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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:lProj/Acc&Pres/ClRC/DateDue.indd

MAPPING QTL FOR AGRONOMIC AND CANNING QUALITY TRAITS IN BLACK
BEAN (PHASEOLUS VULGARIS L.)

By

Evan Michael Wright

A THESIS

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

MASTER OF SCIENCE

Crop and Soil Sciences

2008

ABSTRACT

MAPPING QTL FOR AGRONOMIC AND CANNINGQUALITY TRAITS IN BLACK
BEAN (PHASEOLUS VULGARIS L.)

By
Evan Michael Wright

Quantitative trait loci (QTL) analysis was used to identify QTL for agronomic
performance and canning quality in a recombinant inbred line (RIL) population from the
cross of the black bean cultivar ‘Jaguar’ and the breeding line 115M. A total of 96 RILs
were evaluated for seven agronomic and four canning quality traits in replicated trials at
one location over four years (2004-2007) in Michigan. SSR, TRAP, SRAP, and
phenotypic markers were used to create a genetic map of the population consisting of 119
loci including a locus associated with resistance to a new race of bean rust isolated in
Michigan. The map consisted of 15 linkage groups spanning 460cM (3 8%) of the bean
genome. Composite interval mapping analysis identified a total of 20 QTL for 10 traits
averaged across environments, while an additional 18 QTL were identified in one or
more individual environments. QTL were identified on 10 linkage groups (LG). A major
QTL for seed yield was identified on LG B10. A total of 7 QTL for yield, seed size,
plant height, and canned bean texture showed positive alleles from 115M. Several QTL
co-localized with regions identified in previous studies while others, particularly for
canning quality, were unique. Rust resistance associated with 115M was mapped to LG
B4 and flanked by two TRAP markers, both at an approximate distance of 3 cM. These
results support the utility of TRAP markers to tag disease resistance loci and QTL and
provide a valuable source of rust resistance for future black bean cultivars adapted to

Michigan.

ACKNOWLEDGEMENTS

I would like to thank Dr. James Kelly, for serving as my major professor and
advising this research. I appreciate his patience and understanding over the years. His
enthusiasm for plant breeding and common bean has been inspiring. I feel fortunate to
have had such an outstanding mentor in all aspects of practical scientific research and

professional life.

I also appreciate the members of my guidance committee, Dr. James Hancock, Dr.
Dechun Wang, Dr. Ryan Warner, and Mr. Greg Varner, who applied their varied
expertise to improving this research and thesis. Their advice and helpful conversations

have enhanced this work.

To the members of the bean lab, your fi'iendship and discussions over the years
have been great. Your help with many aspects of my research has also been crucial and
truly appreciated. Thanks for making the field and lab such an enjoyable place to work

and conduct research.

I also thank my wife, April, for her faith and support throughout this season of my
life. I am also thankfiil to my family for inspiring me to learn more about the scientific

aspects of agriculture, and for their interest and encouragement.

Finally, I would like to thank the many friends I made while at Michigan State

University for creating such an enjoyable community in which to work and learn.

TABLE OF CONTENTS

LIST OF TABLES .................................................................................. vii
LIST OF FIGURES .................................................................................. x
CHAPTER 1: LITERATURE REVIEW .......................................................... 1
Introduction ...................................................................................... 1
Domestication ................................................................................ 2

Gene Pools and Races ....................................................................... 3
Preserving Genetic Diversity ............................................................... 5
Utilizing Genetic Diversity ................................................................. 6

Wild Beans to Improve Yield ................................................................. 7
Quantitative Genetic Variation ............................................................ 7
Application of QTL Analysis and Marker Assisted Selection ........................ 10
Markers and Mapping ..................................................................... 10
Biochemical Markers ............................................................. 11

Molecular Markers and Bean Linkage Maps .................................. 11

Mapping and Tagging Genes and QTL ........................................ 14

Disease Resistance ......................................................................... 16
Anthracnose ........................................................................ 16

Common Bean Rust ............................................................... 17

Processing Quality ......................................................................... 19

Color ................................................................................ 20

Texture .............................................................................. 20

Visual Appearance and Washed-Drained Weight ............................ 20

Inheritance of Canning Quality Components ................................. 21

Indirect Screening Methods for Canning Quality ............................. 22

Conclusion .................................................................................. 23
Literature Cited ............................................................................. 24

CHAPTER 2: IDENTIFICATION OF QTL FOR AGRONOMIC AND CANNING

QUALITY TRAITS IN A BLACK BEAN POPULATION .................................. 35
Abstract ...................................................................................... 35
Introduction ................................................................................. 36
Materials and Methods ..................................................................... 40

Plant Materials ..................................................................... 40
Field Trials ......................................................................... 41
Canned Bean Evaluation ......................................................... 42
DNA Isolation and Molecular Marker Analysis .............................. 42
SSR Markers ....................................................................... 42
SRAP and TRAP Markers ....................................................... 43
Phenotypic Markers ............................................................... 43

iv

Data Analysis, Linkage Map Construction, and QTL Analysis ............ 43

Results ........................................................................................ 45
Field Trials ......................................................................... 45
Canning Traits ..................................................................... 46
Markers and Linkage Map ....................................................... 47
QTL Analysis ...................................................................... 48
Yield ........................................................................................................... 49
Seed Size ........................................................................... 49
Days to Flowering ................................................................. 50
Maturity ............................................................................ 50
Lodging ............................................................................. 50
Height ............................................................................... 50
Desirability ......................................................................... 51
Canned Bean Color ............................................................... 51
Texture .............................................................................. 51
Visual Appearance ................................................................ 51
Washed-Drained Weight ......................................................... 52
Co-localized QTL ................................................................. 52
QTL x Environment Interactions ................................................ 52

Discussion ................................................................................... 53
Field Trials ......................................................................... 53
Canning Traits ..................................................................... 55
Linkage Map ....................................................................... 56
Segregation distortion ............................................................ 58
QTL ................................................................................. 59

Yield ........................................................................ 60

Seed Size .................................................................. 61

Days to Flowering ........................................................ 62
Maturity .................................................................... 62
Lodging .................................................................... 63

Height ...................................................................... 63
Desirability ................................................................ 64
Canning Traits ............................................................ 64
Canned Bean Color ............................................. 65

Texture ............................................................ 65

Visual Appearance .............................................. 66
Washed-Drained Weight ....................................... 66

Co-localized QTL ................................................................. 67
Combined and Individual Environments ....................................... 67

Conclusions ................................................................................. 68

Tables ........................................................................................ 70

Figures ....................................................................................... 84

Literature Cited ............................................................................. 90

CHAPTER 3: USE OF TRAP MARKERS TO MAP RESISTANCE TO A NEW RACE

OF COMMON BEAN RUST IN MICHIGAN ................................................. 96
Abstract ...................................................................................... 96
Introduction ................................................................................. 97
Materials and Methods ..................................................................... 99
Results ...................................................................................... 100

Reaction of cultivars ............................................................ 100
Characterization .................................................................. 100
Mapping and Tagging ........................................................... 101
Discussion ................................................................................. 101
Reaction of cultivars ............................................................ 101
Characterization .................................................................. 102
Mapping and Tagging ........................................................... 103
Conclusions ................................................................................ 105
Tables ....................................................................................... 106
Figures ...................................................................................... 109
Literature Cited ........................................................................... 110

APPENDIX A: INDIRECT SCREENING FOR COLOR LOSS IN TWO BLACK BEAN

POPULATIONS ................................................................................... 1 13
Introduction ................................................................................ l 13
Materials and Methods ................................................................... 114
Results and Discussion .................................................................. 114
Conclusions ................................................................................ 1 16
Tables ....................................................................................... 1 18
Literature Cited ........................................................................... 120

APPENDIX B: VALIDATION OF THE SOAK WATER COLOR TEST IN THE

‘JAGUAR’/1 15M RIL POPULATION ......................................................... 121
Introduction ................................................................................ 121
Materials and Methods ................................................................... 121
Results and Discussion .................................................................. 122
Conclusions ................................................................................ 125
Tables ....................................................................................... 126
Literature Cited ........................................................................... 130

