conditions. The identification of markers linked to different QTL for a
complexly inherited trait could represent a unique opportunity to bring together
different plant defense mechanisms and improve non-specific resistance or
horizontal resistance. The genotypes possessing most of these QTL associated
with general plant defense mechanisms could represent important progress
towards improving physiological resistance to white mold.

Use of MAS requires large segregating populations in order to have
sufficient genetic variability for other important traits. MAS can be very efficient in
early generations, but is also challenging if the goal is to select favorable alleles
at specific loci without excessively reducing genetic variability in the rest of the

genome. Considering the selection pressure required to fix alleles at several

69

specific loci of interest, a single MAS step would require working with large
segregating populations (Ribaut and Bertran, 1999).

MAS goes beyond the identification of markers linked to QTL controlling
important quantitative or qualitative traits. An effective strategy is needed in order
to use MAS in a selection process. The appropriate generation and population
size required for MAS, are some of the questions that need resolution before
using MAS technology in a breeding program. Simulation studies have
demonstrated the advantage of using DNA markers as an efficient
complementary tool to conventional selection, especially when the MAS is
applied in early generations of selection (Edwards and Page, 1994).

While many studies have been conducted to map QTL of important traits
in maize (Bernardo, 1999), very few empirical applications of MAS in breeding
programs have been reported. Most studies employing MAS have been
theoretical, using computer simulations (Knapp, 1998; Van Berloo and Stern,
2001; Ribaut et al., 2002). The results of such simulation studies suggest that
MAS can be used to improve quantitative traits in plant breeding programs.

In theory, MAS is proposed to be more efficient than phenotypic selection
when heritability of a trait is low, where tight linkage exists between QTL and
DNA markers (Knapp, 1998), particularly in earlier generations of selection
before recombinational erosion of marker-QTL associations interfere with indirect
selection (Lee, 1995). Comparing MAS and phenotypic selection for quantitative
traits in sweet corn, Yousef and Juvik (2001) observed that the average MAS and

phenotypic selection gain across different populations and selected traits was

70

10.9% and 6.1 %, respectively. Use of MAS is most appropriate when traits are
difficult and costly to measure. They concluded that incorporating DNA markers
into traditional breeding programs could expedite the selection progress and be
cost-effective.

Molecular markers (RAPD and AFLP) linked to white mold resistance
were identified by Kolkman and Kelly (2003) using selective genotyping and
BSA, in a population of recombinant inbred lines obtained from a cross between
8unsi (R) and Newport (S). The identified markers mapped to four linkage
groups, 82, B3, B7 and B8 on the integrated bean map. One major QTL for
disease incidence (DI) and disease severity index (DSI) was identified on linkage
group 82, and was located near RAPD markers 8020.1800 and 015.1800. The
region around 8020.1800 and 015.1800 on the integrated linkage map contains
several identified genes that may be important in disease defense responses.
The largest linkage group associated with white mold resistance was 87
(Kolkman and Kelly, 2003). Markers EAACMCTT130 and EAGGMCTT85 that mapped
in this linkage group were significantly associated with disease severity index,
days to flowering, lodging, yield and seed size. The combination of RAPD
marker 8020.1800 from 82 and AFLP markers, EAACMCTT130 and EAggMcn85
from 87 contributed to 30% of variability for DSI and 27% for DI. The use of
these markers in a strategy for selecting resistance to white mold was suggested
by Kolkman and Kelly (2003).

The goal of this study was to evaluate the potential of MAS to enhance

resistance to white mold, based on selection for RAPD marker 8020.1800 and

71

AFLP markers, EAAcMcrr130 and EAggMcn85 in a RIL mapping population of

8unsi (B) / Midland (S).

Materials and Methods
The MAS strategy was based on RAPD and AFLP markers linked to QTL

for resistance to white mold and identified in a F3-derived family population,
developed from a cross between two navy bean cultivars, 8unsi and Newport
(Kolkman and Kelly, 2003). MAS strategy was conducted on a 96 F47 RIL
population developed from a cross between 8unsi and Midland. 8unsi, with an
indeterminate (type II) growth habit, has physiological resistance and a porous
canopy for avoidance to white mold (Miklas and Grafton, 1992; Park et al., 2001).
Midland is a susceptible cultivar with a determinate (type I) growth habit. Ninety-
six F2 plants were advanced from F2 to F4 through single seed descent method in
the greenhouse. Seeds of individual F4 plants were bulked, and advanced in a
greenhouse. Seeds harvested from three F435 plants were bulked and F46 lines
were increased in the field, in Saginaw, MI, during summer 2000. No selection for

agronomic or phenological traits was made during the population development.

Plant tissue for DNA extraction was collected from parental genotypes and
RIL from F45 plants cultivated in the greenhouse. DNA was extracted from the
plant tissue using a mini-prep procedure (Edwards et at., 1991; Haley et al.,

1994).

The 96 RlLs were scored based on one RAPD marker, 8020.1800
(R2=12.6% for DSI), mapped on linkage group 82 and two AFLP markers

EAACMCTT13O (R2=15.8% for DSI) and EAggMcn85 (R2=16.8°/o for DSI) located

72

on linkage group 87. These three markers combined accounted for 30% of the
phenotypic variability for DSI in 8unsi/Newport RIL across four environments
(Kolkman, 2000). 80201800 and EAegMcn85 are markers in repulsion phase
and marker EAACMCTT130 was detected in coupling phase with resistance to
white mold. The source of the three markers was 8unsi, the resistant parent from
the RIL population where the three markers linked to QTL for resistance to white

mold were identified.

Polymerase chain reaction for the RAPD marker was conducted in a 96-
well FTC-100 programmable thermal controller (MJ Research, Inc) programmed
for 3 cycles of 1 min at 94 °C, 1 min at 35 °C, and 2 min at 72 °C; 34 cycles of
min at 94 °C, 1 min at 40 °C, and 2 min at 72 °C with the final step extended by 1
s for each of the 34 cycles, and a final extension cycle of 5 min at 72 °C (Haley et
al., 1994). RAPD markers are identified by the name of the Operon primer,
followed by the size of the polymorphic fragment, in reference to a 100 base pair
DNA ladder (Invitrogen). The AFLP analysis was carried out as described by
Hazen et al. (2002). Polyacrylamide gel electrophoresis was used to separate
AFLP fragments. Fragments were visualized using a silver staining procedure
(Promega). Gels were scored for band polymorphism, estimating band size in
reference to a 100 and a 10 bp DNA ladder.

Two groups (selected=S and control=C) of 10 RlLs each were selected
based on the MAS strategy used. The S group was selected based on the
presence of three markers in the linkage phase (coupling or repulsion)

associated to the QTL conferring resistance to white mold. The lines include in S

73

group had present the AFLP marker EMCMCTT130 and absence of RAPD marker
80201800 and AFLP marker EAGGMcr-r85. The control group was selected
based on the absence of marker EAACMCTT130 and presence of markers
80201800 and EAegMcn85. To investigate the association of growth habit with
resistance to white mold, five lines selected in each group were either
determinate or Indeterminate. The 20 RlLs and the parents, 8unsi and Midland,
were planted in Montcalm County, MI, in June of 2001 and 2002. Plot rows were
6 m in length, with 0.5 m row spacing. The two center rows of each plot were
planted with the RIL, while the two border rows were planted with the susceptible
parent, Midland.

The soil type at the Montcalm Research Farm sites is a combination of
Eutric Glossoboralfs (coarse-loamy, mixed) and Alfic Fragiorthods (coarse-
loamy, mixed, frigid). Standard agronomic practices for tillage, fertilization, insect
and weed control were applied to ensure adequate plant growth and
development. Plots were irrigated during the whole growing season starting in
the third week after planting with 12 mm of water at least once a week,
depending on rainfall, in order to promote uniform disease pressure across the
field. During 2001 plots were irrigated 8 times, totalizing 127 mm. In 2002 the

plots were irrigated 11 times with approximately 140 mm of water.

Description of the traits:
Plots were rated for disease prior to harvest, when the plants had reached
the physiological maturity. Thirty plants per plot were rated from 0 to 4, where 0 =

no disease present, 1 = 1 to 25 %; 2 = 26 to 50 %; 3 = 51 to 75 %; and 4 = 76%

74

to 100 % of the plant with white mold symptoms in order to estimate DSI and DI
(Hall and Phillips, 1996). DSI was calculated for each plot on a percentage basis,
using the following formula:

DSI (%) = i I 1
4 x (number of plants rated)

Disease incidence was calculated as the number of plants out of thirty evaluated
individuals with white mold infection, based as a percentage. The RlLs were also
evaluated for agronomic and phenological traits, including: yield, seed size, days
to flowering, and days to maturity, canopy height, architecture and lodging. Days
to flowering were characterized by the number of days following planting, when
50% of the plants in a plot have at least one open flower. Around four weeks
after flowering the plots were evaluated for architecture, using a 1 to 5 scale,
where 1 = fully upright, 3 = bush, and 5 = prostrate. Lodging was determined at
the same time that architecture was rated, based on a 1 to 5 scale, where 1 = no
lodging, 3 = moderate lodging, and 5 = excessive lodging. Days to maturity were
calculated as the number of days following planting, to when 90% of the pods
were physiologically mature and drying down. Canopy height was measured at
plant maturity on each individual plot, from six measurements per plot. All plots
were harvested at maturity, when plots were individually pulled and threshed.
Seed yield and weight of 100 seeds were adjusted to 18 % moisture content. In
addition, the 20 RlLs and the parents were screened in the greenhouse based on
resistance to oxalate (ROX), an indirect assay to determine physiological
resistance to white mold (Kolkman and Kelly, 2000). Twenty-one-day old

seedlings (2"d trifoliate emerging) were cut at the base of the stem and placed in

75

a 20 mM oxalate solution (pH 4.0). The seedlings were rated for wilting
symptoms after 12 to 15 h of exposure to the oxalate solution (approximately 6 to
9 h of daylight). A 1 to 6 scale was used to measure wilting, where 1 = no wilting
symptoms and 6 = total plant collapsing, as described by Kolkman and Kelly
(2000)

The experiment was analyzed as a mixed model, using PROC MIXED
(SAS, 1995) considering the groups (selected and control) and growth habit
(determinate and indeterminate) as fixed factors and the genotypes as a random
factor. The Pearson correlation coefficients (7) were calculated by PROC CORR

(SAS, 1995).

Results and Discussion

The potential of MAS based on RAPD marker 80201800 and AF LP
markers, EAACMCTT130 and EAegMcn85 linked to QTL for resistance to white
mold, identified previously in the cultivar 8unsi (Kolkman and Kelly, 2003), was
confirmed in the F4 derived RlLs originated from a cross between 8unsi and
Midland. Based on two years of field evaluation under white mold pressure, the
RlLs included in the selected (S) group developed less disease than the control
(C) group (Table 2.1). The two groups of RlLs resulting from MAS used in this
study were significantly different for DSI (p=0.034) and DI (p=0.057). The
absolute differences either for DSI or DI were around 10%, what is questionable
in terms of efficiency of MAS based on these three markers. However, the
combination of these three markers in the original population where they were

detected, contributed to 30% for DSI and 27% for DI, suggesting that although

76

significant, the genetic gains would not be very large. There was no significant
interaction between groups and growth habit or between groups and years,
indicating the consistency and importance of the QTL associated with the
markers used in this MAS study. The selected group compared to the control
group was more resistant to oxalate (Table 2.1). Growth habit was not associated
with resistance to white mold in both groups, contradicting previous studies that
suggested that determinate growth habit was associated with higher
susceptibility to white mold (Kolkman and Kelly, 2002). The lack of association
between growth habit and resistance to white mold may have been the
consequence of the reduced number of lines evaluated in this study.

The two groups resulting from the MAS were significantly different
(p=0.046) in yield (Table 2.2). Higher yield could be a consequence of less
disease and could also be a direct effect of the QTL close to marker
EMCMCTT130 that mapped to B7 (Kolkman and Kelly, 2003). QTL linked to
marker EMCMCTH 30 marker accounted for 34.4% of phenotypic variability for
yield in the 8unsi/Newport RIL population and was also associated with QTL for

days to flowering (R2=13.9%) and days to maturity (R2: 6.0%).

77

Ill-IE

v 0 C~£flih

.29.. 92; o. 8:228. .o. #0 o. 82:: 85.286

.929: of? new oomfiomom .628... Dn_<m_ .o 8:688 2cm 0998292.”. .929: 3.72 .0 88QO 3:05 6.2.00 a
.22: 9...; o. 8:928. .o. ..8 o. 82:: .mmtozooam

$2.9: n_._n_< 2cm oomfiomom .mxBE 03$. .0 88QO new omFto—zoém .929: 32“? .o 8:685 820.0 nofloflmm N

dozen—.665 m 5 8839.8 83 Soow
2.3. 98 2.95:8: $22me 9 62:9.me 019996 2 2.928.: E0... 088 m :0 comma mm; 8.2me 8 893281 .

 

 

 

2: New EN m. a 3m mam man 342 >0
em QB 98 5.8 8.8 New 28 am. 822
mm 28 8.8 0.8 0.8 5.8 a. 8 2.222
3 Fee $5 3... m. E «.2 5.8 28 28.5...
m: m: w: 62.3.1
em 2: 5.8 «.2 5.8 28 54 _. 83

.52.
2m 8.8 «.8 58 4.8 38 4.8 _ 69C 26.6

Boo .85 48.5 529.0
ea «S 6.8 QR v. z. 8.9. N? m6.2.8
em 28 58 a. 8 5.8 a. B 5.8 N8.92am 955 922

8268 SB Sow 82588 88 Bow
Fxom Qua 34:8

 

.Noow 2cm room E .__2 8:82:22 Cm 6.88.2 20:. 9E; .29.: £505 33... 2:222:23

.0 820.0 922 03. 65 .6 .00”: 2288 o. 8:928. .29 8:885 8822 .chV 825 3:98 8820 SN 63m...

78

Although MAS did not significantly affect days to flowering and days to maturity
(Table 2.2), both traits were associated with DI (Table 2.3), suggesting that
earliness could function as an avoidance mechanism. However, selection for
early genotypes can result in reduction of yield potential, since days to maturity
was positively associated with yield (r=0.39**). Combining acceptable agronomic
traits with resistance to white mold is crucial to select for useful genotypes in a
breeding program.

The selected group exhibited more upright plants compared to the
control group (Table 2.2). The architecture ratings for the selected and control
groups were 2.8 and 3.3 (p=0.043), respectively. In addition, architecture was
associated with DSI (r=0.37**) and DI (r=0.30**) (Table 2.3). Architecture may be
part of the plant avoidance mechanisms that in combination with physiological
resistance could result in higher levels of resistance to white mold. DSI was
negatively associated with yield and seed size, indicating the effect of white mold
disease on those traits. Consequently yield and seed size were positively
correlated. ROX presumed to be associated with physiological resistance, was
only associated with seed size. A biological association between seed size and
resistance to oxalate is not known. The correlation coefficient between ROX and
DSI was 0.38 (p =0.10), suggesting that both traits might be associated. ROX
could play a role in the physiological resistance against S. sclerotiorum and
consequently would affect the field resistance that is measured by DSI. The
correlation for ROX was calculated based on genotype means (N=20), compared

to other traits where the correlation was estimated on a plot basis (N=120).

79

.oFd A 02.9... m

.968. 93898 n m :5 6:58. 9939: u m .mEmno. o: n F 92.3 .m o. P 9.8m v

.memoa u m ucm .53 n m .EQE: 2.2 u .. 29.3 .m o. F 2.8m m
6.0.: 923 o. 8:539 co. ..._.0 2 no.5. date—202m.

5me n...n.< ocm oomfiomom 5me 0&5”. “.o 85on En 820.292.“. 5me n.......< .6 95¢QO d390, .350 N
.29: 2...; 9 8:923: :8 ..ko 0. no.5. date—202m

53:9: n......< ncm oomfiomom 53:9: on.<m. .o 8:38 :5 omrtozoém 8me n...u.< .o 8585 x590 862% c

 

 

:9 2: «:0. mam on E: 3... >0
mod S ace «.9. :2 $8 .3: cm. :85.
Z 3 «.8 N. S at 58 222.2
1m o... 38 a? 2: 5:. .23 955:
«cod me6 w: m: m: w: @5971
em 2.“ m. 2: 09. Q8 38 __ 83
em Z N2: 0.? 02 8mm _ we? .52 53,90
m: 9.5.: m: :3: 3o: 03.: ass§d
Nm 0m «.2: 3... 3: :8 6:80
3 mu m. 2: mm... 08 85 8.8.8 :85 ms).
6 o. 5 a o. 5 5.55.2 9.5.52“. .783, 8:9 8&3.
3:68.. "9202.52. 9 980 0. £60 mum comm 29>

 

.Noom ncm Sow
:_ ...>. .E_mo.:os_ :_ 98on 20.: 2...; 5.3:: €505 €928 ucm $20203 may. “3:322:33 ..o 8305
ms). 025 or: .6 @580. ucm 2:82.55 Ema SE39: 2 9m: 6:526: 2 £8: .m~.m comm 6.9.» cocfiEoo .N.N p.92.