APPENDIX C: SUPPLEMENTAL DATA COLLECTED FROM THE

‘JAGUAR’/1 15M RIL POPULATION FROM 2004-2007 ................................. 131
Tables ....................................................................................... 131
Figures ...................................................................................... 150

vi

LIST OF TABLES

Table 2.1. AN OVA table showing mean squares (p50.0001) for yield, 100 seed weight,
days to flowering, plant height, lodging score, maturity, and agronomic desirability score
for 96 recombinant inbred lines in the Jaguar/115M population combined across four
environments (2004-2007) in Michigan .......................................................... 70

Table 2.2. Pearson correlation coefficients for agronomic and seed quality traits from 96
recombinant inbred lines developed from a ‘Jaguar’ by 115M cross grown in Michigan
during 2004-2007 ..................................................................................... 71

Table 2.3. Phenotypic means and ranges for yield, 100 seed weight, days to flowering,
maturity, lodging score, plant height, and desirability score; canned bean color, texture,
visual appeal, and washed drained weight for 96 recombinant inbred lines in the
Jaguar/115M population combined across four environments (2004-2007) in
Michigan......... ................................................................................................................ 72

Table 2.4. Four year average (2004-2007) seed yield and 100-seed weight of the top ten
and bottom five recombinant inbred lines in the Jaguar/115M population ranked by seed
yield .................................................................................................................................... 73

Table 2.5. Flowering day, maturity, lodging score, plant height, and agronomic
desirability score of the top ten and bottom five recombinant inbred lines in the
Jaguar/115M population ranked by average seed yield from 2004-2007 .................... 74

Table 2.6. Canned bean color, texture, visual appearance, and washed-drained weight of
the top ten and bottom five recombinant inbred lines in the Jaguar/115M population
ranked by average seed yield from 2004-2007 .................................................. 75

Table 2.7. Number of years recombinant inbred lines in the Jaguar/115M population
ranked in the top 10% or bottom 5% based on seed yield ...................................... 76

Table 2.8. Yearly trait means for 2004-2007 and corresponding least significant
differences for the ‘J aguar’/ 1 15M recombinant inbred line population grown in
Michigan ............................................................................................... 76

Table 2.9 Four year mean (2004-2007) yields for 96 recombinant inbred lines in the
Jaguar/115M population, parents, and checks ranked by descending seed yield ........... 77

Table 2.10. Analysis of variance for canning quality traits of a ‘Jaguar’ by 115M RIL
population including canned bean color, texture, washed-drained weight, and visual
appearance ............................................................................................ 78

Table 2.11. Mean values by year for canned bean color, texture, washed-drained weight,
and visual appearance for 2005-2007 ............................................................. 78

vii

Table 2.12. Putative QTL for agronomic and seed quality traits identified in the combined
environment from 96 recombinant inbred lines developed from a ‘J aguar’/ 1 15M cross
and evaluated in Michigan during 2004-2007 ................................................... 79

Table 2.13. Putative QTL for agronomic and seed quality traits identified in the combined
and individual environments from 96 recombinant inbred lines developed from a
‘Jaguar’ll 15M cross and evaluated in Michigan during 2004-2007 .......................... 80

Table 3.1. Reaction of 12 differential common bean cultivars inoculated with race 3:22 of
common bean rust collected from Tuscola county, MI ......................................... 106

Table 3.2. Reactions of selected bean cultivars to inoculation with U. appendiculatus
race 3:22 ............................................................................................ 107

Table 3.3. Reactions of 96 RILs from a ‘Jaguar’ by 115M population to U.
appendiculatus rust race 3 :22 and presence or absence of two TRAP markers (F 7R1 ,
F15R10)............ ........................................................................................................... 108

Table A. 1. Trait means and ranges for five traits in two populations segregating for color
retention based on the ‘Soak Water Color Test’ ............................................... 118

Table A2. Trait means and ranges for five traits measured on 11 pairs of NILs selected
from population 1 .................................................................................. 119

Table B.1. Phenotypic values for seed of the 96 ‘J aguar’/ 1 15M RILs evaluated by the
soak water color test and by visual evaluation of canned bean samples ................... 126

Table B.2. Phenotypic correlations between visual canning score and Hunter L value of
leachate from the soak water color test in a population of 96 recombinant inbred lines
developed from the cross ‘Jaguar’/1 15M grown in Saginaw, MI in 2005 ................. 126

Table B.3. Phenotypic values for seed of the 32 diverse genotypes evaluated by the soak
water color test and by visual evaluation of canned bean samples during 2006. (Canning
scores range from 1=undesirable to 4=neither undesirable nor desirable to

7=desirable) ........................................................................................ 127

Table 8.4. Phenotypic values for all of the 96 ‘Jaguar’/l 15M RILs evaluated by the soak
water color test and by visual evaluation of canned bean samples in 2005 ............... 128

Table O]. 2004 Agronomic and canning data for the ‘Jaguar’ by 115M RIL
population ........................................................................................................................ 13 1

Table C2. 2005 Agronomic and canning data for the ‘Jaguar’ by 115M RIL
population .......................................................................................... 134

Table C3. 2006 Agronomic and canning data for the ‘Jaguar’ by 115M RIL
population .......................................................................................... 137

viii

Table CA. 2007 Agronomic and canning data for the ‘Jaguar’ by 115M RIL
population .......................................................................................... 140

Table C.5. Three year averages for processed bean color (Hunter L-value), texture (Kg-
force), and washed-drained weight (g) measured in the ‘Jaguar’ by 115M RIL
population .......................................................................................... 143

Table C.6. Mean values by year for processed bean color, texture, and washed-drained
weight measured in the ‘Jaguar’ by 115M RIL p0pulation during 2005-2007 ............ 145

Table C.7. Reaction of 96 ‘Jaguar’ by 115M RILs following inoculation with race 73 of
C. lindemuthianum in the greenhouse. (S=susceptible, R=resistant) ....................... 146

Table C.8. SRAP primer sequences used in pairwise combinations to screen for genomic
polymorphisms in a ‘Jaguar’/1 15M RIL population. Sequence information based on Li
and Quiros (2001) TAG 103:455-461 .......................................................... 147

Table C.9. TRAP primer sequences used in pairwise combinations to screen for genomic
polymorphisms in a ‘J aguar’/ 1 15M RIL population. Sequence information based on Hu ‘
and Vick (2003) Plant Mol. Bio. Rept. 21 :289-294 ........................................... 147

Table C.10. Reactions of 96 ‘Jaguar’ by 115M recombinant inbred lines to U.
appendiculatus race 3:22 and presence or absence of two TRAP markers ................ 148

LIST OF FIGURES

Figure 2.1. Monthly precipitation (mm) measured fiom June to September
2004-2007 at the Saginaw Valley Bean and Beet Research Farm, Saginaw, MI ........... 84

Figure 2.2. Linkage map of ‘J aguar’/ 1 15M RIL population and QTL locations for seed
yield (YLD), seed size (SDWT), days to flowering (FLWR), plant height (HT), days to
maturity (MTR), lodging (LDG), agronomic desirability (DS), and canned bean visual
appearance (VA), color (CLR), texture (TXT), and washed-drained weight (WDWT).
QTL are further identified by the last two digits of year (04-07) and QTL with no year
specified were detected in the 4-year combined environment ................................. 85

Figure 3.1. Linkage group B4b of the ‘Jaguar’ by 115M recombinant inbred line
population which contains the resistance locus for U. appendiculatus race 3:22. . . . . ....109

Figure C. 1. Frequency distributions for agronomic and canning quality traits in the
‘Jaguar’ by 115M RIL population ............................................................... 150

Figure C.2. Linkage map of ‘Jaguar’/1 15M RIL population consisting of 119 SSR,
SRAP, TRAP, and phenotypic markers placed on 15 linkage groups covering a combined
distance of 460cM ................................................................................. 154

Chapter 1: Literature Review
Introduction

Common bean (Phaseolus vulgaris L.) is the most important legume for direct human
consumption worldwide. Worldwide production is nearly twice that of chickpea, the second
most important legume (Broughton et al., 2003). Dry beans provide a major source of
protein which accounts for 20-25% of seed weight (Ma and Bliss, 1978). This major storage
protein known as phaseolin has an amino acid profile that complements the deficiencies in
cereal proteins. Beans are superior to cereals in terms of micronutrient content and contain
many important minerals (Ca, Cu, Fe, Mg, Mn, P, Zn) and vitamins (folate) (Welch et al.,
2000). These nutritional properties are a critical component of the diet for over half a billion
people worldwide (Gepts, 2001).