80

.29.. 9:; o. 858.8. .o. #o o. :95. .mmtozoofi
.9..mE n.._u_< ::m oom...owom .9..m:. oo<m ..o 8:89: ::m omrtoioéw .9..m:. n.._n_< .o 8:88 .99: 6:80 .0.
.29.. 9:; o. 888.8. .o. .5 o. 8...: 85.288.
.9..m:. n...u.< ::m oomfiomOm .9..mE 921.: 8:88 ::m ogre—292m. .98.: 3...? .o 8:89: anoa 8.8.8 N

.Smnz. :88 830:8

:o :88 9:.me o. 8:989 .9 8.8.38 82, . 890:3 .Aomruz. 6.: .m::.>.::. :o :88 888:8 8.8.9.00F

 

 

:86 9309.52
.8: 2.9.: - 3.5.9. o. 9m:
8:950:
:8: 2.3.: :Nvd - o. 9m:
m: m: :Rmd m: - o~.w :86
..m to ..NN.o :8: m: :8: - 22>
m: w: w: m: bio- w: - xom
8.: :5: .8: .8: .96. LN..- 9 - 5
..NN: :8: 9 km: :86. :30- m: 2.8.: - .8
9.68.. 93096.4. €3.22 9.532“. 8m 88 22> xom a .8

 

.NOON new room

:. .=>_ 9.8.8.2 :. .9389: :.o:. 9.5, .o::: :265 5.9.8: ::m N:9om_8. 8:9: 92 9.. :. :828. 8.x
::m_:._2\.m::m om 9.. .9 85:0. ::m 93:930.: .52: 5:29: 9 96: .8826: 9 9m: .88 :88 .:.o.>
.Axom. 9:.me 2 888.8. .20. 8:865 088.: émov xo::. 3:88 8:86 .6 F5.6.9.8 cabococn. .md 98..

81

Thus the power of detecting associations is reduced with smaller numbers
evaluated for ROX.

This study was conducted to determine the effectiveness of MAS for
complexly inherited traits such as resistance to white mold in common bean. With
the detection of QTL and discovery of markers linked to these traits and the
application of MAS for white mold resistance should become more routine in
bean breeding programs. MAS can be performed in the absence of the pathogen
and in early generations when genetic resistance is not clearly expressed due to
environmental variation in the field. Usually the contribution of each QTL to a
quantitative trait is small, which makes pyramiding different QTL crucial to

significantly improve the genetic gain for a complexly inherited trait.

A strategy that could be used to improve resistance to white mold would
combine phenotypic selection for favorable agronomic traits with molecular
markers linked to QTL associated with resistance to S. sclerotiorum. Considering
that most of the agronomic traits have a high heritability, selection could
effectively be conducted in early generations and in preliminary yield trials.
Environments with less favorable conditions for white mold could be used to
avoid selection problems regarding cause/effect that frequently occur with traits
such as architecture and lodging that can significantly affect and/or be influenced
by disease occurrence or severity. Scoring agronomic traits without the
occurrence of disease would allow a more efficient selection for these
architectural traits that could be combined with molecular markers associated

with physiological resistance. During the field season, the lines could be

82

screened for the markers included in the MAS strategy and during the final
phenotypic selection procedure; the breeder could select those lines possessing
favorable agronomic traits and the desirable QTL linked to resistance to white

mold.

The MAS study performed in Bunsi/Midland population was based only on
molecular markers without including any agronomic traits related with avoidance
mechanisms or yield. This approach could result in selection of lines that are not
commercially acceptable. The inclusion of agronomic traits in the selection
process is important in a plant breeding program and likely would be the most
effective way to combine resistance to white mold with desirable agronomic
traits. Combining different QTL for white mold resistance and agronomic traits
would require large populations in order to select lines that display the desired

combination of markers and agronomic traits.

A multiple regression analysis for DSI, including agronomic traits and
markers linked to QTL located on 82 and B7, was conducted by Kolkman and
Kelly (2003). The model proposed with the highest R2 (59.4%) was Y= -109.5 +
1.3 (canopy width) + 5 (EAACMCTT130) — 11.0 (8020.1800) + 0.9 (days to
flowering). The same two markers included in this model were used in addition to
EAGGMCTT85. A simulated selection using marker and agronomic variables of this
model was applied to another 98 entry RIL population derived from the cross
between Bunsi and Raven, using ratings for architecture that are related to
canopy width. From those traits measured in the Bunsi/Raven RIL population,

architecture is most similar to canopy width used in the model.

83

The effectiveness of MAS in a population of RIL derived from a cross
between Bunsi and Raven was tested. The selection was initially based only on
markers (Group M) EMCMCTT13O (presence) and BC20.1800 (absence), resulting
in 28 selected lines. The second step in the selection process was to include the
architecture canopy width trait (Group M+A). Lines with score for architecture <
2.5 and > 3.0 were excluded. These values were considered equivalent to a
medium canopy width and branching pattern, in a white mold resistant plant
ideotype proposed for US. Midwest environmental conditions (Kolkman and
Kelly, 2002). When yield was also included as a selection criterion, ten lines with
high yield were selected (Group M+A+Y). The control group (Group C) was
formed by lines with EAACMcn13O (absence) and BCZO.1800 (presence),

regardless of phenological and agronomic traits.

The group M (based on markers) was compared to the control group and
was significantly different for DSI and Di, suggesting that the MAS strategy
based on markers would be effective to select lines with higher levels of
resistance to white mold (Table 2.4). MAS was also effective in improving yield
and seed size, which could be a direct effect of the QTL on those traits or a
consequence of lower levels of white mold disease in the M group. Days to
flowering and days to maturity were affected by the selection strategy, however,
in practical sense the selected lines still fell in an acceptable commercial range.
Including architecture (group M+A) as a selection criterion did not result in RlLs
with higher levels of DSI and DI, which was unexpected, but could be explained

because some of the 10 lines excluded from group M to form group M+A, with

84

Table 2. 4. Disease severity index (DSI), disease incidence (DI), yield (Y), seed
size, days to flowering, days to maturity, lodging, and architecture (A)
of MAS groups derived from 98 entry RIL population of Bunsi/Raven,
grown under white mold pressure at Montcalm, MI, in 2001 and 2002.

Trait Population GroupM1 GroupM+A2 GroupM+A+Y3 Control4

 

mean(98) 5 (28) (18) (10) (14)
DSI (%) 36.1 33.2“6 33.0** 27.7** 40.4
DI (%) 63.9 59.9“ 59.5“ 52.5“ 72.5
Yield(kg/ha) 4315 4484" 4517** 4775** 4192
Seed size 21.6 22.4** 22.7** 22.6** 20.7
(g.100seed")
Days to
Flowering 44.5 44.0" 43.9** 43.9“ 45.7
Days to
Maturity 105.4 106.9“ 107.o** 107.5** 104.7
Lodging7 3.2 3.2ns 3.1ns 2.9ns 3.1
Architecture8 2.7 2.8ns 2.8ns 2.8ns 2.7

 

1 Selection based on presence of AFLP marker EAACMcn130 and absence of
RAPD marker BC20.1800, linked to QTL for resistance to white mold.

2 Selection based on markers (Group M) and architecture ratings between 2.5
and 3.0 were included.

3 Selection based on markers, architecture (GroupM+A) and yield (> 4370 kg/ha).

4 Selection based on absence of AFLP marker EMCMCTT130 and presence of
RAPD marker BC20.1800, linked to QTL for resistance to white mold.

5 Number of selected individuals included in each MAS group.

6 Comparisons against Group control, where **, * and ns indicates p< 0.01, 0.05
and non significant, respectively.

7Scale 1 to 5, where 1 = no lodging, 3 = moderate lodging, and 5 = excessive
lodging.

8 Scale 1 to 5, where 1 = fully upright, 3 = bush, and 5 = prostrate.

85

architecture ratings of 2.1 developed lower levels of disease probably due
avoidance. Progress for resistance to white mold was detected when yield was
included in the multivariate selection process. Therefore yield is an important trait
that should be considered when selecting for resistance to white mold. Selection
for yield under high white mold pressure would be an effective approach to
improve resistance to white mold. Selection only based on disease severity or
incidence can result in lines with limited potential use in breeding due to
combinations of undesirable agronomic traits.

Progress towards the effective use of MAS for quantitative traits in plant
breeding programs depends on the detection of QTL associated with those
complexly inherited traits in different genetic germplasm sources. The
identification of QTL is only one step in MAS. Detection of markers tightly linked
to the QTL and when possible, combinations of markers flanking the genomic
regions are important aspects to develop an effective strategy for MAS.
Pyramiding multiple QTL with individual minor effects is a challenging task, but
the tools are available and it seems that MAS will play a more important role in
breeding for quantitative traits in the future.

Growth habit was not a reliable predictor of the ability to escape white
mold infection under field conditions (Schwartz et al.,1987). Instead, the plant
canopy structure and canopy density associated with the respective growth
habits influence microclimatic conditions within the plant canopy that are more or
less favorable for S. sclerotiorum colonization, infection and spread (Schwartz et

al. 1978; Schwartz et al., 1987). Park (1993) obtained similar results and

86

concluded that narrow canopy, higher flowering nodes, and fewer branches in
the upright plant type might contribute to avoid white mold infection. Narrow
canopy plants allow better light penetration and air movement that results in rapid
drying of dew on leaves and reduced moisture on the soil surface, resulting in a
less favorable microclimate for white mold development (Coyne et. al., 1974;

Blad et al. 1978; Schwartz et al., 1987).

Kolkman and Kelly (2002) proposed that the small-seeded ideotype
resistant to white mold for the Midwest should possess physiological resistance
to white mold and be indeterminate (Type II), resistant lodging, medium canopy
width and branching pattern, and medium height, days to flower and maturity, to
ensure adequate vegetative growth and yield potential without creating a

favorable microclimate within the plant canopy for disease development.

Agronomic and phenological traits should be combined with MAS to select
commercially acceptable lines with a desired architecture, phenological cycle,
and yield potential, in addition to improved resistance to white mold. The
phenotypic selection for all these traits with high heritability could be conducted in
environments where the screening for the disease was not possible. The marker
data in addition to the phenotypic ratings should allow the breeder to more
efficiently select the best lines to be tested for white mold resistance in the next

generation.

87

Ba

0')
(‘1)

Cc

Ed

EC

Ha

Ha

References

Basten, CJ, BS Weir, Z-B Zeng. 2001. QTL Cartographer, Version 1.15. Dep. of
Statistics, Norht Caroline State University, Raleigh, NC.

Bernardo, R. 1999. Selection response with marker-based assortative mating.
Crop Sci. 39:69-73.

Blad, B.L.; J.R. Steadman, and A. Weiss. 1978. Canopy structure and irrigation
influence white mold disease and microclimate of dry edible beans.
Phytopathology 68:1431-1437.

Churchill, GA, and R.W. Doerge. 1994. Empirical threshold values for
quantitative trait mapping. Genetics 138:963-971.

Coyne, D.P., J.R. Steadman, and PM Anderson. 1974. Effect of modified plant
architecture of Great Northern dry bean varieties (Phaseolus vulgaris) on
white mold severity and components of yield. Plant Disease Report
58:379-382.

Edwards, K.C., C. Johnstone and C. Thompson. 1991. A simple and rapid
method for the preparation of plant genomic DNA for PCR analysis. Nuc.
Acids Res. 19:1349.

Edwards, MD, and N.J. Page. 1994. Evaluation of marker-assisted selection
through computer simulation. Theor. Appl. Genet. 88:376-382.

Freyre, R., P.W. Skroch, V. Gelfroy, AF. Adam-Blondon, A Shirmohamadali,
W.C. Johnson, V. Llaca, R.O Nodari, P.A Pereira, S.M. Tsai, J. Tohme, M.
Dron, J. Nienhuis, C.E. Vallejos. P. Gepts. 1998. Towards and integrated
linkage map of common bean. 4. Development of a core linkage map and
alignment of RFLP maps. Theor. Appl. Genet. 97:847-856.

Gimelfarb, A., and R. Lande. 1994. Simulation of marker assisted selection in
hybrid populations. Genet. Res. 63:39-47.

Haley, S.D., P.N. Miklas, L. Afanador, and JD. Kelly. 1994. Random amplified
polymorphic DNA marker variability between and within gene pools of
common bean. J. Am. Soc. Hort. Sci. 119:122-125.

Hall, R., and LG. Phillips. 1996. Evaluation of parameters to assess resistance
of white bean to white mold. Annu. Rep. Bean Improv. Coop. 39:306-307.

Hazen, S.P., P. Leroy, and R. Ward. 2002. AFLP in Triticum aestivum L.:
patterns of genetic diversity and genome distribution. Euphytica. 125:89-
102.

88

Kelly, JD and RN. Miklas. 1999. Marker-assisted selection. pp 93-132. In:
Singh, S.P. (ed). Common bean improvement in the twenty-first century.
Kluver Acad. Pub. Norwell, U.S.A.

Knapp. SJ. 1998. Marker-assisted selection as a strategy for increasing the
probability of selecting superior genotypes. Crop Sci. 38:1164-1174.

Kolkman JM. 2000. Genetic dissection of resistance to white mold (Sclerotinia
sclerotiorum) in common bean (Phaseolus vulgaris) Thesis (Ph.D.)
Michigan State University. Plant Breeding and Genetics Program, Dept. of
Crop and Soil Sciences.

Kolkman, J.M., and JD. Kelly. 2000. An indirect test using oxalate to determinate
physiological resistance to white mold in common bean. Crop Sci. 40:281-
285.

Kolkman, J.M. and Kelly, JD. 2002. Agronomic traits affecting resistance to white
mold in common bean. Crop Sci. 42-693-699.

Kolkman, J.M. and Kelly, JD. 2003. QTL conferring resistance and avoidance to
white mold in common bean. Crop Sci. 43:539-548.

Lander, E.S., P. Green, J. Abrahamson, A. Barlow, M. J. Daly, S.E. Lincoln, and
L. Newburg. 1987. Mapmaker: an interactive computer package for
constructing primary genetic linkage maps of experimental and natural
populations. Genomics 1: 174-181.

Lande, R., and R. Thompson. 1990. Efficiency of marker-assisted selection in
the improvement of quantitative traits. Genetics 124:743-756.

Lander, ES, and D. Botstein. 1989. Mapping Mendelian factors underlying
quantitative traits using RFLP linkage maps. Genetics 121:185-199.

Lee, M. 1995. DNA markers and plant breeding programs. Adv. Agron. 55:265-
344.

Michelmore, R.W., I. Paran, and RV. Kesseli. 1991. Identification of markers
linked to disease resistance genes by bulked segregant analysis: a rapid
method to detect markers in specific genomic regions using segregating
populations. Proc. Natl. Acad. Sci. USA 88:9828-9832.

Miklas, P.N., and K.F. Grafton. 1992. Inheritance of partial resistance to white
mold in inbred populations of dry bean. Crop Sci. 32:943-948.

89

Park, SJ. 1993. Response of bush and upright plant type selections to white
mold and seed yield of common beans grown in various row widths in
southern Ontario. Can. J. Plant Sci. 73:265-272.

Park 8.0., D.P. Coyne, JR Steadman, and PW. Shroch. 2001. Mapping of QTL
for resistance to white mold disease in common bean. Crop Sci. 41 :1253-
1262.

Ribaut, J-M., and J. Bertran. 1999. Single large-scale marker-assisted selection
(SLS-MAS). Molecular Breeding 52531-541.

Ribaut, J-M, C. Jiang, and D. Hoisington. 2002. Simulation experiments on
efficiencies of gene introgression by backcrossing. Crop Sci. 42:557-565.

SAS Institute. 1995. The SAS system for Windows. Release 6.12. SAS lnst.,
Cary, NC.

Schwartz, H.F., J.R. Steadman, and DP. Coyne. 1978. Influence of Phaseolus
vulgaris blossoming characteristics and canopy structure upon reaction to
Sclerotinia sclerotiorum. Phytopathology 682465-470.

Schwartz, H.F., D.H. Casciano, J.A. Asenga, and DR. Wood. 1987. Field
measurement of white mold effects upon dry beans with genetic
resistance or upright plant architecture. Crop Sci. 27:699-702.

Staub, J.E., and PC. Serquen. 1996. Genetic markers, map construction, and
their application in plant breeding. HortScience 31:729-741.

Van Berloo, R., P. Stam. 2001. Simultaneous marker-assisted selection for
multiple traits in autogamous crops. Theor. Appl. Genet. 102:1107-1112.

Young, ND. 1999. A cautiously optimistic vision for marker-assisted breeding.
Molecular Breeding 52505-510.

Yousef, G.G. and A. Juvik. 2001. Comparison on phenotypic and marker-

assisted selection for quantitative traits in sweet corn. Crop Sci.
41:645:655.

90

Chapter III

USE OF INBRED BACKCROSS METHOD TO INTRODUCE RESISTANCE TO
WHITE MOLD FROM EXOTIC GERMPLASM INTO COMMON BEAN

Introduction

White mold, caused by Sclerotinia sclerotiorum, is a serious disease of
common bean (Phaseolus vulgaris) that results in substantial yield loss and
reduced seed quality (Steadman, 1983). Genetic resistance to S. sclerotiorum in
cultivated bean is partial, highly affected by climate conditions, and confounded
by plant avoidance mechanisms (Schwartz et al., 1987; Miklas et al., 1993).
Different approaches to breeding for white mold resistance are needed due to
major differences in climate in different been growing regions of the world where
the disease occurs. Upright plant architecture seems to be effective in avoiding
the disease in more arid western production areas of the US, but offers less
advantage in the more humid Midwest states, including Michigan. Adequate air
movement through the canopy is critical to help reduce white mold infection, and
architectural avoidance needs to be combined with sources of physiological
resistance to enhance overall resistance in bean (Webster and Kelly, 2000).

A strategy to increase the genetic variability for physiological resistance to
white mold in cultivated bean varieties is to introgress genes from primitive
landrace varieties and related wild types of common bean. Moreover, the
potential of wild beans as a source of unique germplasm to improve resistance

and other traits including yield is receiving increased attention by breeders.

91

Exotic germplasm should be considered as source of novel variability for
physiological resistance to white mold. Improved architecture and other traits
considered as avoidance mechanisms are already available in cultivated
commercial bean genotypes. Adequate levels of physiological resistance is
lacking in most of the cultivated bean classes and is the most likely type of
resistance that would be expected to be introduced from exotic bean germplasm.

The genetic base of commercial cultivars within specific market classes is
narrow (McClean et al., 1993).The average world yield of common bean remains
low (<900 kg ha"), and bean production suffers from a wide range of abiotic and
biotic constraints. Improving resistance to abiotic stresses, such as drought and
soil mineral deficiency and toxicity, simultaneously with resistance to biotic
stresses, especially white mold, in high yielding cultivars for North America is a
challenging goal. In order to accomplish this task, breeders should consider that
as much as 90% of the genetic variability available in the primary gene pool of P.
vulgaris and related species remains under utilized (Singh, 1999).

An example of effective use of exotic bean germplasm is the introgression
of resistance to bruchid (Zabrotis subfasciatus) from wild Mexican bean
genotypes. The protein arcelin, a plant-resistance factor that inhibits Z.
subfasciatus growth, was identified in wild Mexican bean populations (Osborn et
al., 1988) and has been successfully bred into cultivated beans (Cardona et al.,
1990) using recurrent backcrossing. This protein is inherited as a single dominant
gene and at least seven co-dominant allelic variants are available for breeding

(Acosta-Gallegos et al., 1998; Cardona and Kornegay, 1999).

92

Compared with their wild progenitors, cultivated plants often show drastic
phenotypic differences in traits such as seed dormancy, seed dispersal
mechanisms, cycle (e.g. earliness), growth habit, photoperiod sensitivity and size
of the harvested parts (Koinange et al., 1996). The combined phenotypic
differences are called the domestication syndrome traits that result from selection
during several thousands of years for adaptation to cultivated environments. The
most important traits of the domestication syndrome in common bean are the
loss of seed dispersal ability conditioned by the presence of fibers in the pods,
and dormancy, conditioned by impermeability of the seed coat. The major genes
controlling the domestication syndrome in common bean were mapped to three
of the eleven linkage groups, B1, B2, and B8 (Freyre et al., 1998) of the bean
genome (Koinange et al., 1996). The major phenotypic effect of individual factors
and the high heritability of individual traits may reflect the domestication process.
Useful mutations with a major effect would have had a higher likelihood of being
detected and selected (Koinange et al., 1996). The simple genetic control of the
domestication syndrome traits suggests that the introgression of diversity from
the wild to the cultivated P. vulgaris should be feasible and not involve special
difficulties. Small genetic changes on B1, 82 and B8 in the wild types should
allow breeders to return to cultivated bean and therefore would permit the
evaluation of novel traits in a cultivated background. The potential offered by
landraces and wild types as sources of quantitative traits can only be tested

when QTL associated with the trait of interest are transferred into adapted

Qermplasm.

93

In terms of resistance to S. sclerotiorum, field resistance is the final goal of
breeders. Putative sources of resistance are detected by greenhouse tests but
need to be tested under high disease pressure in field conditions to verify the
reaction to S. sclerotiorum. However, before field screening can be conducted,
germplasm conversion may be needed to adapt many of these accessions to
temperate North American conditions where photoperiod sensitivity, late maturity,
and climbing growth habit interfere with field evaluation (Webster and Kelly,
2000)

Germplasm conversion consists of backcrossing insensitivity to
photoperiod and other agronomic traits from an adapted donor parent to the
unadapted accessions having putative resistance to S. sclerotiorum. Germplasm
conversion is a long term project. The level of adaptation of the resulting lines
can affect the breeders ability to identify superior sources for quantitative traits,
such as resistance to white mold. Lack of complete adaptation may result in
avoidance to white mold, but may be associated with undesired agronomic
and/or phenological traits. Alternatively, plant introductions can be backcrossed
into adapted cultivars and BCnFn progenies developed without selection can be
tested for resistance under field conditions. The inbred backcross method has
recently received attention as an effective method to transfer more complex
quantitative traits from unadapted or diverse germplasm into otherwise adapted,
productive cultivars (Bliss, 1993; Tanskley and McCouch, 1997; Hartman and St.
Clair, 1998). Different percentages of the donor parent can be incorporated into

the inbred backcross line population. The percentage of plant introduction

94

germplasm that should be used in a breeding population was studied in soybean
by Thorpe and Fehr (1970) using landraces and adapted varieties. The results
indicated that the use of 25% unadapted germplasm was more successful than
50% unadapted germplasm in soybean crosses.

Very large collections of germplasm from around the world have been
assembled in efforts to conserve the extant genetic variation of numerous
species (Miklas et al., 1999). To evaluate and utilize effectively these collections,
it is first necessary to identify a smaller subset or core collection for testing that
represents most of the genetic variation in the entire collection (Skinner et al.,
1999). Establishment of core collections, by including a representative sample
from the total accessions maintained in germplasm banks, facilitates the
evaluation of a wide array of genetic diversity for resistance to a complex
pathogen such as S. sclerotiorum. Once a core accession is identified with
potential resistance to certain pathogen, an expanded search for resistance can
be conducted among accessions with similar passport data in the active
collection (Miklas et al., 1999). On the basis of several statistical models, Brown
(1989) suggested that at least 70% of the alleles presented in the entire
collection will be represented in a core collection comprised of at least 10% of the
total accessions. The importance of core collections is recognized, but the core
accessions still need to be evaluated to detect useful genes. Since most of the
core accessions cannot be tested directly in the field due problems with
adaptation, it is necessary to use indirect screening methods to select for

resistance to S. sclerotiorum. Considerable efforts have been made to develop

95

greenhouse and laboratory screening methods in order to identify physiological
resistance to S. sclerotiorum (Hunter et al., 1981; Miklas et al., 1992a; Miklas et
al., 1992b; Petzoldt and Dickson, 1996; Steadman et al., 1997; Kolkman and
Kelly, 2000). The identification of putative sources of resistance is the step to
effectively exploit exotic germplasm in breeding programs. Only those selected
core accessions with putative resistance to S. sclerotiorum, based on different
screening methods, would be used in developing inbred backcross lines.

Crosses between cultivated and wild bean genotypes that contrast for
important traits could generate mapping populations that may permit more
detailed analysis about the genetic control of these traits. As the results
accumulate from molecular mapping studies in crop plants, patterns are likely to
emerge concerning the chromosomal positions of key loci controlling yield and
other important traits. Once this information is available, exotic germplasm could
be sampled for targeted areas of the genome through backcross breeding and
the generation of allelic series in near isogenic lines that would then be evaluated
agronomically (Tanskley and McCouch, 1997).

Attempts to use exotic germplasm have been conducted in different plant
species. Studies conducted in tomato and rice indicated that exotic germplasm
contains many new and useful genes that can significantly enhance agricultural
production, even for complex traits like yield (Tanskley and Nelson, 1996;
Tanksley and McCouch, 1997). The same studies also demonstrated that the
phenotype of exotic germplasm is a poor predictor of its genetic potential.

Combining molecular mapping techniques with inbred backcross was proposed

96

by Tanskley and Nelson (1996) as a method for simultaneously discover and
transfer valuable QTL from unadapted germplasm into elite breeding lines. Use
of molecular mapping techniques like the advanced backcross QTL method
could be crucial to identify important QTL and explore the genetic resources
available.

The introgression of a QTL for yield from Glycine soja into commercial
soybean cultivars was tested by Concibido et al. (2003). The detection of a QTL
from the exotic germplasm accounting for 9% of phenotypic variability for yield
demonstrated the potential of using exotic germplasm to improve yield in
soybean. The availability of a number of SSR markers around the QTL region
makes them good candidates for MAS. Thompson and Nelson (1998) evaluated
the genetic diversity and agronomic performance of experimental lines derived
from plant introductions (Pls) maintained in the USDA Soybean Germplasm
Collection. Experimental lines possessed 25, 50, 75 or 100% of PI germplasm
(based on pedigrees). The increased genetic diversity and yield provided
evidence that exotic germplasm can contribute genes with the potential to
increase yield.

Most of the important and unique alleles are masked by the
preponderance of undesirable alleles in the wild species which restricts the use
of the wild species for the genetic improvement of quantitative traits. Initially, wild
germplasm from within the same gene pool of P. vulgaris should be used to
increase genetic variability of quantitative traits, prior to using the wild germplasm

from the other gene pool. After successfully introducing resistance to S.

97

sclerotiorum from wild germplasm to adapted cultivars within the same gene
pool, the next step would be to transfer the sources of resistance between gene
pools. The use of exotic germplasm to introduce novel genetic variability for
important qualitative and quantitative traits is supported by plant breeders (lninda
et al., 1996; Concibido et al., 2003). The identification of sources of physiological
resistance to white mold in bean is the main goal where use is made of landraces
and wild genotypes as novel sources of genetic resistance to S. sclerotiorum. A
better understanding of the different mechanisms involved in plant defense
against S. sclerotiorum, and how they can affect the spread of fungus throughout
the plant are important aspects that could help to identify sources of physiological

resistance to S. sclerotiorum.

Physiological Resistance and Plant Defense Related Proteins

Physiological resistance is controlled by factors such as increased
activities of plant defense-related enzymes (Miklas et al., 1993) and phytoalexin
accumulation (Sutton and Deverall, 1984) that limit spread of the pathogen in
plant tissue. Miklas et al., ( 1993) measured the phenylalanine ammonia-lyase
(PAL) in detached stems inoculated with growing mycelium of S. sclerotiorum in
five bean cultivars with differing levels of resistance to S. sclerotiorum. Greater
PAL activity was observed in the resistant cultivar when compared to the
susceptible cultivar, suggesting that PAL activity may be involved in physiological

resistance of common bean to S. sclerotiorum.

98

In order to gain insight into physiological resistance and quantitative host-
pathogen relationships in general, the activity of plant defense related proteins

should be assayed in different genotypes with different levels of resistance to S.
sclerotiorum. B-1,3-Glucanases, chitinases and peroxidases are some of these

plant defense related proteins that might be involved in general plant defense

mechanisms (Van Loon, 1999).

B-1,3-Glucanases are capable of degrading B-1,3-glucans (Leubner-
Metzger and Meins, 1999) and are implicated in various functions since the
substrate is widespread in fungal cell wall (Gooday, 1994). [3-1,3-Glucanases
may be directly involved in fungal resistance by inhibiting fungal growth through

the degradation of fungal cell wall. B-1,3-Glucanases may also be involved
indirectly in fungal resistance by producing B-glucan elicitors by fungal cell wall

degradation. B-glucan elicitors have been reported to induce resistance against
fungal pathogens by inducing plant defenses (Bhandal and Paxton, 1991).
Chitin is a linear homopolymer of B-1,4-linked N-acetyl-D-glucosamine

residues and is a structural component of fungi and insects but is not found in
higher plants (Gooday, 1994). Endochitinase , the enzyme that randomly
hydrolyses internal B-1,4- linkages of chitin, is constitutively expressed in many
crop plants, including beans (Boller et al., 1983). The exact physiological function
of chitinases in higher plants is not known, but their existence has been

considered a measure of defense to pathogens and insects rather than a normal

metabolite essential for plant metabolism (Huang, 2001). Chitinases may be

99

involved directly in defense by breaking down fungal cell walls and indirectly
involved by producing elicitors that induce other plant defenses. Chitinase activity
in plants significantly increases in numerous incompatible host-parasite
interactions. Increases in chitinase activity were detected in bean leaves

inoculated with alfalfa mosaic virus (de Tapia et al., 1986). Many fungal
pathogens contain [3-1,3-glucan, chitin, and chitosan as cell wall components. It

is conceivable that enzymes hydrolyzing these fungal cell wall components are
able to inhibit fungal growth.

Peroxidases catalyse a number of reactions that contribute to cell wall
strengthening (Hammerschmidt et al., 1982). These reactions include the
incorporation of phenolics into cell walls and lignification and suberization of plant
cell walls. Anionic peroxidase is located in the cell wall and catalyzes the
polymerization of cinnamyl alcohols into Iignin (Peng and Kuc, 1992).
Peroxidases may also have a role in defense by producing toxic levels of
hydrogen peroxide and phenolic free radicals that restrict pathogen growth
(Nicholson and Hammerschmidt, 1992).

The objectives of this work were to: 1. lntrogress genetic variability from
landrace and wild bean genotypes, selected as putative sources of resistance to
S. sclerotiorum, into adapted genetic background through inbred backcross
method;. 2. Test the reaction of the plant introductions, adapted parents and
inbred backcross lines to S. sclerotiorum inoculation; and 3. Evaluate the activity
of endogenous and induced defense related proteins in a group of bean lines

including wild, landrace and cultivated genotypes.

100

Materials and Methods

Development of inbred backcross lines

Plant Material

Seven populations were developed through the inbred backcross method
by crossing four plant introduction (Pl) accessions, including wild genotypes and
landraces, with three adapted dry bean cultivars (Table 1). The plant accessions
Pl 318695, Pl 313850, Pl 325685, and PI 313609 were originally selected as
potential sources for resistance to S. sclerotiorum based on positive greenhouse
tests conducted at Michigan State University, North Dakota State University and
University of Nebraska-Lincoln (Kolkman, 2000). The greenhouse tests included
the straw test (Petzoldt and Dickson, 1996), the leaf-agar plug assay (Steadman
et al., 1997) and the oxalic acid assay (Kolkman and Kelly, 2000). The
accessions Pl 325685, and PI 318695 are wild type accessions and are
considered potential sources of novel genetic diversity for resistance to S.
sclerotiorum. Accession Pl 313850 was also identified by Miklas et al. (1999), as
having putative physiological resistance to S. sclerotiorum, by screening a sub
sample of the core collection of P. vulgaris accessions using straw test. The core
sub sample represents the active USDA National Plant Germplasm System
Collection of 1,698 bean accessions from Central America and South America.

Due to photoperiod sensitivity, the crosses between the accessions and
adapted navy (Bunsi and Huron) and black (Tacana) bean cultivars were made
under short days (10 hours light I14 hours dark) in a growth chamber during

summer of 2000. Field screening for resistance to S. sclerotiorum of unadapted

101

germplasm is not practical, since morphological and phenological traits, such as
photoperiod sensitivity and climbing growth habit prevent the evaluation for field
resistance. The inbred backcross (IBC) method (Bliss, 1993) was conducted to
develop inbred backcross lines to transfer potential valuable genes from
unadapted genotypes in a commercial cultivar background where they can be
effectively evaluated. The lBC method differs from the recurrent backcross in that
no selection for donor traits is practiced during the backcrossing. To retain most
of the genetic variability from the exotic germplasm represented in the IBL,
different BC1F1 individuals need to be used as parents in the backcrossing. To
overcome the problem of lack of adaptation within the wild accessions, two
backcrosses to the recurrent parent were made in order to introgress quantitative
traits from the wild genotypes into the adapted cultivated background, and one
backcross was made to introgress quantitative traits from cultivated landraces Pl
313850 and PI 313609 into local cultivars. Therefore, the average percentage of
the landraces and wild genotypes in the resulting inbred backcross lines (IBL)
were 25 and 12.5%, respectively. During the development of IBL from crosses
involving wild genotypes, the objective was to use all BCrFr plants available to
obtain the BC2F1 generation to have represented the maximum genetic variability
derived from the wild parent. For example, more than 35 different BC1F1 plants
were utilized in crosses to generate the BC2F1 population. Lines were advanced
to BCng in the greenhouse through single seed descent and BC2F3;4 IBL were
tested in the field. For those populations derived from crosses with landraces,

involving only one backcross, five seeds from each individual BC1F1 plant were

102

bull
ant
glE

20

W

I01

bulked and the BCer generation was advanced in the field, during summer 2001
and individual plants were harvested. Lines were advanced to BC1F 4 in the
greenhouse by single seed descent and BC1F4;5 IBL were tested in the field in

2002.