In areas of Latin America and Africa, beans are the primary protein source in the diet
of many people with a per-capita consumption of over 60kg per year in parts of eastern
Africa (Miklas and Singh, 2007). Developed countries concerned with healthy diets are
steadily increasing consumption (Acosta-Gallegos et al., 2007), although consumer
preference for bean size, shape, and color vary widely. Sustained bean consumption has been
linked with reduced cholesterol levels and a lower risk of heart disease (Anderson et al.,
1984; Winham et al., 2007) and certain cancers (l-Iangen and Bennink, 2003).

Beans are grown on more than 14Mha worldwide. Collectively, the Americas
produce 6.7MMT. Brazil is the largest producer (2.5MMT) and consumer (>10kg/yr per-
capita), followed by the US (1 .3IVH\/IT) and Mexico (1 .OMMT) (Singh, 1999). Production
practices vary based on social and ecological factors. In Latin America, with the exception
of Argentina, more than half the production takes place on farms <20ha, ofien on small

parcels of marginal land. Similar production constraints exist in South and East Africa,

where diverse assortments of beans are often intercropped on marginal soil, and significant
losses due to biotic and abiotic stress are common (Broughton et al., 2003).

The farm value of the US. dry bean crop for 2007 was approximately $677 million
with an estimated 1.15 MMT harvested from 598,429 ha for an average yield of l736kg/ha.
These figures represented an increase in overall yield and a 22% increase in value of the
crop, but a decrease in area harvested compared with the previous season. In Michigan the
2007 crop was valued at $88 million, with 141,500 metric tons produced on 78,917 hectares
for an average of l800kg/ha, representing a decrease in area harvested and yield, but an
increase in crop value over previous years (USDA, 2008).

Domestication

Beans (Phaseolus spp. L.) are among the oldest crops in the New World. Along with
maize and cassava, they have been a staple for generations of people in the Americas and
around the world. Beans are extremely diverse in terms of morphological variation and
environmental adaptation, resulting in crops which are suited for a range of agronomic niches
and consumer preferences (Broughton et al., 2003).

Common bean was domesticated from a wild legume distributed from northern
Mexico to northeastern Argentina (Gepts et al., 1986). P. vulgaris is one of five cultivated
Phaseolus species native to the Americas. Domestication occurred over a period of 7000-
8000 years (Gepts and Debouck, 1991), and radiocarbon dating of ancient beans confirmed
this process was occurring 4000 years ago (Kaplan and Lynch, 1999). This process took
place independently in South America and Middle America from morphologically and
biochemically distinct populations (Chacon et al., 2005).

From domestication centers in South America, Mexico, and Central America,
common bean production has expanded into a range of environments. In the Americas, it

occurs from 35°S to >50°N latitude and from sea level to >3000masl (Gepts et al., 1988;

2

Beebe et al., 1997). Beans were introduced to Africa, Asia, Europe, and Oceana during early
explorations of the Americas by European traders (Gepts and Bliss, 1988). Although
common bean has diverged into diverse environments, hybrids between wild and
domesticated beans are fully fertile. This presents opportunity for barrier-free introgression
of favorable allelic diversity into cultivated bean (Koinange et al., 1996).

During domestication, common bean was transformed from a climbing, indeterminate
vine to less vigorous indeterminate vines and determinate bush habits. Sensitivity to long
day photoperiod was lost; leaves, pods, and seeds increased in size; and seed colors evolved
from speckled gray, brown, beige, and cream colors to brighter solid, striped, or spotted
colors. Pod structure was altered from highly fibrous to less fibrous, resulting in a loss of
shattering at maturity (Gepts and Debouck, 1991). Several major genes and quantitative trait
loci (QTLs) that influenced these domestication traits have been identified and mapped
(Koinange et al., 1996; Freyre et al., 1.998; Gepts, 1999).

Gene Pools and Races

Wild populations, as well as modern cultivars, can be grouped into two major gene
pools, one located in Middle America and the other in the Central and Southern Andes.
Based on recent DNA sequence information, it appears that these two gene pools originated
from an ancestral group in Ecuador and northern Peru that spread both north and south,
resulting in the evolution of the two geographically distinct groups (Debouck et al., 1993;
Kami et al., 1995). Each gene pool can be recognized by differences in seed size, seed
proteins (phaseolin), allozymes, morphological traits, and molecular markers (Gepts, 1988;
Beebe et al., 2000; Blair et al., 2006b). Strong evidence exists for multiple, independent
domestications of some races of Middle American beans, based upon multiple chloroplast

haplotypes (Chacon et al., 2005). However, all Andean beans examined share a common

haplotype, supporting the hypothesis of a single domestication before diverging into their
present domesticated races.

Each gene pool has been subdivided into races based on environmental adaptation as
well as plant and seed morphology. These races are recognized by their specific
physiological, agronomic, biochemical, and molecular characteristics. The Middle American
gene pool consists of races Mesoamerica, Durango, Jalisco, and Guatemala, while the
Andean group consists of races Chile, Nueva Granada, and Peru (Singh et al., 1991a; 1991b;
Beebe et al., 2000). The fourth Mesoamerican race, Guatemala, was not initially identified
by Singh et al. (1991a), but later proposed by Beebe et al. (2000) as a less well defined group
encompassing accessions from Guatemala and areas of southern Mexico. Members of this
race possess an indeterminate climbing growth habit and small seed size most similar to race
Mesoamerica but nonetheless unique.

During domestication, the genetic base of many crop species has become increasingly
narrow (Papa et al., 2005). At least 16-18% of the common bean genome was influenced by
selection during domestication (Papa et al., 2007). While this process eliminated many genes
for undesirable traits such as dispersal of seeds or excess vegetative growth, usefirl genes
located in close proximity to these loci were excluded from the modern cultivated gene pool.
Analysis of genomic regions surrounding these major domestication genes suggests they .
harbor much more genetic diversity in wild beans and represent an untapped source of
genetic variability waiting to be introgressed into cultivated germplasm (Papa et al., 2005,
2007). Traits present in wild populations that are absent or underrepresented among
domesticated germplasm include insect resistance and increased nutritional value
(summarized in Table 2, Acosta-Gallegos et al., 2007).

Although significant progress has been made in improving yield and disease

resistance, breeders have increasingly relied upon crosses between genetically related elite

4

germplasm to develop new cultivars (Sonnante et al., 1994). This practice has increased the
probability of producing improved progeny but further narrowed the genetic base of the crop.
Voysest et al. (1994) reported that among a subset of 130 bean cultivars each from both a
Mesoamerican race and an Andean race, over 75% of the genes present originated in that
same race, and in most cases could be traced back to about 12 parental lines based on
analysis of pedigrees. However, they also noted a tendency toward inter-racial crosses in
more recently developed bean cultivars, suggesting breeders are making an effort to maintain
or increase the genetic diversity of future cultivars.