Field evaluations

The BC2F3;4 and BC1F4:5 populations were planted in the field as single
rows in Saginaw, Michigan, in 2002. The goals of the experiment in Saginaw
were to evaluate the adaptation, agronomic traits of the 638 IBL derived from
seven different crosses (Table 3.1), and also increase seed of the selected
populations for future studies. Plot rows were 6m in length, with 0.5 m row
spacing. The plots were spaced planted with 20 seeds per 6m row. The
experiment consisted of single row plots per IBL without replication due to limited
seed availability. Standard agronomic practices for tillage, fertilization, and insect
and weed control were applied to ensure adequate plant growth and
development. No attempt was made to create favorable conditions for white mold
development and no disease was observed during the growing season.

The lBLs were visually rated for agronomic and phenological traits,
including: architecture and lodging, maturity and overall desirability. At
physiological maturity, plots were evaluated for architecture, using a 1 to 5 scale,
where 1 = fully upright, 3 = bush, and 5 = prostrate. Lodging was determined at
the same stage, based on a 1 to 5 scale, where 1 = no lodging, 3 = moderate

lodging, and 5 = excessive lodging. Maturity was rated comparing the IBL to the

103

adapted parents and was classified in early (E), normal (N) and late (L).
Desirability was rated using a scale from 1 to 9 where 1=undesirable plant type
and 9= highly desirable plant type. Both populations of Tacana, with PI 313850
and PI 318695 were selected along with the cross of Huron and PI 318695 for
further studies because they were agronomically adapted. Selected populations
were harvested at maturity, and then plots were individually pulled and threshed.
The 100 seed weight of individual IBL was determined after adjusting to 18 %

seed moisture content.

Greenhouse Straw Tests

Two Tacana derived IBL populations developed with Pl 313850 and PI
318695 were screened in two different experiments for physiological resistance
to S. sclerotiorum mold using the straw test (Petzoldt and Dickson, 1996). The
straw test was carried out with four replications for the Tacana/Pl 318695 IBL
population and three replications for the Tacana/Pl 313850 population. The
experiment was conducted in greenhouse maintained at 20-24°C. The
temperature control is crucial for a consistent an effective screening, since higher
temperatures affect the ability of the pathogen in developing the disease. A 15
cm diameter clay pot containing Bacto potting planting mix substrate (Michigan
Peat Co.), with 4 plants, represented a replicate that were randomized in
complete blocks. Plants were inoculated 28 days after planting. The growing tip
of the main stem was removed and a plastic straw containing a potato dextrose

agar (PDA) plug of three-day-old growing S. sclerotiorum mycelium was placed

104

over the intact cut stem. Only the edges of the growing fungus on the petri dish
plates were used to inoculate the plants. Eight days after inoculation, the white
mold reaction was scored from 1 to 9, suggested by Petzoldt and Dickson (1996)

and modified by Miklas et al., (1999).

Analysis of Plant Defense Related Proteins

Plant material and inoculation with S. sclerotiorum

Common bean cultivars Tacana, Raven, Bunsi, and PI 318695 and PI
313850 accessions were grown in 15 cm diameter pots containing Bacto potting
planting mix substrate (Michigan Peat Co.) in the greenhouse. The second fully
expanded trifoliate leaves of 28 day-old plants were excised and brought to the
laboratory. Leaves were harvested from three pots per cultivar, and each pot
contained two plants, thus providing six replicate samples per trifoliate per
cultivar. Half of the leaves (three trifoliates per cultivar) were inoculated with a 5
mm plug containing a 3-day-old S. sclerotiorum culture grown in PDA. The
inoculated trifoliates were kept in a petri dish containing wet filter paper
(Whatman # 1; 125 mm diameter). The other half of the trifoliate leaves were
kept in a petri dish with wet filter paper and a 1 cm PDA plug without the fungus
was placed into its surface (control). The size of the lesions was measured using
a digital caliper (Mitutoyo Corp.) at 48 hours after inoculation and the leaf tissue

surrounding the lesions was collected, weighed, frozen in liquid nitrogen and
stored at -80°C until required for use in enzymatic assays of peroxidase, B-1,3-

glucanase and chitinase. Twenty-day old plants of bean cultivars Tacana, Raven,

105

Bunsi, and accessions Pl 318695 and PI 313850 were also tested against S.
sclerotiorum by performing a straw test. The plants were grown under the same
time and conditions of those plants used to assay the defense related proteins.
Clay pots were used with three plants represented the experimental unit and ten

replications in a completely randomized design.

Sample preparation and protein extraction

The frozen leaf material was homogenized on ice in 0.08M potassium
acetate buffer, pH 5.0, containing 1 mM EDTA (3 ml.g‘1 fresh weight). Reduced
glutathione was added to the buffer immediately prior to homogenization to
achieve a final concentration of 5 mM. The extracts were centrifuged at 10,000 g
for 15 minutes and the supernatants were collected. Protein content in crude
extracts was determined using the Bio-Rad Protein Assay, with bovine serum
albumin (BSA) as standard. Samples of these supernatants with volumes that
varied depending on initial mass of fresh weight were frozen at -80°C for

subsequent use in the enzymes assays.

Spectrophotometric assays of peroxidase and B-1,3-glucanase

Soluble peroxidase activity was determined at room temperature by
measuring the appearance of pink/brown color resulting from guaiacol oxidation
in the presence of hydrogen peroxide, based on the method described by
Hammerschmidt et al (1982). The reaction mixture consisted of 50 pl guaiacol

0.02 M (Sigma), 0.5 ml 0.38 M H202 and 2.0 ml 0.2 M sodium phosphate buffer

106

pH 6.0 that were added to a 3.0 ml glass cuvette and the spectrophotometer was
set to zero at 470 nm. Suitable diluted plant extract (50 pl) was added and the
cuvette inverted to mix ingredients prior to measuring optical density at 470 nm
every 0.25 minute for 4 minutes. The slope of the optical density versus time was
used as a relative measure of peroxidase activity. Results were expressed as
AOD470.g fresh weight“1 min". Three replicates were used for each sample and

results averaged.

The B-1,3-glucanase assay was performed according to Dann and
Deverall (2000) which relies on the release rate of a measurable blue dye when
the dye-bound substrate is hydrolyzed by endo-1,3-B-glucanase. The substrate,

azurine-crosslinked pachyman (AZCL-Pachyman), was obtained from Megazyme

Pty. Ltd. (Wicklow, Republic of Ireland). The substrate hydrates in water but is
water insoluble. Hydrolysis by endo-1,3-B-glucanase releases water-soluble dyed

fragments and the rate of release can be directly related to the enzyme activity.
Samples of the crude extract preparation (0.1 ml) were added to eppendorf tubes
and 0.4 ml potassium acetate buffer 10 mM, pH 5.0 was added. A separate tube
containing 0.5 ml of the buffer was included as a substrate blank. After pre-
equilibration for 5 minutes at 30°C, the reaction was initiated by adding 0.1 ml of
AZCL-pachyman substrate (prepared at the rate of 0.1 g substrate per 3.0 ml of
10 mM potassium acetate buffer, pH 5.0) to each tube. The reaction was stopped
at 10 minutes by adding 0.7 ml of a 20% w/v Tris (Sigma Chemical Co, USA)
solution. The tubes were vortexed, left at room temperature for 5 minutes and

then centrifuged at 9,000 G for 2 minutes. The optical density of the supernatants

107

was measured at 595 nm in the spectrophotometer (Dann et al., 1996). B-1,3-

Glucanase activity was determined by using a standard curve prepared with
laminarinase (EC 3.2.1.6, Sigma) with amounts that varied from 1.25 to 20
units.ml". The enzyme activity was expressed as units.g fresh weight". Three

replicates were used for each sample and results averaged.

Assay for chitinase activity

Chitinase activity was determined for diluted concentrated extracts as
described by Trudel and Asselin (1989), modified by Velasquez and
Hammerschmidt (personal communication). A gel containing 40 ml sodium
phosphate buffer 0.01 M, pH 6.0 and 0.4 g agarose and glycol chitin (Sigma, 10
mg.ml") was poured in a petri dish plate (125 mm). Glycol chitin was used as
substrate and was prepared by acetylation of glycol chitosan ( Molano et al.,
1979). After the gel polymerized, small wells (2 mm diameter) were opened into
the gel and 50 pg of the samples protein extract were poured into the openings.
The gel was incubated for 4 hours at 37°C in water bath. After incubation, 50 ml
of Tris-HCI 0.5 M, pH 8.9 containing 0.01% w/v calcofluor white m2R (Sigma)
was added into the gel. This staining solution was kept for 10 minutes, and then
the gel was rinsed with distilled water. The petri dish plate was filled with distilled
water for overnight period of time. The gel was placed under UV light, where dark
lytic zones of chitinase activity could be visualized in a clear background. A
standard curve of chitinase activity was prepared using a serial dilution from 1 to

20 units.ml'1 of commercially available chitinase prepared from Streptomyces

108

griseus (Sigma). Pictures of the gels were taken under UV light, and the

diameters of the lytic zones with chitinase activity were measured using image
analysis software (Sigma Scan). Activity could be compared by the size of the
lytic zone. The enzyme activity was expressed as units.g fresh weight". Three

replicates were used for each sample and the results were averaged.

Native polyacrylamide gel electrophoresis (PAGE) assays for
peroxidase, B-1,3-glucanase and chitinase activity

Electrophoresis was carried out under non-denaturing conditions.
Polyacrylamide resolving gels (7.5%, 1.5 mm) were prepared by mixing 4.9 ml
deionized water, 2.5 ml 1.5 M Tris-HCI (pH 8.8), 2.5 ml 30% acrylamide/bis
(29:1 ), 50 pl 10% (w/v) ammonium persulfate. The solution was degassed for 15
minutes, 5 pl of TEMED was added, and the gel was poured. The stacking gel
(4.0%) was prepared by mixing 6.1 ml deionized water, 2.5 ml 0.5 M Tris-HCI
(pH 6.8), 1.3 ml 30% acrylamide/bis, and 50 pl of 10% (w/v) ammonium
persulfate. The solution was degassed for 15 minutes, then 10 pl of TEMED was
added and the gel was poured. The upper tank buffer was prepared by mixing
5.2 g Tris base, 3.5 g glycine and 1 liter of distilled water. The lower tank buffer
was prepared by mixing 14.5 g Tris base, 60 ml of 1 N HCI, and diluting to 1 liter
with distilled water (Laemli, 1970). For chitinase detection gels, 0.01% (w/v)
glycol chitin was added prior to polymerization of the gel. Equal amounts of

protein (30 pg) were loaded onto the gels, and a constant current of 60 mA.gel'1

109

was applied for 5 hours and the temperature set to 4°C. A Mini-Protean lll
electrophoresis system from Bio-Rad was used to run the gels.

After electrophoresis, acidic peroxidase isoenzymes were visualized by
soaking the gels in a staining solution that contained 200 ml of 0.1 M sodium
acetate buffer, pH 5.0; 2 ml of 3-amino-9-ethyl-carbazole (prepared as 25 mg/ml
of N,N-dimethylformamide), and 200 pl of 30% H202. After development of red-
brown bands, the gels were refrigerated in 7% acetic acid until scanned or
photographed (Graham et al., 1965).

Endo-1,3-B-glucanase isoenzymes were visualized by washing the
electrophorized gels with distilled water three times and preincubated in 0.05 M
potassium acetate buffer, pH 5.0, for 5 minutes with slow shaking. Then the gels
were incubated at 40°C for 30 minutes in a solution containing 75 ml 0.05 M
potassium acetate buffer, pH 5.0 and laminarin (Sigma) predissolved in distilled
water (1g/75 ml). The gels were washed three times with distilled water and
placed onto a glass tray containing 200 ml 1.0 M NaOH and 0.3 9 2,3,5-
triphenyltetrazolium chloride. The gels were kept in a boiling water bath until red
bands appeared. The gels were transferred to a solution containing 3% glycerol,
40% methanol, 10% acetic acid, and 47% water (v/v) until photographed (Pan et
aL,1989)

Chitinase isoenzymes were detected after incubating the gels for 2 hours
at 37°C with reciprocal shaking in 0.1 M sodium acetate buffer, pH 5.0 containing
1% (v/v) Triton X-100. After the incubation, the gels were stained with a solution

of 0.01% (w/v) calcofluor white m2R (Sigma) dissolved in 0.5 M Tris-HCI, pH 8.9

110

for 5 minutes. The dye solution was removed and the gels were kept in distilled
water for about one hour at room temperature. Lytic zones were visualized by
placing the gels under a UV light source, and the gels were photographed
(Trudel and Asselin, 1989).

Data from the five genotypes including Pl 313850, Pl 318695, Tacana,
Bunsi and Raven were analyzed for significance by one-way analysis of variance
using the SigmaStat software. The straw test conducted in the two IBL
populations was analyzed using PROC GLM (SAS, 1995) in a randomized
complete block design with three replications for Tacana/Pl 313850 and four

replications for Tacana/Pl 313695 IBL populations.

Results and Discussion

Development and evaluation of inbred backcross lines

Six hundred and thirty-eight inbred backcross progenies (BC2F3;4 and
BC1F4;5) were developed from seven different crosses (Table 3.1). The inbred
backcross lines may provide a unique opportunity for the evaluation of novel
sources for resistance to S. sclerotiorum in an adapted cultivated genetic
background. Field evaluation of landrace and wild types is not possible due to
problems of adaptation of wild bean germplasm lacking the domestication
syndrome traits (Koinange et al., 1996). Inbred lines can only be tested for field
resistance against S. sclerotiorum if they are fully adapted to the local
environmental conditions. Field screening for white mold resistance of unadapted

germplasm is ineffective, since morphological and phenological traits, such as

111

Te

59

ad

photoperiod sensitivity and climbing growth habit have undesired avoidance
mechanisms and are could be falsely confounded with physiological resistance.
The preliminary field test carried out in 2002, based on spaced plants and single
plots, was the first opportunity to evaluate the IBL in terms of agronomic traits
and adaptation to local conditions. The main objective was to identify the best
populations in terms of adaptation and agronomic traits for further studies to
detect enhanced levels of resistance to S. sclerotiorum among IBLs.
Tacana/Pl313850 and Tacana/Pl 318695 IBL populations were selected from
seven populations based on visually rated agronomic traits. Tacana was the
adapted parent in both selected populations. Tacana (Lopez-Salinas et al., 1997)
is an upright black bean that has more erect plant architecture compared to the
other cultivated parents, Bunsi and Huron (Kelly et al., 1994) were used as
parents to develop the other five backcross inbred populations. Lines within the
two selected populations that showed extreme late flowering and maturity were
discarded. Out of 97 and 116 IBL from Tacana/Pl313850 and Tacana/Pl 318695
populations, respectively, 75 and 89 IBL were chosen to be tested with the straw

test to detect resistance to S. sclerotiorum.

112

 

ow - - men. F0m Ndo m_nEo_oO Umum>E=O momm Fm

 

- oor - vumumom Pm 02me 25> mwommm
.. m P r 3 3“. Fom o. 5 Eon. 852230 omwm 5
nm mm m: X”“214.qu Om 02on 25> 38 5
$390 335995 mama E
8390 :05... wommozo Baum memos. cozmzmcow nomm o3 59.0 EmEm>an_ co_mmmoo<

 

.551 new $55 6:83 .2953 coon bu _m_o.oEEoo 695 new EEozokoBm .m 2 85369 5062293.“
9:93 53> wcofimooom E nmfimumca c8253 8308 E “382923 8:: $98.08 855 ho .mnEsz 6m 223.