Preserving Genetic Diversity

Vavilov (1940) described the value of collecting and preserving wild crop relatives as
sources for genetic improvement for modern agriculture. This potential motivated the
creation of seed banks to preserve the wild, weedy, and landrace ancestors of crops so that
they would be available to future generations of plant breeders. At least 700 such collections
worldwide are estimated to contain over 2.5 million accessions of wild and domesticated
germplasm (Tanksley and McCouch, 1997). Recently an underground vault in Norway was
built to store duplicate samples of these accessions to further protect plant genetic resources
on a long term basis (Charles, 2006).

A number of core collections, relatively small subgroups representing the diversity
contained in these large collections, have been assembled to facilitate the use of these
resources (Logozzo et al., 2007). Other genetic resources have been conserved in situ, either
in the wild (Debouck et al., 2008) or in some cases on farms as cultivated landraces or
farmer-maintained varieties (Lioi et al., 2005; Tiranti and Negri, 2007). Gomez (2004) found
differences between some Nicaraguan bean landraces conserved in situ, and samples of those
landraces conserved ex situ, suggesting that the natural environment may be the preferred

location to maintain these resources. Due to the highly heterogeneous nature of landraces,

5

diversity can be lost during successive cycles of seed renewal in environments that differ
fi'om the area where this germplasm was collected.

Utilizing Genetic Diversity

Although germplasm collections have been established, many have been
underutilized (Singh, 2001). Many breeders believe that there are useful genes available that
could be utilized to further improve modern cultivars, but attempts at identifying and
extracting these resources have been limited. One promising example was reported by
Lippman and Tanksley (2001), where six QTL (collectively accounting for 67% of the
variation for fruit weight) were detected in a population derived from a single inter-specific
cross between a wild tomato and the largest cultivated tomato. Although the QTL had all
been previously reported, this was the first time they had been detected together in the
progeny from a single cross.

In order to move useful diversity for genetically complex, quantitative traits such as
yield from unadapted backgrounds into elite cultivars, there is a need for prebreeding efforts.
Through prebreeding, beneficial traits can be moved into an intermediate, adapted
background, which facilitates the transfer into elite material without bringing along many
undesirable characteristics associated with the original unadapted germplasm (Gepts, 2005).

Kelly et al. (1998) proposed a pyramid scheme for organizing pre-breeding in bean.
The top tier consists of elite by elite crosses within market class, growth habit, and maturity
groupings, which ensures short term progress in yield potential. The intermediate level
would involve genetically distant but adapted crosses between market classes, growth habits,
maturity groups, and races, requiring more time to obtain commercially accepted cultivars.
At base level, few restrictions would govern breeding activities, presenting opportunity to

introgress wild or unadapted genetics requiring several crosses to achieve an adapted

phenotype.

Wild Beans to Improve Yield

Wild beans have been largely untouched as a genetic resource for increasing yield
(Singh et al., 1995). Wild germplasm contains alleles, especially near loci associated with
the domestication syndrome, which are not represented in domesticated cultivars (Papa et al.,
2007). Some of these alleles have a positive effect on key traits such as yield, and should be
transferred to a cultivated background and exploited to increase yield. When working in
populations derived from crosses between domesticated and wild parents, the diversity of
progeny can become problematic. Due to the large variation in traits related to
domestication, alleles with relatively small effect are difficult or impossible to detect
(Acosta-Gallegos et al., 2007). The inbred backcross line (IBL) (Bliss, 1993) combined with
the advanced backcross-QTL (Tanksley and Nelson, 1996) method offers an opportunity to
overcome this barrier and identify minor loci that would otherwise go undetected. The IBL
method has been used extensively at the Centro Intemacional de Agricultura Tropical (CIAT,
Cali, Colombia) to introgress diversity from wild beans into domesticated germplasm (Beebe
et al., 2004). Progeny from these crosses have been widely tested in Mexico and Michigan
where some lines have surpassed all local checks and exceeded the yield of the domesticated
recurrent parent by up to 27% (Beaver et al., 2003; Kelly, 2004). These results have
stimulated further research to dissect the genetic architecture underlying the increased
performance of progeny derived from wild by domesticated crosses.

Quantitative Genetic Variation

Through phenotypic selection, bean breeders have made significant progress in
developing cultivars with superior performance across a range of environments by selecting
for agronomic traits such as yield, disease resistance, and improved plant architecture (Kelly,
2001; Beaver et al., 2003; Miklas et al., 2006). To ensure continued progress in

understanding these complexly inherited traits, phenotypic selection can be complemented by

7

the use of molecular markers to pinpoint specific locations within the genome that condition
individual genetic components of complex traits, along with their relative individual effects
(Beaver et al., 2003). Information from molecular analyses, combined with phenotypic
evaluation, will contribute to continued development of agronomically superior cultivars.

In crop plants, quantitative variation affects many important traits, including yield,
resistance to pathogens, and seed quality (Kelly et al., 1998; Miklas et al., 2006; Posa-
Macalincag et al. 2002). Therefore breeding requires a means to analyze and discern the
genetic basis of these complexly inherited traits. In the early twentieth century, Sax (1923)
developed concepts to study genes affecting quantitative traits. He demonstrated that
quantitative variation resulted from the combined effects of multiple genes with
environmental effects by investigating the co-segregation of bean seed size with Mendelian
markers for seed color and pattern (Gepts, 2001). Thoday (1961; as cited by Young, 1996)
suggested if the segregation of simply inherited genes could be used to detect linked QTLs,
eventually all QTLs involved in complexly inherited traits could be characterized.

However, until about 1980, quantitative traits analysis remained a statistical exercise
rather than a detailed analysis of individual gene effects. The assumption remained that
although multiple genes contributed to the expression of a quantitative trait, their individual
effects were relatively equal and allelic differences were minimal compared to the effect of
the environment. The varying effects of individual loci on the complex trait were largely
overlooked under this system. Using this minimal framework, substantial progress was made
in better understanding the genetics controlling many complex traits and making appropriate
selections based on this knowledge (Asins, 2002).

Dissecting the genetic control of complex traits requires a joint analysis of genotypic
and phenotypic data. Molecular linkage mapping and subsequent analysis of genetic regions

affecting quantitative traits, or QTL, has been a useful approach to determine the minimum

8

number of genes involved, their relative phenotypic effects, as well as linkages that may exist
between traits (Koinange et al., 1996).

Modern QTL analysis entails hybridizing parents that differ significantly in one or
more quantitatively inherited traits, followed by phenotyping and genotyping of the progeny.
Genotypic data are used to construct a linkage map of the population, which is subsequently
used as a framework to locate regions of the genome significantly associated with phenotypic
variation. Adjacent markers on the linkage map delineate the approximate location of the
QTL; consequently, shorter intervals between markers will result in a more precise estimate
of QTL size and location while indicating a tighter linkage between the markers and the
genetic locus (Collard et al., 2005).

Once the location of the QTL and tightly linked markers is determined, plant breeders
can use this information to make indirect selections based on the genotypes of the markers
linked to the QTL. A selection process where marker data contributes to the selection
decision is called marker-assisted selection (MAS), and has contributed to more rapid
introgression of favorable alleles for disease resistance and agronomic traits into improved
cultivars from various genetic backgrounds (Asins, 2002; Kelly et al., 2003; Kelly and.
Vallejo, 2006). These selections can be made despite environmental conditions that hinder
the expression and selection of the linked phenotype. The process may also be more efficient
and economical than phenotypic selection (Ender et al., 2007; Yu et al., 2000b). However,
genotypic selections must be verified for phenotype to eliminate false positive or negative
selections due to recombination between markers and their associated QTL (Collard et al.,

20051

Application of QTL Analysis and Marker Assisted Selection

QTL analysis, coupled with MAS, is a tool increasingly used by breeders to locate
genes associated with quantitative traits in plant genomes and select individuals containing
desirable combinations of those genes. This approach has the ability to overcome some of
the limitations encountered when selecting for quantitative traits by conventional phenotypic
selection (Blair et al., 2007). Using phenotypic selection alone, it can be difficult to identify
and select individuals that carry a series of different beneficial alleles influencing a
quantitative trait (Schneider et al., 1997; Tar’an et al., 2003). The situation is further
complicated if the desirable phenotype is masked by the presence of undesirable alleles,
which often occurs when genetic materials from the wild are introgressed. Tanksley and
Nelson (1996) proposed the advanced backcross QTL (ABC-QTL) analysis to introgress and
identify favorable alleles from wild relatives that are masked by unfavorable genotypes. This
technique was utilized by Gur and Zamir (2004) to improve yield in tomato by up 50% using
a wild species as a donor of favorable alleles. Similar improvements were obtained in rice
(Tanksley and McCouch, 1997) and in bean (Blair et al., 2006), demonstrating that
phenotypic selection can be enhanced by MAS.
Markers & Mapping

P. vulgaris is a diploid species that has 2n = 2x = 22 chromosomes. The 11
chromosomes are relatively small, and have all been identified (Cheng and Bassett, 1981).
Arumuganathan and Earle (1991) determined the genome size was 0.65pg/haploid genome or
635mbp, one of the smallest in the legume family. Pedrosa et al. (2003) assigned all 11
chromosomes to their respective linkage groups (LGs) using fluorescence in situ
hybridization.

To determine where genes are located in the genome, molecular linkage maps based

on molecular markers have been developed. These maps provide approximate locations of

10

individual loci relative to each other. A number of different marker types have been used for
different purposes and as new marker systems became available, they have often replaced
older systems that had inherent limitations. Gepts et al. (2008) recently reviewed advances in
marker technology for bean.

Biochemical Markers

Weeden (1984) first described allozymes in bean as a method to differentiate
cultivars based on genotype. These early biochemical markers were used to confirm the
geographic distribution of the wild common bean gene pools. Singh et al. (1991a) used
allozymes to definitively divide the two major gene pools of P. vulgaris into three races each.
Debouck et el. (1993) showed that in addition to the Middle American and Andean gene
pools, an ancestral gene pool exists in Ecuador and northern Peru that is distinct from the
other two gene pools. Seed protein markers (phaseolin) have also been used to characterize
diversity among beans and provide evidence for multiple bean domestications based on
differences in electrophoretic patterns (Gepts and Bliss, 1986). These early biochemical
markers were useful, although their limited genome coverage and level of polymorphism
imposed limitations on their application for characterizing genetic diversity in closely related
groups of beans.

Molecular Markers and Bean Linkage Maps

Molecular markers based on random variation in genomic sequences later became
available and expanded the application of genetic markers in bean breeding. Randomly
amplified polymorphic DNA (RAPD) markers (Welsh and McClelland, 1990; Williams et
al., 1990) have been widely used to tag and map disease resistance genes (Kelly, 1995) and in
linkage map construction. Amplified fragment length polymorphism (AF LP) markers (V 05
et al., 1995) are also based on arbitrary primer sequences and have been used widely for

mapping and to assess genetic diversity. A number of sequences linked to resistance genes

11

and amplified by RAPD or some by AF LP markers have been converted to sequence
characterized amplified region (SCAR) markers for use in resistance gene pyramiding (Kelly
et al., 2003; Miklas et al., 2006).

Restriction fragment length polymorphism (RF LP) markers were used as framework
markers in early linkage maps, which also integrated the information from the earlier
biochemical markers. Vallejos et al. (1992) constructed a linkage map based primarily on
RFLP markers and estimated the size of the bean genome at 1200 cM. Freyre et al. (1998)
published a consensus map of the 11 LGs of bean that integrated several previous maps
(Vallejos et al., 1992; Nodari et al., 1993; Adam-Blondon et al., 1994; Jung et al., 1996,
1997; Skroch et al., 1996) based on shared RF LP and RAPD markers. This map consisted of
550 RAPD, RFLP, SCAR, isozyme, and phenotypic markers, in addition to another 500
markers in common with the other bean maps, resulting in an average distance of 1-2cM
between adjacent markers (Kelly et al., 2003). Linkage maps also delineate the locations of
genes for phenotypic traits such as disease and insect resistance, seed size, color, storage
proteins, and pod color.

Yu et al. (2000a) developed the first 37 common bean simple sequence repeat (SSR)
markers, successfully assigned 15 of them to the Freyre et al. (1998) consensus map, and
determined that SSR sequences were abundant in common bean. SSR markers have an
advantage of being co-dominant, PCR based which allows automation, usually multi-allelic
and hyper-variable, randomly and uniformly distributed throughout the genome, and
accessible to multiple researchers as published primer sequences (Yu et al., 1999).

Since the introduction of SSRs for bean in 1999, additional markers utilizing a
number of different sources of sequence information have been developed (Gaitan-Solis et
al., 2002; Blair et al., 2003; Yaish and Vaiga, 2003; Guerra-Sanz, 2004; Caixeta et al., 2005;

Frei et al., 2005; Buso et al., 2006; Benchimol et al., 2007; Hanai et al., 2007; de Campos et

12

al., 2007; Grisi et al., 2007). A small portion of these markers have been mapped to the
consensus map, but most have not been widely utilized for mapping. Expanding the
consensus map, Blair et al. (2003a) constructed the first map of bean based solely on simple
sequence repeat (SSR) markers and then integrated those markers into the Freyre et al. (1998)
and Vallejos et al. (1992) maps.

Despite recent interest in development of SSR markers for common bean, the bean
genome has not been saturated so other marker systems must be used to construct an efficient
linkage map from genetically related mapping populations. Sequence-related amplified
polymorphism (SRAP) markers were originally developed as a simple, reliable, moderate
throughput, reproducible, dominant marker system for Brassica oleracea (Li and Quiros,
2001). These markers have been utilized in diverse crops such as potato, rice, lettuce,
Chinese cabbage, rapeseed, garlic, apple, citrus, and celery. The markers are based on pair-
wise combinations of 17 or 18 nucleotide long primer sequences that target genomic
sequences in open reading frames and have shown equivalent genome coverage as AF LP
markers in Brassica spp. (Li and Quiros, 2001).

Hu and Vick (2003) developed the target region amplification polymorphism (TRAP)
technique for use in Helianthus annuus. Miklas et al. (2006b) suggested tagging and
mapping common bean genes involved in disease resistance using TRAP markers. TRAP
markers use two primers of 18 nucleotides, one designed fi'om expressed sequence tag (EST)
sequence information, and the other of arbitrary sequence with either an AT- or GC-rich core
targeted to an intron or exon, respectively (Hu and Vick, 2003). These markers have been
used successfully for a number of crops and purposes, including fingerprinting of lettuce
cultivars (Hu et al., 2005), gene tagging in sunflower (Rojas-Barros et al., 2005), and QTL

mapping in a RIL population of wheat (Liu et al., 2005). In wheat, this marker system was

13

an efficient and robust technique to rapidly generate markers distributed across the genome
and proved as usefirl as SSR markers for assigning linkage groups to chromosomes.

Mapping and Tagging Genes and QTL

The development of a bean consensus map has facilitated comparison between
individual mapping and gene tagging studies. For example, clusters of resistance genes for
bean rust, anthracnose, common bacterial blight, and white mold have been detected on LGs
Bl, B4, B7, and B11 (summarized by Miklas et al., 2006a). Tagging these genes with
markers has allowed for indirect selection for disease resistance in both domestic and
overseas breeding programs. In addition to tagging major resistance genes, marker assisted
techniques have increased understanding of complexly inherited traits such as stress tolerance
(Schneider et al., 1997), root architecture (Beebe et al., 2006), and quantitative disease
resistance (Park et al., 2001; Miklas et al., 2007) through QTL analysis. Another group of
genes associated with the domestication syndrome of common bean that influences,
photoperiod insensitivity, lack of seed dormancy, seed color patterns, and increased seed size,
were identified and mapped to LG Bl (Koinange et al., 1996). A summary of the -
populations used for tagging and mapping a variety of genes and QTL between '1992 and
2004 was recently compiled by Miklas and Singh (2007). Mapping studies will continue to
be useful to identify additional genetic diversity fi'om wild or exotic germplasm introgressed
into domesticated beans.