113

Seed size was an important agronomic trait that was expected to be recovered
with two backcrosses in the population derived from the cross between Tacana
and the wild genotype Pl 318695. Seed size ranged from 13.4 to 26.1g per 100
seeds in 89 IBL compared to 3.5 and 21 .59 from the parents Pl 318695 and
Tacana, respectively (Table 3.2). Lodging and desirability scores indicate that
lines with acceptable commercial architecture were developed and can be
effectively tested in the field for resistance to S. sclerotiorum. On average the
lBLs resulting from two backcrosses have 12.5% of contribution from the wild
genotype. A third backcross was not considered necessary because it would
reduce the average contribution of the exotic germplasm to 6.25% among the
resulting IBLs. The goal was to retain as much genetic variability as possible
from the wild source within IBL lines while maintaining adaptation, agronomic and
seed size traits of the recurrent parent. The population derived from Tacana and
the landrace Pl 313850 was developed with one backcross, resulting in inbred
backcross lines with 25% of genetic contribution from the landrace genotype. A
landrace is already a cultivated genotype and therefore one backcross should be
sufficient to recover adapted lines with acceptable combination of agronomic and
seed size traits. Despite the gene pool differences between the Tacana (Middle
American) and PI 313850 (Andean) parents, the population was well adapted
and did not show the typical genetic incompatibility associated with gene pool
crosses. The desirability scores of IBL indicated they possessed adequate

adaptation and architecture (Table 3.3).

114

Table 3.2. Lodging, desirability coefficient, maturity, seed size and reaction to
straw test in BC2 IBL derived from the cross between Tacana and PI
318695, grown in Saginaw, MI, in 2002.

 

IBL code Lodging1 082 Days to Seed size Straw

Maturity3 (g.100seedfl Test4
8571 3 4 M 18.1 2.3
8540 4 4 E 19.0 2.5
8563 2 6 E 23.4 2.6
8553 3 5 M 20.2 2.7
8600 3 3 L 19.1 2.7
118-2 3 4 M 18.7 5.9
8501 4 3 E 16.2 6.3
8507 3 3 M 17.7 6.8
8599 3 4 M 18.4 7.1
8508 3 4 E 17.3 7.2
Average(89) 2.9 4.3 - 18.7 4.3

Range 2 — 4 3 — 7 - 13.4 — 26.1 2.3 — 7.2

Tacana 2.0 5.0 M 21.5 3.4
Pl-318695 - - - 3.55 4.5
Raven° 2.9 3.9 107 19.0 8.5
Bunsi° 3.5 4.0 109 22.8 6.6

 

1Lodging is rated in a 1 to 5 scale, where 1= no lodging, 3 = moderate lodging,
and 5 = excessive lodging.

2 Desirability score, the higher the score (from 1 to 9) the more desirable is the
line.

3 E=early, M= mid season and L=late, relative to days to maturity in Tacana.

4 Straw test is scored in a scale from 1 to 9 where 1=no symptoms and 9=
complete collapse of the plant.

5 Seeds from plants grown in the greenhouse.

6 Grown in Montcalm, MI, in 2002.

115

Pl 313850 could be an important source of resistance to white mold for future

breeding of Andean Kidney and Cranberry bean classes.

Greenhouse straw tests

Based on the reaction to S. sclerotiorum measured by the straw test, PI
313850 was the most resistant genotype, followed by Pl 318695 and Tacana
(Figure 3.1).The two populations derived from Tacana crossed with PI 318695
and PI 313850 were tested in relation to their reaction to S. sclerotiorum in the
greenhouse using the straw test. In both populations significant differences
(P<0.001) for reaction to S. sclerotiorum were detected among the IBL within
each population. Many lines with low ratings for straw test were identified (Table
3.2, Figure 3.2 and Table 3.3 and Figure 3.3), suggesting that based on the
reaction to S. sclerotiorum, putative resistance to white mold existed within both
populations. The heritability estimative for the reaction to S. sclerotiorum in both
populations were 0.65 and 0.59, suggesting that selection for reaction to straw

test can be performed effectively in a breeding program.

Evaluation of defense related proteins activity in detached leaf assays

The response of been genotypes Pl 318695, Pl 313850, Tacana, Bunsi
and Raven to S. sclerotiorum and activity of defense-related enzymes was
determined. The plant introductions and Tacana were used as parents to develop

two inbred backcross population selected in Saginaw in 2002.

116

Table 3.3. Lodging, desirability coefficient, maturity, seed size and reaction to
straw test in BC2 IBL derived from the cross between Tacana and
Pl313850, grown in Saginaw, MI, in 2002.

 

IBL code Lodging1 DS2 Days to Seed size Straw

Maturity3 (9.10036661 ) Test“
8607 4 3 M 25.8 1.3
8673 3 3 L 24.4 1.4
8601 2 3 M 27.1 1.8
8636 3 4 M 23.4 2.0
8604 2 4 M 23.2 2.0
8639 3 4 M 27.5 4.2
8690 2 5 M 23.9 4.3
8638 3 5 E 21.1 4.4
8695 3 3 L 28.6 5.1
8652 3 3 M 24.6 5.7
Average(76) 2.7 4.2 - 24.0 3.0

Range 2-4 2—6 - 18.3—31.9 1.3-5.7

Tacana 2 5 M 21.5 2.8
PI 313850 - - - 61 .o5 3.3
Raven° 2.9 3.9 107 19.0 8.5
Bunsi° 3.5 4.0 109 22.8 6.6

 

1 Scale from 1 to 5, where 1: no lodging and 5 = excessive lodging.

2 Desirability score, the higher the score (from 1 to 9) the more desirable is the
line.

3 E=early, M=mid season and L=late, relative to Tacana.

4 Scale from 1 to 9, where 1=no symptoms and 9: collapse of the plant.

5 Seeds from plants grown in the greenhouse.

6 Grown in Montcalm, MI, in 2002.

117

Bunsi and Raven were included as checks and they were chosen because they
are the parents of the mapping population discussed in chapter 1.

The straw test results observed in the Pl accessions and adapted bean
varieties (Figure 3.1) were supported by the lesion size measured in detached
leaves 48 hours post inoculation (Figure 3.4). Leaves of PI 313850 inoculated
with S. sclerotiorum developed smaller lesions compared to all the other
genotypes. Bunsi was considered a susceptible genotype based on the reaction
to S. sclerotiorum with the detached leaf (Steadman et al., 1997) and straw
(Pedzoldt and Dickson, 1996) tests. Bunsi has field resistance and is known as a
source of resistance to S. sclerotiorum. Moderate correlation between field
resistance to S. sclerotiorum and straw test rating has been reported (Hall et al.,
1999; Myers et al., 1999; Miklas et al., 2001). However, the resistant genotypes
Bunsi and G 122, normally scored greater than five in greenhouse straw tests. It
is likely that the straw test measures only a single component (physiological
resistance of stem internode and nodal tissue to pathogen invasion) of a complex

of factors involved in physiological resistance (Miklas et al., 1999).
The activity of defense related enzymes peroxidase, B-1,3-glucanase and

chitinase was evaluated 48 hours post inoculation. Pl 313850 showed the
highest peroxidase activity after infection, compared to the other four genotypes
tested (Figure 3.5). Peroxidase activity was also higher in Pl 313850 without the

presence of the pathogen, suggesting that peroxidase was constitutively
expressed in higher levels in this genotype. Tacana showed the highest B-1,3-

glucanase activity with and without inoculation (Figure 3.6).

118

STRAW TEST RATING

 

 

Bunsi Raven Pl 313850 Pl 318695 Tacana

Figure 3.1. Reaction of adapted and exotic bean genotypes to inoculation with S.
sclerotiorum using the straw test. Data represent mean of 10
replicates and the bars indicate standard deviation for each mean.
P<0.001 for bean genotypes. The straw test was rated in a scale
from 1 to 9 where 1: no symptoms and 9: total plant collapse.

119

Tacana

3O - ¢ P1 318695

1

     

25-

20A

15-

Frequency

10-

 

 

 

1 2 3 4 5 6 7 8 9
Straw test rating

Figure 3.2. Reaction of 89 Tacana/Pl 318695 inbred backcross lines (IBL) to
inoculation with S. sclerotiorum using the straw test. P < 0.001 for
IBL. The straw test was rated in a scale from 1 to 9 where 1: no
symptoms and 9: total plant collapse.

120

'Tacana

1

P1313850

l

  

30-

 

25-

20-

15d

Frequency

10-

 

 

1 2 3 4- 5 6 '7 8 9

Straw test rating
Figure 3.3. Reaction of 76 Tacana/Pl 313850 inbred backcross lines (IBL) to
inoculation with S. sclerotiorum using the straw test. P < 0.001 for

IBL. The straw test was rated in a scale from 1 to 9 where 1: no
symptoms and 9: total plant collapse.

121

25—

20—

Diameter of lesion (mm)

 

 

Bunsi Raven Pl 313850 Pl 318695 Tacana

Figure 3.4. Diameter of lesion in detached leaves of five common bean
accessions at 48 hours after inoculation with S. sclerotiorum. Data
represent mean of 9 replicates, and the bars indicate standard
deviation. P=0.003 for bean genotypes.

122

- Control
[ZZZ Inoculated with Sclerotinia scleroa'orum

35-

25-l

20—

 

15—

10—

Peroxrdase Actrv Ity
(AODW .min'1.g FW’)

   

 

 

 

 

 

 

4

01

  

Figure 3.5. Peroxidase activity in detached leaves of five common bean
accessions at 48 hours after inoculation with PDA plug (control), and
PDA plug with S. sclerotiorum mycelia. Data represent mean of 9
replicates, and the bars indicate standard deviation. P=0.02 for
genotypes when inoculated and non significant for genotypes without
inoculation (control).

123

280 _ - COWOI
r221 Inoculated with Sclerotinia sclerotiorum

 

B-L3-Gbcalase Activity
(uunits.g PW")

 

 

 

 

 

 

 

Bunsi Raven Pl 318695 Pl 313850 Tacana

Figure 3.6. B-1,3-Glucanase activity in detached leaves of five common bean
accessions 48 hours after inoculation with PDA plug (control), and
PDA plug with S. sclerotiorum mycelia. Data represent mean of 9
replicates, and the bars indicate standard deviation. P=0.01 for
genotypes when inoculated and P=0.03 for genotypes without
inoculation (control).

124

The activity of the defense related enzyme chitinase was not increased when the
leaves were inoculated with S. sclerotiorum and the five genotypes showed no

significant differences in its activity (Figure 3.7).
Polyacrylamide gel electrophoresis (PAGE) for peroxidase and B-1 ,3-

glucanase activity supported the previous data from spectrophotometric assay of
enzymatic activity. Pl 313850 showed the high peroxidase activity, but similar to
Bunsi, after being inoculated with S. sclerotiorum (Figure 3.8). The peroxidase
activity observed through PAGE is only associated with the acid isoforms and

represent only part of the total peroxidase activity measured in the
spectrophotometric assay. Tacana had the highest B-1,3-glucanase activity

(Figure 3.9) and PAGE gel for chitinase activity confirmed that the five genotypes
tested were not significantly different (Figure 3.10).

The different levels of enzymatic activity of defense related proteins
detected in this study indicated that genetic variability for plant defense
mechanisms exist among the five bean genotypes. Combining high activity of
different defense related enzymes could result in higher levels of non-specific
plant defenses. This hypothesis could be tested by producing recombinants from
crosses between parents such as Tacana and PI 313850 and combining high
activity of different defense related enzymes from different sources. The inbred
backcross lines with putative resistance to S. sclerotiorum mold might represent
a unique opportunity to introduce novel sources of resistance into different bean

commercial classes.

125

Chitinase activity
(units.g fresh weight")

Figure 3.7.

 

- Control
[E Inoculated with Sclerotinia sclerotiorum

 

 

A

 

/

 

 

 

 

 

Bunsi Raven PI 313850

 

 

 

Pl 318695

 

 

Tacan a

Chitinase activity in leaves of five common bean accessions 48 hours
after inoculation with PDA plug (control), and PDA plug with S.
sclerotiorum mycelia. Data represent mean of 9 replicates, and the
bars indicate standard deviation. Differences were not significant
among genotypes at P=0.05, with or without inoculation.

 

C = Control
I = inoculated with
S. sclerotiorum

 

 

 

P1318695 P1313850 Bunsi Tacana Raven

Figure 3.8. Enzymatic activity of acidic peroxidase isoforms in detached leaves of
five common bean accessions 48 hours after inoculation with PDA
plug (control), and PDA plug with S. sclerotiorum mycelia.

127

 

C = Control
I = inoculated with
S. sclerotiorum

 

 

 

P1318695 P1313850 Bunsi Tacana Raven

Figure 3.9. Glucanase activity in detached leaves of five common bean
accessions 48 hours after inoculation with PDA plug (control),
and PDA plug with S. sclerotiorum mycelia.

128

 

C = Control
I = inoculated with
S. sclerotiorum

 

 

 

    

          

PI31869SHP‘13-13850 B'Ensi 'Tacana Raven

Figure 3.10. Chitinase activity in detached leaves of bean 48 hours after
inoculation with PDA plug (control), and PDA plug with S.
sclerotiorum mycelia.

129

Considering that Bunsi does not display high levels of resistance when
inoculated using the straw test, the inbred lines identified using the straw test
may have different mechanisms of defense other than those present in Bunsi.
Tacana displayed higher levels of resistance when inoculated with S.
sclerotiorum, similar to Pl 318695, but lower than Pl 313850. The differences in
activity of the defense related enzymes in Tacana, Pl 313850 and Pl318695, and
the range of reaction to S. sclerotiorum observed in the populations
Tacana/Pl313850 and Tacana/318695, suggest that different mechanisms of
defense to white mold could be present in those genotypes. To identify and
understand the defense mechanisms involved in resistance would open the
possibility of developing new strategies to select new sources of resistance to S.
sclerotiorum. The study of other plant defense-related genes and the putative
association with physiological resistance to white mold should be considered.
Polygalacturonase-inhibiting protein (Pgip) is particularly noteworthy, since
polygalacturonase is generated by the pathogen during S. sclerotiorum infection
(Favaron et al., 1994) and QTL for resistance to white mold were detected (see
Chapter 1) in the same region on linkage 82 where Pgip was mapped (Kolkman
and Kelly, 2003). The new sources of resistance could be combined with those
currently available and result in enhanced levels of resistance to S. sclerotiorum

in common been.

130

References

Acosta-Gallegos, J., C. Quintero, J.Vargas, O. Toro, J.Tohme, and C. Cardona.
1998. A new variant of arcelin in wild common bean, Phaseolus vulgaris
L., from southern Mexico. Genet. Resources and Crop Evol. 45:235-242.

Bhandal, LS, and JD. Paxton. 1991. Phytoalexins biosynthesis induced by the
fungal glucan polytran l in soybean, pea, and sweet pepper tissue. Journal
of Agriculture and Food Chemistry 39:2156-2157.

Bliss, FA. 1993. Breeding common been for improved biological nitrogen
fixation. Plant and Soil 152:71-79.

Boller, T., A. Gehri, F. Mauch., and U. Vogeli. 1983. Chitinase in bean leaves:
induction by ethylene, purification, and possible function. Planta 157:22-
31.

Brown, A.H.D. 1989. Core collection: a practical approach to genetic resources
management. Genome 31:818-824.

Cardona, C., J. Kornegay, C.E. Posso, F. Morales, and H. Ramirez. 1990.
Comparative value of four arcelin variants in the development of dry bean
lines resistant to the Mexican bean weevil. Entomol. Expt. Appl. 56:197-
206.

Cardona, C., and J. Kornegay. 1999. Bean germplasm resource for insect
resistance . pp. 85-99. In: S.L. Clement and SS. Quisenberry (ed.) Global
plant genetic resources for insect-resistance. CRC Press, Boca Raton, FL.

Concibido, C., B. La Vallee, P. Mclaird, N. Pineda, J. Meyer, L. Hummel, J.
Yang, K. Wu, and X. Delannay. 2003. lntrogression of a quantitative trait
locus for yield from Glycine soja into commercial soybean cultivars. Theor.
Appl. Genet. 106:575-582.

Dann, E.K., P. Mewly, J.P. Métraux, and B.J. Deverall.1996. The effect of
pathogen inoculation or chemical treatment on activities of chitinase and
B-1,3-glucanase and accumulation of salicylic acid in leaves of green

bean, Phaseolus vulgaris L. Physiological and molecular Plant Pathology
49:307-319.

Dann, E.K. and B.J. Deverall. 2000. Activation of systemic disease resistance in

pea by an avirulent bacterium or a benzothiadiazole, but not by a fungal
leaf spot pathogen. Plant Pathology 49:324-332.

131

De Tapia, M., P. Bergmann, P. Awade, and G. Burkard. 1986. Analysis of acid
extractable bean leaf proteins induced by mercuric chloride treatment and
alfalfa mosaic virus infection. Partial purification and characterization.
Plant Sci 45:167-177.

Favaron, F., R. D’Ovidio, E. Porceddu, and P. Alghisi. 1994. Purification and
molecular characterization of soybean polygalacturonase-inhibiting
protein. Planta 195:80-87.