Introgression of wild bean germplasm into cultivated backgrounds has received
considerable attention from scientists at CIAT, located in Cali, Colombia (Beebe et al.,
2003). Based on information available from molecular diversity analyses of diverse bean
germplasm (Tohme et al., 1996), a core collection was established and unique wild
accessions were incorporated into a breeding program in order to transfer genetic diversity

into a cultivated background for further analysis of desirable variation. Blair et al. (2006a)

14

identified 13 QTL associated with a wild Colombian bean that had a positive effect on plant
height, yield and yield components in a population derived from a cultivated by wild cross.
Guzman-Maldonado et al. (2003) identified 14 QTL and determined that a Mexican wild
bean contributed alleles that increased seed mass, content of Ca, Fe, Zn, and tannins in the
seed when crossed with a cultivated bean.

Studies using diverse genetic backgrounds from both gene pools of bean have
identified QTL for varied agronomic traits. Park et al. (2000) identified QTL for seed size
and shape. Tar’an et al. (2002) identified 14 QTL for yield and other agronomic traits.
Beattie et al. (2003) used QTL analysis to identify 21 genomic regions associated with
agronomic and architecture traits of a bean ideotype. Checa and Blair (2008) examined
climbing ability and identified 23 QTL for growth habit components. Tsai et al. (1998)
identified 6 QTL for nodule number involved in N-fixation.

Although QTL for disease resistance have shown practical application for MAS
(Miklas et al. 2006a), application of QTL studies for other polygenic traits has been limited
(Blair et al., 2007.). Complex agronomic traits targeted by QTL studies are often controlled
by many minor loci, rather than a few regions with major effect, which increases the
investment required to implement routine MAS.

The limited genome coverage of molecular maps in narrow intra-gene pool crosses in
bean typically used to generate elite germplasm further restricts the ability to detect QTL
related to major economic traits. However, markers can be useful in breeding programs even
if the application is associating a particular phenotype with a trait that can then be targeted
for phenotypic selection. This approach has been used to study nutrient uptake
characteristics of different root structures and determine which root features should be
selected to increase the efficiency of nutrient uptake in phosphorous deficient soils (Beebe et

al., 2006; Tsai et al., 1998; Yan et al., 2005).

15

Disease Resistance

Anthracnose

A number of diseases reduce the productivity of common bean, so an important part
of any breeding program involves resistance breeding. Among the diseases affecting bean
production, anthracnose, caused by Colletotrichum Iindemuthianum (Sacc. & Magnus) Briosi
& Cav., is considered the most serious disease of common bean worldwide (Kelly and
Vallejo, 2004). This status is largely related to the seed-bome nature of the disease, and
highly variable pathogenicity. Common bean and C. lindemuthianum co-evolved, leading to
both Andean and Middle American pathogen groups (reviewed by Pastor-Corrales, 2004).
This variability is classified by race with an internationally recognized binary code based on
the disease reaction of 12 differential host cultivars that each carry a binary code number
from 1 to 2048 (Pastor-Corrales, 1991). Under this system, a number is assigned to each
virulent race based on the summation of the binary numbers of those differential cultivars
that are susceptible to the race.

Genetic resistance is the most effective management strategy for dealing with bean
anthracnose. Resistance to anthracnose is conferred by twelve major independent genes,
denoted Co-I to C0-13 (Co-3/C0-9 are allelic), and follows the gene-for-gene theory. All but
00-8 behave as dominant genes, and various authors have demonstrated Co-I, C0-3, and C0-4
to be part of an allelic series at three different loci (Table 2, Kelly and Vallejo, 2004;
Goncalves-Vidigal et al., 2009). Co-I and C0-12 are the only resistance genes of Andean
origin, while the others originated in Middle American germplasm (Kelly and Vallejo, 2004;
Goncalves-Vidigal et al., 2008). Historically, various letters have been used to denote these
different resistance genes, but those designations have since been replaced with the Co

symbol followed by a number, as proposed by Kelly and Young (1996).

16

Common Bean Rust

Another significant disease affecting bean production is common bean rust, caused by
one of the most pathogenically variable rust fungi, Uromyces appendiculatus (PerszPers)
Unger (Stavely et al., 1994). Pathogenic races of U. appendiculatus in common bean were
first reported in 1935 (I-larter et al., 1935; as cited by Stavely et al., 1994). Due to the
variability of this pathogen, breeding for genetic resistance to rust has been complicated by
the rapid breakdown of major resistance genes deployed in new cultivars. Efforts to prolong
the life of currently known resistance genes include pyramiding of multiple genes and
incorporation of different resistance characteristics (specific, slow rusting, reduced pustule
size, age-dependent resistance, and pubescence) (Miklas et al., 2005).

The effectiveness of this strategy was confirmed in Honduras where a cultivar with
single gene resistance succumbed to rust infection but another cultivar with additional
resistance genes did not become infected with a newly emerging rust pathotype (Mmbaga et
al., 1996). Specific races of rust exhibit patterns of virulence that reflect the division between
the Andean and Middle American bean gene pools, suggesting a history of co-evolution
between the rust pathogen and its host (Pastor-Corrales, 2004; Acevedo et al., 2008).
Molecular analysis of the pathogen also confirms this pattern (Araya et al., 2004). Therefore
the strategy to manage resistance to a wide range of rust races has been to pyramid major Ur-
genes with overlapping resistance spectrums from both gene pools to provide durable rust
resistance across a range of environments (Miklas et al., 2005).

To classify the variability of the rust pathogen, Steadman et al. (2002) proposed a
new differential series of twelve bean cultivars, six each from the two gene pools. Each
cultivar in this series was assigned a binary value, with the two gene pools considered
separately. The sums of the binary values of the susceptible cultivars, determined for each of

the two gene pools, are used to assign a race number to an unknown isolate of the pathogen.

17

This classification system better reflects the gene pool differences of rust isolates and
resistance genes compared with the previously implemented differential series (Stavely et al.,
1983)

Nine named resistance genes (Ur-3, Ur-4, Ur-5, Ur-6, Ur- 7, Ur-9, Ur-I I, Ur-12, and
Ur-13) and four unnamed genes (one each in BAC6 (Ur-BAC6) and ‘Ouro Negro’ (Ur-ON),
two in ‘Dorado’ (Ur-Dorado-53, Ur-Dorado-108) have been characterized, tagged with
RAPD or SCAR markers, and mapped to five linkage groups (reviewed by Miklas et al.,
2006a). Recently, Pastor-Corrales et al. (2008) tagged and mapped an additional unnamed
gene from P1260418 to LG B4 that confers resistance to all but one known race of U.
appendiculatus.

Based on these results and additional inheritance studies, it also seems that rust
resistance genes are more clustered within the genome than anthracnose resistance genes
(Miklas et al., 2005; 2006). Some rust resistance genes have been characterized as clusters of
tightly linked loci as with Ur-5 which is inherited as a complex of single dominant genes
linked tightly in coupling (Stavely, 1984). Ur-3, which conditions slightly different reactions
to the rust pathogen depending on the resistant cultivar used as a source, may also consist of a
similar complex block of tightly linked genes (Miklas et al., 2006a).

Furthermore, Ur-5 appears to be in close proximity, if not linked to Ur-Dorado-108
resistance (Miklas et al., 2000). Ur-ON is independent of these genes, but also resides on the
same region of B4 (Alzate-Marin et al., 2004). A similar cluster resides on B11, where Ur-3
and Ur-I I are linked (Stavely, 1998), and Ur-Dorado-53 was mapped to the same region
(Miklas et al., 2000). Ur-6, Ur- 7, and Ur-BAC6 reside nearby on B11, but independent of
the Ur-3/Ur-11 cluster (Miklas et al., 2006a).

In addition to clustering of Ur-genes, resistance genes for anthracnose and rust co-

localize by gene pool of origin at several genomic locations. The Andean genes Co-I and

18

Ur-9 co-localize on B1, while Mesoamerican genes Co-3/Co-9, Ur-5, Co-10, and Ur-Dorado-
108 co-localize on B4. Additional Mesoamerican genes Co-2, Ur-3, Ur-l 1, and Ur-Dorado-
53 co-localize on B1] (Miklas et al., 2006a). These associations suggest a mechanism such
as duplication of ancestral gene sequences may have lead to these resistance gene clusters.
Geffroy et al. (1999) examined the molecular basis of the genome at the B4 cluster and found
the region was characterized by leucine-rich-repeats (LRRs) and possessed 11 resistance
gene analogs, supporting the hypothesis of ancestral gene duplication and divergence at
complex resistance clusters.

To date, all rust resistance genes characterized are dominantly inherited. Genes
identified from the Mesoamerican gene pool include Ur-3, Ur-5, Ur- 7, Ur-I I, Ur-Dorado-53,
Ur-Dorado-108, Ur-ON, and Ur-BAC6. These genes have conferred broader resistance to
different rust races than those from the Andean gene pool, which include Ur-4, Ur-6, Ur-9,
Ur-12, and Ur-13. (Kelly et al., 1996; Miklas et al., 2006a).

In temperate production areas of North America, Ur-3 has shown impressive
durability against the highly variable pathogen U. appendiculatus (Singh, 2005). However,
this resistance could inevitably break down in the future, so additional information about
relationships among resistance genes and more precise map locations will be needed to breed
future rust resistant cultivars (Kelly et al., 2003).

Processing Quality

In addition to the agronomic traits, seed quality traits are also scrutinized by bean
breeders. New cultivars must possess acceptable color, texture, and visual appearance when
canned. A favorable combination of these traits is critical in determining whether a cultivar
will be accepted by consumers (Hosfield and Uebersax, 1991). Commercial bean canners are
constrained by these expectations, and additionally require beans with rapid, uniform

hydration and a high water holding capacity. These characteristics ensure an efficient

19

canning process and result in increased washed-drained weight, therefore increasing
processor yield (Hosfield, 1991). Cultivars that consistently fail to maintain a desirable
color, texture, and visual appearance after the canning process may be discarded despite their
agronomic merits.

Color

Consumers have specific preferences about the color and appearance of canned beans
(Hosfield, 1991). Pigments in the seed coat of beans determine the absorption and
reflectance of different wavelengths of light. During canning these pigments, especially
anthocyanins, leach out of the bean and into the brine, which results in black beans that
appear brown and unappealing (Bushey et al., 2000). Color of dry or canned beans can be
measured on the Hunter L-scale, where 1=pure black and 100=pure white using the Hunter
Lab Color and Color Difference meter (Hunter Laboratories, Reston, VA).

Texture

Texture is measured to quantify the consumer perception of chewing the cooked bean
product (Ghaderi et al., 1984). This attribute is measured using a shear press in terms of kg
force applied to a 100g sample of canned beans at a constant rate (I-losfield and Uebersax,
1980). An increased force required to shear a sample of beans corresponds to increased bean
firmness (Bolles et al., 1990). A desirable texture for black beans is 45-75 kg force, and
samples with texture measurements beyond this range may be perceived when chewed as
being too soft or firm (BIC, 2008).

Visual Appearance and Washed-Drained Weight

Visual appearance is a subjective rating that considers the sum of individual quality
components such as color, clumps, and splits, as well as the starchiness and consistency of
the brine. Visual appearance provides a general index of a cultivar’s suitability for

commercial canning referenced to cultivars with demonstrated quality attributes (Hosfield

20

and Uebersax, 1984). Visual appearance correlates positively with texture, but negatively
with washed-drained weight in navy beans (Walters et al., 1997). Although high washed-
drained weight increases processor yield, the volume of canned beans produced from a given
dry weight, the negative correlation indicates increasing this value excessively may decrease
consumer acceptance.

Inheritance of Canning Quality Components

Each individual component of canning quality is moderately to highly heritable, but
behaves in a complex, quantitative manner (I-Iosfield et al., 1984; Wassimi et al., 1990;
Walters et al., 1997). In addition to this complexity, environmental effects such as location
(Ghaderi et al., 1984; Shellie and Hosfield, 2001) or year (Hosfield et al., 1984) interact with
individual components of canning quality in some studies while others have shown
insignificant interactions (Wassimi et al., 1990). The effect of location or year may influence
these components more than genotype in some seasons (Walters et al., 1997).

Due to the inherent environmental effects associated with evaluating canning quality
traits, MAS for individual components has been proposed. This technique has been used
successfully to improve disease resistance and abiotic stress tolerance (Kelly et al., 2003;
Miklas et al., 2006). Previous attempts to identify QTL associated with canning quality traits
have shown varied results. Walters et al. (1997) identified a group of RAPD markers
associated with visual appearance, texture, and washed-drained weight in navy bean, but also
found many of these associations were location and population specific, limiting their
widespread use in breeding programs.

Posa-Macalincag et al. (2002) later screened two populations of red kidney beans
with the same RAPD markers, but could not significantly associate them to any quality trait.
Instead, two different QTL were detected, each associated with both visual appearance and

splitting, and in different genomic regions than the markers identified by Walters et al.

21

(1997). These QTL were also population and environment specific. One QTL was located on
B8 of the bean core map, which has been previously been associated with seed traits
including the C locus for seedcoat pattern (McClean et al., 2002), the R gene for dominant
red seedcoat pattern (Miklas et al., 2000; Bassett, 1998) and QTL for seed size and shape
(Park et al., 2000). The other QTL was detected for the same traits in a different population,
but on a different linkage group not aligned with the core map. These results suggest
population, environment, and gene pool specificity of markers associated with seed quality
traits, which underscores the difficulty in identifying reliable markers that are useful across a
wide range of genetic backgrounds and seed types.

Indirect Screening Methods for Canning Quality

Black beans are especially prone to loss of seed color during the canning process and
may appear brown rather than black, making them visually unappealing to consumers.
Current canning methods require a minimum of three years to generate a sufficient quantity
of seed before the first evaluation can be made for canning quality (Bushey and Hosfield,
2007a). Limited work has been undertaken to develop an informative early-generation
screen to allow processing evaluation at an earlier generation in the breeding process.
Ruengsakulrach et al. (1991) approached this problem both directly by canning a smaller
sample (28.5 g solids) and indirectly by correlating pasting torque values of whole bean flour
(2g sample) to shear texture values. Those methods were both successful in predicting
processing quality of later generations. Lu et al. (1996) examined the chemical composition
of navy beans and found a significant correlation between soluble-pectin content measured in
a small quantity of seed and visual score of a much larger canned sample of the same variety.
Bushey and Hosfield (2007b) soaked a small quantity of black beans in a hot brine to
simulate the blanching that occurs prior to canning, and then measured the color of the brine

to predict color loss during canning. Although several of these methods showed good

22

correlation to canned bean quality, they still require a substantial investment of time and skill
to complete, and have not been adopted by the bean community.
Conclusion

After review of previous work related to common bean agronomic and quality traits,
the present study was undertaken to further dissect key economic traits by studying
quantitative variation for yield and canning quality traits in a recombinant inbred line
population of black beans. The goal of this work was to identify genomic regions associated
with these traits that could be utilized by bean breeders working to enhance yield while

maintaining canning quality of future black bean cultivars.

23

Literature Cited

Acosta-Gallegos, J .A., J .D. Kelly, and P. Gepts. 2007. Prebreeding in common bean and use
of genetic diversity from wild germplasm. Crop Sci. 47: 844-59.

Acevedo, M., J .R. Steadman, J .C. Rosas, and J. Venegas. 2008. Coevolution of the bean rust
pathogen Uromyces appendiculatus with its wild, weedy, and domesticated hosts at a
center of diversity. Annu. Rep. Bean Improv. Coop. 51: 22-23.