Freyre, R., P.W. Skroch, V. Gelfroy, A.F. Adam-Blondon, A. Shirmohamadali,
W.C. Johnson, V. Llaca, R.O Nodari, P.A Pereira, S.M. Tsai, J. Tohme, M.
Dron, J. Nienhuis, C.E. Vallejos, and P. Gepts. 1998. Towards and
integrated linkage map of common bean. 4. Development of a core
linkage map and alignment of RFLP maps. Theor. Appl. Genet. 97:847-
856.

Gooday, G.W. 1994. Cell walls. In: N.A. R. Gow and GM. Gadd (eds). The
growing fungus. Chapman and Hall: London. P. 43-62.

Graham, R.C., U. Lundholm, and M.J. Karnovsky. 1965. Cytochemical
determination of peroxidase activity with 3-amino-9-ethylcarbazole.
Journal of Histochemistry and Cytochemistry 13:150-152.

Hall, R., LG. Phillips, and D. Mooij. 1999. Field and greenhouse evaluation of the
straw test for resistance of dry bean to white mold. Ann. Rep. Bean
Improv. Coop. 42:55-56.

Hammerschmidt, R., E. M. Nuckles, and J. Kac. 1982. Association of enhanced
peroxidase activity with induced systemic resistance of cucumber to
Colletotrichum Iagenarium. Physiological Plant Pathology 20:73-82.

Hartman, J.B., and DA. St. Clair. 1998. Variation for insect resistance and
horticultural traits in tomato inbred backcross populations derived from
Lycopersicon pennelli. Crop Sci. 38:14501-1508.

Huang, J-S. 2001. Plant pathogenesis and resistance: biochemistry and
physiology of plant-microbe interactions.

Hunter, J.E., M.H. Dickson, and A. Cigna. 1981. Limited-term inoculation: a
method to screen bean plants for partial resistance to white mold. Plant
Disease 65:414-417.

Ininda, J., W.R. Fehr, S.R Cianzio, and SR. Schnebly. 1996. Genetic gain in

soybean populations with different percentages of plant introduction
parentage. Crop Sci. 36:1470-1472.

132

Kelly, J.D., G.L. Hosfield, G.V. Varner, M.A. Uebersax, M.E. Brothers, and J.
Taylor. 1994. Registration of ‘Huron’ navy bean. Crop Sci. 34:1408.

Koinange, E. M.K., S.P. Singh, and P. Gepts. 1996. Genetic control of the
domestication syndrome in common bean. Crop Sci. 36:1037-1045.

Kolkman J.M. 2000. Genetic dissection of resistance to white mold (Sclerotinia
sclerotiorum) in common bean (Phaseolus vulgaris).Ph.D. Dissertation,
pp. 124. Michigan State University. East Lansing, MI.

Kolkman J.M, and JD Kelly. 2000. An indirect test using oxalate to determine
physiological resistance to white mold in common bean. Crop Sci. 40:281-
285.

Kolkman, J.M. and Kelly, JD. 2003. QTL conferring resistance and avoidance to
white mold in common bean. Crop Sci. 43:539-548.

Laemmli, UK. 1970. Cleavage of structural proteins during the assembly of the
head of bacteriophage T4. Nature 227: 680-685.

Leubner-Metzger, G., and F. Meins.. 1999. Functions and regulation of plant [3-
1,3-glucananes (PR-2). In: S. K. Datta and S. Muthukrishnan (eds).
Pathogenesis-related proteins in plants. CRS Press: New York. p 49-76.

Lopez-Salinas, E., J.A. Acosta-Gallegos, E.N. Becerra-Leor, G. Frayre-Vazques,
S.H.Orozco, and SE. Beebe. Registration of “Negro Tacana” common
bean.

McClean, P.E., J.R. Myers, and J.J. Hammond. 1993. Coefficient of parentage
and cluster-analysis of Nort-American dry bean cultivars. Crop Sci.
33:190-197.

Miklas, P.N., K.F. Grafton, and BB. Nelson. 19923. Screening for partial
physiological resistance to white mold in dry bean using excised stems. J.
Amer. Soc. Hort. Sci. 117:321-327.

Miklas, P.N., K.F. Grafton, G.A. Secor, and PE. Mcclean. 1992b. Use of
pathogen filtrate to differentiate physiological resistance of dry bean to
white mold disease. Crop Sci. 32:310-312.

Miklas, P.N., K.F. Grafton, and P. E. McClean. 1993. Estimating phenylalanine

ammonia-lyase activity in common beans inoculated with Sclerotinia
sclerotiorum. HortScience 28:937-938.

133

Miklas, P.N., R. Delorme, R. Hannan, and M.H. Dickson. 1999. Using a
subsample of the core collection to identify new sources of resistance to
white mold in common bean. Crop Sci. 39:569-573.

Miklas PN, R Delorme, WC Johnson, and P Gepts. 2001. QTL conditioning
physiological resistance and avoidance to white mold in dry bean. Crop
Sci. 41 :309-315.

Molano, J., l. Polacheck, A. Duran, and E. Cabib. 1979. An endochitinase from
wheat germ. Activity on nascent and preformed chitin. The Journal of
Biological Chemistry 254: 4901-4907.

Myers, J.R., B. Gilmore, and D. Kean. 1999. Correlation between the field and
straw test for white mold resistance in common bean. Annu. Rep. Bean
Improv. Coop. 42:57-58.

Nicholson, R.L., and R. Hammerschmidt. 1992. Phenolic compounds and their
role in disease resistance. Annual Review of Phytopathology 30:369-389.

Osborn, T.C., D.C. Alexander, S.S.M. Sun, C. Cardona, and FA. Bliss. 1988.
Insecticidal activity and lectin homology of arcelin seed protein. Science
240:207-210.

Pan, S.Q., X.S. Ye, and J. Kim. 1989. Direct detection of B-1,3-glucanase
isozymes on polyacrylamide electrophoresis and isoelectrofocusing gels.
Analytical Biochemistry 182:136- 140.

Peng M., and J. Kim. 1992. Peroxidase-generated hydrogen peroxide as a
source of antifungal activity in vitro and on tobacco leaf disks.
Phytopathology 82:696-699.

Petzoldt, R., and M.H. Dickson. 1996. Straw test for resistance to white mold.
Annu. Rep. Bean Improv. Coop. 39:142-143.

SAS Institute. 1995. The SAS system for Windows. Release 6.12. SAS lnst.,
Cary, NC.

Schwartz, H.F., D.H. Casciano, J.A. Asenga, and DR. Wood. 1987. Field
measurement of white mold effects upon dry beans with genetic
resistance or upright plant architecture. Crop Sci. 27:699-702.

Singh, PS. 1999. Integrated Genetic Improvement. lnzCommon bean

improvement in the twenty-first century. S.P. Singh (ed). Kluver Academic
Publishers, Dordrecht, The Netherlands.

134

Skinner, D.Z., G.R. Bauchan, G. Auricht, and S. Hughes. 1999. A method for the
efficient management and utilization of large germplasm collections. Crop
Sci. 39:1237-1242.

Steadman. JR. 1983. White mold- a serious yield-limiting disease of bean. Plant
Disease 67:346-50.

Steadman, J.R., K. Powers, and B. Higgins. 1997. Screening common bean or
white mold resistance using detached leaves. Annu. Rep.Bean Improv.
Coop. 402140-141.

Sutton, DC, and B.J. Deverall. 1984. Phytoalexin accumulation during infection
of bean and soybean by ascospores and mycelium of Sclerotinia
sclerotiorum. Plant Pathology 33:377-383.

Tanksley, SD, and JC. Nelson. 1996. Advanced backcross QTL analysis: a
method for the simultaneous discovery and transfer of valuable QTL form
unadapted germplasm into elite breeding lines. Theor. Appl. Genet.
92:191-203.

Tanksley, SD, and SR. McCouch. 1997. Seed banks and molecular maps:
Unlocking genetic potential from the wild. Science 277:1063-1066.

Thompson, J., and R.L. Nelson. 1998. Utilization of diverse germplasm for
soybean yield improvement. Crop Sci. 38:1362-1368.

Thorne, J.C., and W.R. Fehr. 1970. Exotic germplasm for yield improvement in 2-
way and 3-way soybean crosses. Crop Sci. 10:677-678.

Trudel, J., and A. Asselin. 1989. Detection of chitinase activity after
polyacrylamide gel electrophoresis. Analytical Biochemistry 178: 362-366.

Van Loon, LC, 1999. Occurrence and properties of plant pathogenesis-related
proteins. In: S. K. Datta and S. Muthukrishnan (eds). Pathogenesis-
related proteins in plants. CRS Press: New York. p 1-19.

Webster, D., and JD Kelly. 2000. Comentary on germplasm enhancement and
cultivar development. pp. 55-59. In: S. Singh (ed). Bean research,
production and utilization. Proc. of the Idaho Bean Workshop, August 3
and 4, 2000. Twin Falls, Idaho.

135

GENERAL DISCUSSION AND FUTURE RESEARCH

General Discussion

Complete field resistance to white mold does not exist in bean, but
genotypes with partial resistance that includes physiological resistance and/or
morphological avoidance mechanisms are available. The goal of the research
reported here was to enhance the resistance of current bean varieties, develop
methods to more efficiently transfer resistance to a wide range of seed types and
identify new and novel sources of resistance to white mold in exotic bean
germplasm.

The first objective was to identify QTL associated with resistance to white
mold using a multi-trait bulked segregant analysis approach. The RIL population
used in this study was derived from a cross between Bunsi and Raven. Bunsi is
recognized as a source of resistance to white mold and Raven, despite its upright
plant architecture that suggests avoidance to white mold, is highly susceptible
and represents a low level of physiological resistance. A QTL for DSI was
detected on linkage group B2, confirming the QTL detected previously by
Kolkman and Kelly (2003) in a different RIL population. The data support the
existence of a QTL for resistance to white mold on B2, in a region of the bean
genome where several host response genes related to general defense
mechanisms that may play an important role in disease resistance reside. PvPR-
2, a low molecular weight acidic protein induced during fungal elicitation (Walter

et al., 1990), a polygalacturonase-inhibiting protein (Toubart et al., 1992), and

136

chalcone synthase, (Ryder et al., 1987), an isoflavonoid-derived phytoalexin
involved in host defense are all located in this region on B2. Plant defense-
related genes may be triggered as a general resistance response to S.
sclerotiorum infection. QTL associated with resistance to common bacterial blight
(Nodari et al., 1993), and root rot (Schneider et al., 2001) were also identified in
this region of B2.

A novel QTL associated with resistance to white mold detected in this
study was mapped to BS. A third QTL for DSI mapped to B7 in the same region
where Kolkman and Kelly (2003) identified a major QTL for resistance to white
mold and other associated traits, including days to flowering and maturity,
architecture, lodging, seed size and yield. The genomic region on B7 where the
QTL was detected corresponded to the linkage B7-b in this study. Miklas et al.,
(2001) and Park et al., (2001) also detected QTL for resistance to white mold on
B7. However, those QTL were mapped close to the Phaseolin gene, a different
region on B7, that corresponds to linkage B7-a in this study, suggesting that the
QTL identified for DSI in Bunsi/Raven RIL population is different from the QTL
detected on B7 in Andean genotypes G122 and PC-50 by Miklas et al., (2001)
and Park et al., (2001).

The fact that different QTL were identified in 2001 and 2002 suggested
that depending on the environmental conditions, different mechanisms were
playing a role in resistance to white mold. The environmental conditions in 2001
were more favorable for white mold and plant structure was not detected as an

important factor in disease development. Plant architecture played an important

137

role in avoiding the disease in 2002 where high temperatures and low
precipitation prevailed. The results suggested that under conditions of high
disease pressure, avoidance mechanisms may not be sufficient to restrict
disease development and physiological resistance would play an increasingly
important role in field resistance. Bunsi and Raven are contrasting genotypes for
architecture, and the effect of upright plant type on resistance to white mold was
evaluated in different years with different environmental conditions. Associations
between the different agronomic traits and DSI identified in this study, indicate
that plant architecture and resistance to lodging are traits that need to be
considered when selecting for resistance to white mold. An upright structure and
medium plant width that ensures a porous canopy should represent ideal plant
architecture with avoidance to white mold. However, avoidance mechanisms
have to be combined with physiological resistance to obtain high levels of
resistance to white mold.

Resistance to white mold is a quantitative trait, problematic to measure,
significantly influenced by other traits and environment. Complex traits exhibiting
low heritability make MAS an important approach to enhance resistance. The
identification of different QTL is crucial because they may represent an
opportunity to combine through MAS, different plant strategies for resistance to
white mold that might result in an enhanced and stable resistance across
environments. MAS provides an opportunity to enhance resistance by combining
QTL regardless of function in resistance to white mold. However, a better

understanding of the defense mechanisms directly related to resistance to white

138

mold is needed. Since the contribution of each QTL to a quantitative trait is small,
pyramiding different QTL is necessary to significantly improve the genetic gain
for a complexly inherited trait. The potential of using MAS for resistance to white
mold was confirmed in a F4 derived RIL population originated from a cross
between Bunsi and Midland.

A strategy proposed to improve resistance to white mold in common bean
would combine phenotypic selection for agronomic traits with molecular markers
linked to QTL associated with physiological resistance to S. sclerotiorum.
Considering that most of the agronomic traits have a high heritability, selection
could effectively be conducted in early generations and in preliminary yield trials.
Environments without favorable conditions for white mold could be used to avoid
problems regarding cause/effect that frequently occur with traits such as
architecture and lodging that can significantly affect or be influenced by disease
occurrence. Scoring agronomic traits in the absence of disease would allow a
more efficient selection for these traits that could be combined with molecular
markers for physiological resistance. Combining QTL for white mold resistance
on different linkage groups with agronomic traits would require large populations
in order to select lines that display the desired combination of markers and
agronomic traits. Selection for yield under high white mold pressure might be an
effective final step to improve resistance to white mold in agronomically
acceptable genotypes. Selection only based on disease severity or incidence can
result in lines with limited potential use in breeding due to combinations of

undesirable agronomic traits. Pyramiding multiple QTL with individual minor

139

effects is a challenging task, but the tools are available and MAS should play a
more important role in plant breeding for quantitative traits in the future.

The third aspect of this study was to introgress putative sources of white
mold resistance into cultivated bean varieties. The plant introductions were
selected based on different greenhouse tests used to detect physiological
resistance to S. sclerotiorum. Bean has a wide genetic diversity among
landraces, wild types and species that is an invaluable source for important traits
including resistance to white mold. Field screening for white mold resistance of
unadapted germplasm is ineffective, since morphological and phenological traits,
such as photoperiod sensitivity and climbing growth habit prevent field evaluation
and could be confounded with physiological resistance. The putative sources of
resistance to white mold have to be introduced into adapted genetic background
for testing under field conditions. From seven populations developed using inbred
backcross method, two populations, Tacana/Pl 313850 and Tacana/Pl 318695,
were selected in the field for further studies based on acceptable adaptation and
plant architecture.

In both selected populations significant differences for reaction to S.
sclerotiorum were detected among the IBL within each population. Many lines
with low ratings for the straw test were identified, suggesting that based on the
reaction to S. sclerotiorum, both populations have putative resistance to white
mold. Based on desirability score, architecture, lodging and seed size, the inbred

backcross method was effective to recover adaptation and acceptable agronomic

140

traits. The IBL can be tested in the field under white mold pressure to identify
sources of resistance.

The different levels of enzymatic activity of defense related proteins
detected in this study suggested that genetic variability for plant defense
mechanisms may exist among the plant introductions and bean varieties tested.
High activity of different defense related enzymes could result in higher levels of
non specific plant defenses which could influence the reaction to S. sclerotiorum.
This hypothesis could be tested by measuring the activity of defense related

enzymes in the Tacana/Pl 313850 IBL selected in the straw test, since Tacana
and PI 313850 expressed high B-1,3-glucanase and peroxidase activity,

respectively, when inoculated with S. sclerotiorum. The interactions of different
defense related enzymes with white mold resistance remains to be elucidated.
The inbred backcross lines with putative resistance to white mold might
represent an unique opportunity to introduce novel sources of resistance into
different bean commercial classes. Considering that Bunsi does not display high
levels of resistance when inoculated with S. sclerotiorum in the straw test, the
resistant inbred lines identified using the straw test may have different
mechanisms of defense other than those present in Bunsi. To identify and
understand the defense mechanisms involved in resistance would provide the
possibility of developing new strategies to select novel sources of resistance to
S. sclerotiorum. The novel sources of resistance could be combined with those
currently available sources and result in enhanced levels of resistance to white

mold in common bean.

141

Future Research

White mold is a complex disease and a combination of approaches is
needed to improve resistance to S. sclerotiorum in common bean. Suggestions

of further studies that could follow the current research are discussed.

1. Select the recombinant inbred lines derived from Bunsi/Raven cross with the
best marker combination associated with resistance to S. sclerotiorum. Test
those lines in the field in different environments to confirm the efficiency of the
QTL identified in this study. RIL lacking the markers linked to QTL for

resistance could be used as controls.

2. Convert AFLP markers linked to QTL for resistance to white mold into SCAR
markers to increase the throughput of MAS selection and help reduce
screening costs. The regions where the QTL mapped were relatively
saturated with markers. The selection of an adequate AFLP marker size (>

200bp) to be converted is therefore possible.

3. Test the IBL derived from Tacana/Pl 313850 and Tacana/Pl 318695, selected
in the straw tests and based on agronomic traits, for the reaction to S.
sclerotiorum under high disease pressure in field conditions. Indirect tests to
detect physiological resistance including inoculation of detached leaves, straw

test, and resistance to oxalate should also be conducted. Activity of defense

142

 

related enzymes should be assayed, especially B-1,3-glucanase and

peroxidase activity in lBLs of Tacana/PI 313850 and correlations with

resistance could be determined.

4. Select from the Tacana/Pl 313850 and Tacana/PI 318695 populations, IBL
that combine resistance to white mold and desired agronomic traits, for QTL
detection studies. Developing mapping populations by crossing with a white
mold susceptible genotype but with an acceptable upright plant structure,
such as Raven, should be an effective strategy to detect QTL associated with

physiological resistance, since both parents would be upright.

5. Identify QTL associated to resistance to S. sclerotiorum in genetic sources in
both gene pools including landrace and wild accessions. Bunsi is the only
reported Middle American bean which has QTL detected for white mold.
Pyramiding different QTL from different genetic sources could be an effective
strategy to improve resistance to white mold in bean. QTL identified in Pl
313850 could be combined with QTL already detected in large seeded
Andean genotypes G 122 and PC 50 and be introduced into Andean
Cranberry and Kidney bean classes. Transfer QTL between gene pools has
to be considered but may show limitation due the incompatibility in crosses
involving genotypes from Andean and Middle American gene pools. The
cross between Tacana and PI 313850 resulted in IBL with acceptable

adaptation and agronomic traits suggesting the inbred backcross method may

143

be effective to transfer QTL for resistance to white mold between the Andean
and Middle American gene pools. To identify adapted varieties, landraces or
wild types with resistance to white mold is an important contribution to bean
breeding programs but the incorporation of new sources of resistance to S.
Sclerotiorum into new bean varieties is essential. However, plant
introductions already identified as putative sources of resistance to white
mold, such as those used in this study, should be first exploited. In a medium
and long term, more screening of core collections should be conducted and
maybe based on better understanding of physiological resistance. Searching
for resistance in related species is an alternative, but genetic diversity in bean

primary gene pool should be exploited first.

6. Study polygalacturonase-inhibiting protein (Pgip) and putative association with
physiological resistance to white mold should be considered. Pgip is
particularly noteworthy since polygalacturonase is generated by the pathogen
during S. sclerotiorum infection and QTL for resistance to white mold were
detection in the same region on B2 where Pgip was mapped. The limitation is
to develop a high throughput protocol to measure Pgip in beans, since most
of the methods currently available are based on measuring mRNA derived
from Pgip gene. The detection of Pgip by use of polyclonal antibodies could
be an alternative methodology that would allow screening of large

populations.

144

7. Transform bean plants with oxalate oxidase germin gene from wheat might be
an effective approach to increase resistance, since oxalic acid is the primary
pathogenicity factor of S. sclerotiorum. Results obtained with soybean
transformed with oxalate oxidase gene were promising (Donaldson et al.,
2001). However technical limitations in bean transformation and marketing

genetically modified organisms (GMO) are issues that have to be considered

in this approach.

145

References

Donaldson, P.A., T. Anderson, B.G. Lane, A.L. Davidson, and DH. Simonds.
2001. Soybean plants expressing an active oligomeric oxalate oxidase
from the wheat gf-2.8 (germin) gene are resistant to the oxalate-secreting
pathogen Sclerotinia sclerotiorum. Physiological and Molecular Plant
Pathology 59:297-307.

Kolkman, J.M. and JD. Kelly. 2003. QTL conferring resistance and avoidance to
white mold in common bean. Crop Sci. 43:539-548.

Miklas P.N., R. Delorme, W.C. Johnson, and P. Gepts. 2001. QTL conditioning
physiological resistance and avoidance to white mold in dry bean. Crop
Sci. 41 :309-315.

Nodari, R.O., S.M. Tsai, P. Guzman, R.L. Gilbertson, and P. Gepts. 1993.
Towards an integrated linkage map of common bean. llI. Mapping genetic
factors controlling host-bacteria interactions. Genetics 134:341-350.

Park 80., DP. Coyne, J.R. Steadman, and PW. Shroch. 2001. Mapping of QTL
for resistance to white mold disease in common bean. Crop Sci. 41 :1253-
1262.

Ryder, T.B., S.A. Hedrick, J.N. Bell, X. Liang, S.D. Clouse, and OJ. Lamb. 1987.
Organization and differential activation of a gene family encoding the plant
defense enzyme chalcone synthase in Phaseolus vulgaris. Mol. Gen.
Genet. 210:219-233.

Schneider, K.A., K.F. Grafton, and JD. Kelly. 2001. QTL analysis of resistance to
Fusarium root rot in bean. Crop Sci. 41 :535-542.

Toubart, P., A Desiderio, G. Salvi, F. Cervone, L. Daroda, and G. De Lorenzo.
1992. Cloning and characterization of the gene encoding the
endopolygalacturonase-inhibiting protein PGIP) of Phaseolus vulgaris L.
Plant J. 22367-373.

Walter, M.H., J. Liu, C. Grand, C.J. Lamb, and D. Hess. 1990. Bean-
pathogenesis-related proteins deduced from elicitor-induced transcripts
are members of a ubiquitous new class of conserved PR proteins
including pollen allergens. Mol. Gen. Genet. 222:353-360.

146

APPENDICES

147

APPENDIX A

148

 

 

149

0.6. Em o: 9. QB 88 me E 8.
mam 3. mos ow 3m SR 8 8 mm
3 5. we 3 3w 3.8 mm mm mm
3 0m 62 me ism 88 B we &
mam 3. BF 3 08 at. we to. om
3 ma 2: 9V mam No; mm mm 2
9m 3 BF 9. 3m 594 om mm 8
3 3 no? m... emu Ron 8 8 t
3. 3 SF me mom 3 3. 8 mm 8
nm 0m «2 9. 3m 88 mm mm 2
on on 2: 3. EN 85 on 8 3
nm 3 mo? we 4. no. 8% we 3. 2
mm mm .2: 9. 0.3 BR 8 mm NF
3. 3 we 3. m. E. 59. 2 we 5
ma 3 m: we NNN 85 8 B or
mew ma k: 3 EN $8 on mm m
3 am m2 mm o. 8 KB 8 9. m
ed «a 2: 9. F. E No; 8 mm e
9m 3 o: 3 new man 2 t. m
on E 2: 3 3m 8% 8 mm m
on 3 me 9. EN 5:. mm 3” v
ow 0m 62 m4 mam 83 E 3. m
3 em 2: 3 0mm 2% 8 8 N
a. F mm 2: 3. gm 82. mm mm P

3:36;. 8:632”. $888th 859; g g A -8:
msg#83262 .6583 2 6.60 2 mac 6% 88 22> a .8 5E

 

.58 5 .=2 .E_m8co_2 Hm 23on 20E 923 Love: 5505 3.1 co>mm>mcam .6 9:86:65 new @5930.
.btamE 9 9:8 .9526: 2 9:3 .mflw Room .263 .29 85205 ommmwfi .cmov xmuE 3:93 mmmowfi .2 ofimh

 

ON o.m vov 0v _..©N VFNV 5 on av

 

QN mm 8 me o. a «3. we 5 m...
3 3 mo: 3 3m 89. 8 mm 2
ON 3. 8 3 Km 88 2 E 9.
em 3 2: 9 6.8 88 2 we we
3 3 o: 3 EN 8:. 5 m... 3.
0m 3. 9: 3 new 05. 2 mm 2.
2.. 3 2: 6... mm: 2:: mm mm N4
3 3 mo: 3 4.8 8% m: we 3
rem 0m E 2. 0mm 4mm... 5 mm ow
3 9m 2: me «mm 684 8 NV 8
o.m 3V 8 mm 2: 88 E me mm
on ma 0: S. New 8:. 8 on em
ma 3. 2: we a. E omen E B on
mam we 9: me 3m 38 No .8 mm
3 we 8 3 m. N 9.9. 2 cc em
3 tn 8 B mam 34m .2 9. mm
3 4.4 2: mm ZN ~me 8 mm mm
ca ma 2: me 98 85 3 mm 5
em 3 x: 9. NS 86m 8 E om
Qm Qm 2: .3. 9mm moon we 8 mm
3. 0m 2: S. emu mum». E 5. 8
am am 2: :5 v.8 8m... mm 8 R
9m em 2: m4 0% 85 8 m... 8
9m Nm 2: 44 New «mt. we 3. mm
3:22. 5.5626: $868on $59; 3.: g A -8:
@8262 .8663 2 £8 9 98 86 68m 29> a .8 ..E

 

.2 6.8: Co 8:62:80

150

0025520: 0* cha >4.

 

 

m... cm; 0. <55 mama 93 036 8 03m 8 58:62 203.6088»
5 om- V 33 GE e633 a. A common- p 392ng Bmflczq
mo mm mm wSoN mad Sm 8m ab Nm
9 3 mo wwom mob Sm 3m who mb
wm mm mm SEN mob S SS 9N wb
mm Sm Nm SSNN name we 2: “N Nm
mS SN Nm um”: NSb SS 3m S.S Nm
mm mm mm mmmm B .m Sm 8N Sm S.m
mm am we SONS N. .m SN 3N wb Nb
mN mm mo 3 No B .m So So Wm Nm
mm mo mo wOSm 8b SS :5 S.S wb
mm Sm mN Sumo NPN So om S; Na
mo ".1 mm wowS B .o Sm 3m S.» S.o
Q. 3 mo 33 SS So 8w Nm Hm
mm mm NA SNom Nob S. SS wb Hm
mu So N4 mmém mm.» Sm 8S Nm Nm
mS mm mS wmmS mob SN 3m ab mb
mm mm m: SNAS mob Sm 8» Wu Pm
mm SN mo S03 MNS SS 8S NN No
mN Sm No SwOS B .4 SS 3m SN No
mm SN No mama mob SS 8w Nm Pm
me “S 3 mm: MNm Sm 8w N.» No
No Sm NN 3mm 3 .m Sm 8S Na ._.m
NA mm S mem NOS Sm 8m wb No
NM S._ we mmmm B .o Sm 3o wS Po
Nw m; we SmSm mob Sm 3m NN Wm
NS Sm mm SSom mmb Sm 8w No Wm

 

151

.szmoanm new Earn: 2:3": 92.? m 9 F 225 m .9230. o>_mmooxoum new @580. 0:": 92.2, m 2 F 2QO :

 

 

152

3 on 2: 2. 3m 82. B 2. 8
3 2m 2: 2. 28 2.2. 2. 2m 3
2m 3. 2: 2. 3: SE 8 2 8
on S. 2: 2. EN 8 5 em 8 mm
3.. 3. 2: 2. Nam 42....” E t. .3
6: 2m 2: 3. 4.2.. 85 R S 8
3 Sn 2: 2. EN 2.2. 8 B 2..
3 2m 2: 2. SN 2.2. 2 2. a
2m 5 2: 2. 24.6. R2. E on om
2m 2.” 2: 2. 3m 2:. 8 2. 2.
3. ON 2: 2. 52 32. 5 8 2w
3 3. 2: 2. 22.. 2.2. E 2. 5
3 2m 2: em v.8 5:. E 2. 2w
Om S 2: 2. 2% mmmm mm 8 2w
3 Sm 2: 2. 28 232” E B 8
2m 2m 2: 3. 28 32. 8 8 2w
3. 2m 2: 2. 2cm 3% mm 2. mm
a: 2m 2: 3. 3m 82. 2K 3. 5
2m 2m 2: 3. Km 82. E 2. 8
6.: 2m 2: t. 4.8. 28 8 mm 2
3 2m 2: 3. SN 22. NS 2. E
3 2.6. 2: s. N. :N 23. E on R
6: ON 8 2. 2mm 82. NS 3. E
3 2m 2: 2. 3m 22. 2. mm E
2.32: 9:95.“. 2768822 829; 32 30v 2 -25
«926229.... .9663 9 28 9 28 6% 88 22> a an ..E

 

._.< 953. So cozmnczcoO

 

 

153

3 2m 2: 2. 22: «mm... mm 2. vm
ON 3 mm 2. 2.8 82. 2. R 8.
2m 2. 2: 2. 22: RS 5 2.. mm
3 2m 2: 2. Nam 82. 2. :N a
3 2m «5 8 EN 22. E :m om
3 ca 23 2. 3m 82. 2.. a 2:
3 S E 2. 2% :2. mo mm 2
m: a: 2: 2. 23 22. mm 2: 2
2m 2m 2: 2. 2.2: 28.. mm m: 2
ma 3 2: 2. 2% $2. 2. 2.. 2
2m 3 2: 2. RN 622. B ..N E
:.m ma «3 2. 2mm 82. 2. t 2:
2m 3 2: .3. 2mm 82. 2. mm N:
2m 2m 2: 2. 25 2.2. R t :
2m 3 2: 5 tom 2.2.” mo 3 2
2m 3 E 2. 28 232. 8 mm m
3 2m 2: 2. 28 $2. 8 mm m
2~ 3 2: 2. 2:: 82. mm mm 2
5 on t: 2. 2mm 82. B an o
2.6. S N: 8 2E 32. 2 mm m
3 2m 2: 2. 28 «22. 5 mm 2.
2m 3 2: 2. NNN 22.. ..m m: m
3 SN 2:: 2. 2% :2. mm ..N N
2m 2.. 2: 2. 32.. 02.6 mm 3 :

25.2: 626262 2768829 2659; 2.2 22v 2 .2:%

236262 .852... 2 28 9 .28 6% 68m 29> a .8 2m

 

.Noom c_ ._s_ .E_moEo_2 Sm masmmoa 20E 223 525 $505 3.x cm>mm=mc=m So 920928.... new @5930.
2:5me 2 22mm .9th0: 9 260 .36 88 .22.» .29 85202 mmmmmfi .cwov xons bto>mm @2820 .2 2an

 

 

2N N.N 2: :.v EN 22” 2 2. 2.
N.N «N 2: m. 28 23. 2. N 2.
:.N 2m 2: 2. 22 2.2. R. MN 2.
N.N 2m 8 2. NdN :2... 2. oN 2.
Nm 3. N:: 2. :..ON 2:... 2 2 2.
2m 3 N:: 2. 22 2N2. 2. on 3.
2N 2N 2: 2. 28 32. :.N 2 2.
N.N 2m ::: 2. :..2 mN2. No N N:.
:.m 2N 2: 2. N2 82. :.o Nm ::.
2m :..N N: : Nm 2NN 22. :2 R 2.
2N N.N ::: 2. 2:N N22. :o mN N.N
3 :.m 8 2. 22 22. «N N: 8
N.N :.N 2: 2. QNN mNE 2., 2 B
2N N.N 2: t. 28 NE. :o :.N on
:.m 2:. 2: 2. 2:N 22. 2 2. mm
3 3 :.N 2. 22 2:. 8 ::. ..m
:.N 2N 3 2. :.:.N oN2. «N 2 mm
:.m 2m 2: 3. v.2 82. NN 2.: NN
N.N N.N 2: 2. o.mN 2N2“ mN :: :m
2m N.m :.2 3. 22 88 3. NN om
N.N 2m ::: 2. 2mN NE. 8 2H mN
..m ..m 2: on 22 22. 8 :m mN
:.N :.N 2: 2. 2NN 22. 2. Nm N
:.N 2N B: 2. NdN mon 8 2 8
2N :.N 2: 2. N.oN :22. NN m mN

2.22: 8:25.“. 2782022 329: g 3.; 2 2:2

.2325... .252... 2 28 2 28 $7. 88 22> a .8 ..E

 

N... 2%: :o 8:95on

154

 

 

155

5 mm NS 3 new a? me E E
3 3 o: t. «.8 3% mm mm 2
3 2” N: 3 tom $3. 2 cm Q
ad tn we 9. m8 8:. 3 mm E
2 N.N 2: 3 m2 «m9. 8 S 2
mm 3 08 S o. a 39. Q on 8
mam N.N 2: 3. 0.2 on? on 8 8
3 mm 2: 8 3m 3% 8 8 E
E N.N 2: 3 3m 59. S cm 8
3 ma 5. 8 wow 53 9. E 8
ma 1m 2: 9. 92 8mm 5 am we
o.~ 3. m9 3 Now 3% E B 8
nu ma 2: 3 92 89. on 5. mo
3 3 2: 3. v.8. we? 8 m 5
Na 3 BF S 02 9.9 E on 8
3 0m 3 «w 3: 83 9. mm am
am 3 x: 9. wt 88 8 «m mm
on o.m 5 3 a? 8% mm 2 no
2 N.N o: 2 EN 5: 3. mm 8
mm 2. 2 P E «.8 mm? mm 8 mm
Nm 3 z: 3 m.- $3 3‘ mm E
3 mm mm 3. EN 59. 2. mm mm
3 in we NV New 33 E 9. mm
3 mm 2: av 98 R5 5 B B
N.N 2., m2 9. 02 38 mo vm om

€39: @5526: $889.9 3:35 304 g A -8:

Nmss§§< 3583 2 £3 9 98o 36 8mm 22> a .8 4E

 

.2 298 8 85528

.memoanm ES Ewtms >=El 20:3 m 8 _ 28m N .mevB omeQonum can wEwco. 9.“— 82? m 8 _ 28m _

 

 

156

3 ma N9 9 tom 88 S mm 8
ma tm mm 3. m. 5. ES mo 3 S
cm S we 3. 3: 58 3 mm 8
3.. 3 m2 3 New mSm cm 8 mm
mam 3. m2 8. 3m 88. E S Sm
o.~ 3 2: 3 wow 8S 5 2 mm
on on SE S New 83 3 mm mm
0.0. 3 N: on 08 8% E S 5
N.N mm 2 F ow 0%.. 8S S. mm om
3.. 3 m2 3 3m NBS S m: 8
SN 3 SP 3 «Nu mm? mm 9 mm
on Qm N: 3 SN 85 E .S S
Qm 3 m2 2 2: OS? on S 8
mm mm 2: .3. 3m 2S 3 mm mm
0.0. mam m2 3 9mm 35 3 mm Sm
Qm 3. o: S 2: ES 8 S 8
hm Qm S: 8 a? E S 2 5 mm
N.N 3.. BF 9. now mm? mm mm E
3 mm o: 3 Now mono S M: 8
3 on o: e. «.2 8S 8 mm 2
S mm 2: 3. 0% ~me 8 R 3
So. 3 2: 3. «.2 BS 5 NF R
a; 3 mm 9. EN 88 Sm 3 2
SN .3 2: .3 o. E 5? mm M: E
3:22. @5532“. $888va 8&9: 3.; 3.; «-8:
«9202202 3:68.. 9 £8 2 $8 gm 8% 29> 5 .8 4E

 

.2 9an .6 8:95:50

Table A3 Primer combination and fragment size of AFLP markers tested in
Bunsi/Raven RIL population.

 

AFLP marker Primer combination1 AFLP marker Primer combination
1 EAACMCTT130 27 EAGCMCAA480
2 EMCMCTT207 28 EAGCMCAASZO
3 EAACMCTT223 29 EAGAMCT6190
4 EAACMCAABB 30 EAGAMCTCAOO
5 EAACMCAA380 31 EAGAMCTGMO
6 EAACMCAA147 32 EAGAMCAH 02
7 EAACMCAABZO 33 EAGAMCAH 35

8c2 EMCMCAA177/2223 34 EAGAMCATMB
9 EAegMCTG17O 35 EAGAMCAfl 65
1 0 EAGGMCTG34O 36 EAGAMCAH 8O
1 1 EACTMCTT1O3 37 EAGAMCATZOO
1 2 EACTMCTfl 70 380 EAGAMCA7349/350
1 3 EACTMCTT21 O 39 EATAMCAG1 30
14 EACTMCTTZBO 4O EATAMCAG157
1 5 EACTManQO 41 EATAMCAG85
1 6 EACTMCTT67O 42 EATAMCAGQO
1 7 EACTMCTflOO 43 EMTMCAA285
1 8 EACTMCTT720 44 EMTMCAAZSO
19 EACCMCTTZQO 45 EAATMCAA134
20 EACCMCTT33O 46 EACTMCCA190
21 EACCMCTTBSO 47 EACTMCCA120
22 EACCMCTTSSO 48 EACTMCCA1 10
23c EAGTMcn21 9/220 49 EAATMCAA4OO
24 EAGTMCTTBOO 50 EMTMCAA1 15
25 EAGTMCTTSOO 51 EAATMCAA1O5
26 EAGCMCMAOO 52 EAGGMCAA17O

 

157

Continuation of Table A3.

 

AFLP marker Primer combination‘ AFLP marker Primer combination
53 EAGGMCAA85 79 EAAAMCAASSO
54 EACAMCCG1OO 80 EMAMCAA6OO
55 EACAMccg120 81 EAAAMCAA650
56 EACAMCCGZOO 82 EAmMCTC1 10
57 EACAMCC346O 83 EATGMCTCZZO
58 EACCMCTGZ55 84 EATGMCTc33O
59 EACCMCTG350 85 EAmMCTCBZ
6O EATCMCCT33O 86 EATGMCTG120
61 EATCMCCTQO 87 EATGMCTG4OO
62 EACCMCTC45O 89 EAGCMCAA1O5
63 EACCMCT386 90 EAGCMCAA128
64 EAGTMCAT22O 91 EAGCMCAA190
65 EAGTMCA128O 92 EACTMCAC120
66 EAGTMCA736O 93 EACTMCACZO5
67 EAGTMCAT48O 94 EACTMCAC4OO
68 EAGTMCAT75 95 EACTMCA122O
69 EAGTMCATQO 96 EACTMCAT26O
70 EAAAMCAA1 04 97 EACTMCATSSO
71 EAAAMCAA17O 98 EACTMCAT85
72 EAAAMCAA1 90 99 EAAAMCAG1OO
73 EAAAMCAA195 100 EAAAMCAG1O3
74 EMAMCAA223 101C EAAAMCAG195/197
75 EAAAMCAA225 102 EAAAMCAGZ1O
76 EAAAMCAA227 103 EAAAMCAG41O

77c EAAAMCAA269/27O 105 EACCMCCT31O
78 EAAAMCAASOO 106 EACCMCCTBZ

 

158

Continuation of Table A3.

 

AF LP marker Primer combination1 AFLP marker Primer combination

1 O7 EACCMCCT97 1 14 EACTMCAA46O

108 EACCMCGA142 1 16 EACTMCAA75

1 09 EACCMCGA3OO 1 1 7 EACTMCAASBO

1 1O EACCMCGA4OO 1 18 EACTMCAA550
1 1 1c EAGGMCAC145/147 1 19 EACTMCAASBO

1 12 EAGGMCACZS6 120 EACTMCAAQO

1 13 EAGGMCAC78 121c EACTMCAA33O/320

 

1 The letters E and M indicate the restriction enzymes EcoRI and Msel. The three
letters following E indicate the EcoRI (ANN) selective nucleotides and the three
letters following M indicate the Msel (CNN) selective nucleotides. The number
represents the approximate size of the polymorphic band.

2 0 indicates codominant marker

3 The two numbers indicate the size of both fragments of a codominant marker.

159

APPENDIX B

160

 

Table 31. Lodging, desirability coefficient, maturity, seed size and reaction to
straw test in 89 BC2 IBL derived from the cross between Tacana and
PI 318695, grown in Saginaw, MI, in 2002.

 

 

IBL code Lodging1 082 Days to Seed size Straw
(02T- ) Maturity3 (g.1005eed“) Test4
8571 3 4 M 18.1 2.3
8540 4 4 E 19.0 2.5
8563 2 6 E 23.4 2.6
8553 3 5 M 20.2 2.7
8600 3 3 L 19.1 2.7
8580 2 5 M 22.8 2.7
8554 3 5 E 21.2 2.8
8533 2 4 M 16.9 2.8
8523 2 6 E 21.4 2.9
8537 3 4 M 22.6 3.0
8535 3 5 E 18.8 3.0
8509 3 4 M 18.5 3.0
139-1 3 4 E 20.0 3.1
8568 2 5 M 21.5 3.1
8595 3 4 L 17.7 3.2
8575 4 3 E 20.1 3.2
8596 2 5 M 20.7 3.3
8506 3 4 M 19.0 3.3
8577 3 4 L 26.1 3.3
8570 3 4 M 20.5 3.3
8574 2 7 E 21.4 3.3
8515 3 6 E 21.3 3.3
8527 3 4 E 19.6 3.3
8592 2 5 L 17.3 3.5
8551 2 4 M 16.7 3.6
8550 3 4 E 20.0 3.6
8561 3 5 E 15.2 3.6
8543 2 5 E 24.3 3.6
122-1 3 4 E 18.0 3.7
8526 3 3 L 15.5 3.7
8582 2 4 E 18.3 3.8
8503 3 5 E 19.4 3.8
8566 3 4 L 20.1 3.8
8558 3 6 E 18.3 3.8
97-3 3 6 E 17.6 3.8
95-3 3 4 E 18.0 3.9
8597 3 3 M 16.9 3.9
111-1 3 3 L 18.0 3.9

 

 

161

Continuation of Table B.1

 

 

 

IBL code Lodging1 082 Days to Seed size Straw
(02T- ) Maturity3 (g.1OOSeed") Test4
8573 4 3 E 19.2 3.9
122-3 2 5 M 17.5 4.0
8588 3 4 L 19.4 4.1
8545 3 4 E 18.0 4.1
8586 2 4 M 17.8 4.1
127-2 3 4 M 18.8 4.1
8518 3 4 E 16.8 4.2
8536 3 4 E 13.4 4.2
8549 2 4 M 18.5 4.3
139.54 3 4 M 16.3 4.3
8524 3 5 E 20.5 4.4
8517 2 5 E 18.8 4.4
8581 2 5 E 22.0 4.4
8541 2 5 M 19.1 4.5
8531 3 4 E 17.7 4.5
8579 4 3 M 16.4 4.6
8547 3 4 L 20.9 4.8
8585 3 4 M 17.7 4.8
8532 4 3 M 16.8 4.8
8562 4 3 E 14.6 4.8
8556 3 7 E 16.1 4.8
8589 3 4 L 19.1 4.8
8598 4 3 M 18.1 4.8
8590 3 4 L 15.9 4.9
8576 3 5 M 22.2 4.9
8534 2 4 M 15.7 4.9
8587 3 3 M 18.8 4.9
8548 4 3 M 18.1 5.0
8530 2 6 E 20.7 5.1
97-2 4 3 M 16.4 5.1
114-4 3 4 M 16.4 5.2
8522 3 5 E 16.1 5.2
8564 3 4 E 20.7 5.3
8542 2 6 M 16.5 5.3
8544 2 4 E 14.4 5.3
8578 3 4 L 18.9 5.3
8520 3 4 M 17.4 5.4
8593 3 4 M 17.5 5.4
8529 3 5 E 19.2 5.4
8572 3 4 M 20.1 5.5
8512 3 4 M 19.9 5.6

 

162

 

Continuation of Table 81

 

 

 

 

 

IBL code LodgingT DS2 Days to Seed size Straw

(02T- ) Maturity3 (g.1OOSeed") Test“
8538 4 3 E 15.3 5.6
8557 3 5 E 17.3 5.7
8504 2 4 M 18.8 5.8
8516 2 4 L 19.7 5.9
126-2 2 5 M 21.0 5.9
118-2 3 4 M 18.7 5.9
8501 4 3 E 16.2 6.3
8507 3 3 M 17.7 6.8
8599 3 4 M 18.4 7.1
8508 3 4 E 17.3 7.2
Average(89) 2.9 4.3 - 18.7 4.3

Range 2 — 4 3 - 7 - 13.4 -— 26.1 2.3 — 7.2
Tacana 2.0 5.0 M 21.5 3.4
Pl-318695 - - - 3.55 4.5

 

1Lodging is rated in a 1 to 5 scale, where 1: no lodging, 3 = moderate lodging,
and 5 = excessive lodging.

2 Desirability score, the higher the score (from 1 to 9) the more desirable is the
line.

3 E=early, M= mid season and L=late, relative to days to maturity in Tacana.

4 Straw test is scored in a scale from 1 to 9 where 1=no symptoms and 9=
complete collapse of the plant.

5 Seeds from plants grown in the greenhouse.

163

Table 82. Lodging, desirability coefficient, maturity, seed size and reaction to
straw test in 76 BC2 IBL derived from the cross between Tacana
and PI 313850, grown in Saginaw, MI, in 2002.

 

 

IBL code Lodging1 DS2 Days to Seed size Straw
(O2T- ) Maturity3 (g.1003eed'1) Test“
8607 4 3 M 25.8 1 .3
8673 3 3 L 24.4 1 .4
8601 2 3 M 27.1 1.8
8636 3 4 M 23.4 2.0
8604 2 4 M 23.2 2.0
8668 2 3 E 23.9 2.1
8659 2 5 E 24.3 2.2
8683 2 5 E 23.6 2.2
8605 3 5 M 24.3 2.2
8621 2 4 L 24.0 2.3
8649 3 4 E 22.3 2.3
8679 2 4 M 27.5 2.3
8635 2 3 E 26.5 2.3
8655 4 3 L 25.5 2.3
8629 2 4 M 23.6 2.3
8640 4 4 L 22.6 2.3
8645 3 3 L 18.3 2.3
861 1 3 3 E 23.8 2.4
8628 4 3 M 26.5 2.4
8633 2 3 E 22.7 2.5
8657 3 2 M 26.3 2.5
8627 2 4 E 21 .3 2.6
8631 2 4 M 24.1 2.6
8684 3 4 M 25.7 2.6
8692 2 5 M 23.2 2.7
8606 4 4 M 23.8 2.7
8613 3 4 M 24.2 2.7
8615 2 4 M 23.4 2.7
8693 2 4 M 22.8 2.7
8642 3 5 E 23.8 2.7
8622 2 4 E 22.7 2.8
8641 2 5 E 24.4 2.8
8691 2 5 E 24.1 2.8
8632 3 4 L 31 .9 2.8
8644 3 4 M 23.9 2.8

 

164

 

 

Continuation of Table 8.2.

 

 

IBL code Lodging1 DS2 Days to Seed size Straw
(02T- ) Maturity3 (g.100seed") Test4
8678 3 5 E 23.7 2.9
8660 2 5 M 23.0 2.9
8680 2 5 E 23.0 2.9
8647 4 4 M 21.3 2.9
8616 2 4 M 23.3 3.0
8670 2 5 E 23.9 3.1
8675 3 5 M 23.3 3.1
8685 2 5 M 21.6 3.1
8688 4 3 M 24.7 3.1
8686 2 5 M 22.3 3.2
8612 2 4 M 24.5 3.2
8651 3 4 M 25.1 3.2
8662 3 4 M 24.3 3.2
8694 2 4 M 21.6 3.2
8650 2 5 E 25.1 3.3
8676 3 5 M 22.5 3.3
8681 3 5 E 23.5 3.3
8689 2 5 M 21.6 3.3
8656 3 4 E 27.5 3.3
8696 2 5 M 22.8 3.3
8624 3 4 M 24.9 3.3
8663 3 5 M 24.6 3.4
8630 3 4 M 23.4 3.5
8677 2 6 M 25.0 3.6
8666 3 5 E 24.6 3.6
8623 3 3 L 25.6 3.6
8625 2 4 E 23.6 3.7
8626 2 5 E 23.5 3.8
8637 3 2 L 25.1 3.8
8667 3 5 E 22.8 3.9
8671 3 5 M 24.1 3.9
8661 3 5 M 23.5 3.9
8653 3 5 M 21.9 4.0
8672 3 6 M 24.7 4.1
8608 3 4 M 23.0 4.2
8639 3 4 M 27.5 4.2
8690 2 5 M 23.9 4.3

 

165

Continuation of Table 8.2.

 

 

 

 

 

IBL code Lodging1 082 Days to Seed size Straw
(02T- ) Matu rity3 (g.1005eed") Test4
8638 3 5 E 21.1 4.4
8695 3 3 L 28.6 5.1
8652 3 3 M 24.6 5.7
Average(75) 2.7 4.2 - 24.0 3.0
Range 2—4 2—6 - 18.3—31.9 1.3—5.7
Tacana 2 5 M 21.5 2.8
Pi 313850 - - - 61.05 3.3

 

1 Scale from 1 to 5, where 1= no lodging and 5 = excessive lodging.

2 Desirability score, the higher the score (from 1 to 9) the more desirable is the

line.

3 E=early, M=mid season and L=late, relative to Tacana.
4 Scale from 1 to 9, where 1=no symptoms and 9= collapse of the plant.
5 Seeds from plants grown in the greenhouse.

166