Adam-Blondon A., M. Sevignac, and M. Dron. 1994. A genetic map of common bean to
localize specific resistance genes against anthracnose. Genome 37: 915-924.

Alzate-Marin, A.L., T.L.P.O. Souza, V.A. Ragagnin, M.A. Moreira, and E.G. Barros. 2004.
Allelism tests between the rust resistance gene present in common bean cultivar
‘Ouro Negro’ and genes Ur-5 and Ur-II. Phytopathology 152160-64.

Anderson, J .W., L. Story, B. Sieling, W.J. Chen, M.S. Petro, and J. Story. 1984.
Hypocholesterolemic effects of oat-bran or bean intake for hypercholesterolemic
men. Amer. J. Clin. Nutr. 40:1146-1155.

Araya, C.M., A.T. Alleyne, and J .R. Steadman. 2004. Phenotypic and genotypic

characterization of Uromyces appendiculatus from Phaseolus vulgaris in the
Americas. Plant Dis. 88: 830-836.

Asins, MJ. 2002. Present and future of quantitative trait locus analysis in plant breeding.
Plant Breed. 121:281-291.

Bassett, M.I. 1998. A test cross protocol for determining the genotype of dark red seedcoat
colors in common bean. J. Amer. Soc. Hort. Sci. 123:1048-1052.

Beattie, A.D., J. Larsen, T.E. Michaels, and KP. Pauls. 2003. Mapping quantitative trait
loci for a common bean (Phaseolus vulgaris L.) ideotype. Genome 46: 411—422.

Beaver, J .S., J .C. Rosas, J. Myers, J. Acosta, J .D. Kelly, S. Nchimbi-Msolla, R. Misangu, J.
Bokosi, S. Temple, E. Amaud-Santana, and DP. Coyne. 2003. Contributions of the

Bean/Cowpea CRSP to cultivar and germplasm development in common bean. Field
Crops Res. 82:87-102.

Beavis, W.D. 1998. QTL analyses: Power, precision, and accuracy. In: A.H. Patterson (ed.),
Molecular dissection of complex traits, 145-162. CRC Press, Boca Raton, FL.

Becerra Velasquez, V.L., and P. Gepts. 1994. RF LP Diversity in Common Bean (Phaseolus
vulgaris L.). Genome 37:256-263.

Beebe, S., O. Toro, A.V. Gonzalez, M.I. Chacon, and D.G. Debouck. 1997. Wild-weed-crop
complexes of common bean (Phaseolus vulgaris L., F abaceae) in the Andes of Peru

and Colombia, and their implications for conservation and breeding. Genet. Resour.
Crop Evol. 44: 73-91.

24

Beebe, S., P.W. Skroch, J. Tohme, M.C. Duque, F. Pedraza, and J. Nienhaus. 2000. Structure
of genetic diversity among common bean landraces of Middle American origin based
on correspondence analysis of RAPD. Crop Sci. 40: 264-273.

Beebe, S., J. Rengifo, E. Gaitan, M.C. Duque, and J. Tohme. 2001. Diversity and origin of
Andean landraces of common bean. Crop Sci. 41:854-862.

Beebe, S., J. Tohme, J. Nienhuis, F. Pedraza, J. Rengifo, E. Tovar, and A. Islam. Studies in
Phaseolus germplasm diversity: a review of work at CIAT. 2004. Annu. Rep. Bean
Improv. Coop. 47: 33-34.

Beebe, S., M.Rojas-Pierce, X. Yan, M.W. Blair, F. Pedraza, F. Munoz, J. Tohme, and J. P.
Lynch. 2006. Quantitative trait loci for root architecture traits correlated with
phosphorous acquisition in common bean. Crop Sci.46:413-423.

Benchimol, L.L., T. Campos, S.A.M. Carbonell, C.A. Colombo, A.F. Chioratto, E.F.
Formighieri, L.R.L. Gouvea, and AP. Souza. 2007. Structure of genetic diversity ,
among common bean (Phaseolus vulgaris L.) varieties of Mesoamerican and Andean

origins using newly developed microsatellite markers. Genet. Resour. Crop Evol. 54:
1747-1762.

BIC. 2008. [Online] Bean Processing. Verified July 21, 2008: http://www.css.msu.edu/bic/

Blair, M.W., F. Pedraza, H.F. Buendia, E. Gaitan-Solis, S.E. Beebe, P. Gepts, and J. Tohme.
2003a. Development of a genome-wide anchored microsatellite map for common
bean (Phaseolus vulgaris L.). Theor. Appl. Genet. 107:1362-1374.

Blair, M.W., G. Iriarte, and S. Beebe. 2003b. QTL Analysis of an Andean advanced
backcross population for yield traits derived from wild P. vulgaris. Annu. Rep. Bean
Improv. Coop. 46:7-8.

Blair, M.W., G. Iriarte, and S. Beebe. 2006a. QTL analysis of yield traits in an advanced
backcross population derived from a cultivated Andean x wild common bean
(Phaseolus vulgaris L.) cross. Theor. Appl. Genet. 112:1149-1163.

Blair, M.W., M.C. Giraldo, H.F. Buendia, E. Tovar, M.C. Duque, and S. E. Beebe. 2006b.
Microsatellite marker diversity in common bean (Phaseolus vulgaris L.). Theor.
Appl. Genet. 113: 100—109.

Blair, M.W., M.A. Fregene, S.E. Beebe, and H. Ceballos. 2007. Marker-assisted selection in
common beans and cassava. In: E. Guimaraes, J. Ruane, B. Scherf, A. Sonnino, and
J. Dargie (eds.), Marker Assisted Selection: Current status and future perspectives in
crops, livestock, forestry, and fish, 81-115. FAO, Rome.

Bolles, A.D., M.A. Uebersax, G.L. Hosfield, and RC. Hamelink. 1982. Textural parameters

derived from shear curves of processed dry edible beans. Mich. Dry Bean Digest 6:
21-23.

25

Broughton, W.J., G. Hernandez, M. Blair, S. Beebe, P. Gepts, and J. Vanderleyden. 2003.
Beans (Phaseolus spp.) — model food legumes. Plant Soil 252: 55-128.

Bushey, S.M., L.S. Harris, and G.L. Hosfield. 2004. Development of an early generation test
for predicting color loss of black beans. Annu. Rep. Bean Improv. Coop. 47:139-140.

Bushey, SM. and G.L. Hosfield. 2007a. Absorbance of the soak water to predict canning
quality. Annu. Rep. Bean Improv. Coop. 50:39-40.

Bushey, SM. and G.L. Hosfield. 2007b. A test to predict color loss in black bean during
thermal processing. Annu. Rep. Bean Improv. Coop. 50:41-42.

Buso, G.S.C., Z.P.S. Amara], R.P.V. Brondani, and ME. Ferreira. 2006. Microsatellite
markers for the common bean Phaseolus vulgaris. Mol. Ecol. Notes 6: 252-254.

Campos, T., L.L. Benchimol, S.A.M. Carbonell, A.F. Chioratto, E.F. Formighieri, and A.
Pereira de Souza. 2007. Microsatellites for genetic studies and breeding programs in
common bean. Pesq. Agropec. Bras. 42: 589-592.

Chacon, M.I., S.B. Pickersgill, and D.G. Debouck. 2005. Domestication patterns in common

bean (Phaseolus vulgaris L.) and the origin of the Mesoamerican and Andean
cultivated races. Theor. Appl. Genet. 110:432-444.

Charles, D. 2006. A forever seed bank takes root in the arctic. Science 312:1730-1731.

Debouck, D.G., R.T. Herrera, and R.V. Araya. 2008. New populations of wild common
bean disclosed in Nicaragua. Annu. Rep. Bean Improv. Coop. 51: 120-121.

Debouck, D.G., O. Toro, O.M. Paredes, W.C. Johnson, and P. Gepts. 1993. Genetic diversity
and ecological distribution of Phaseolus vulgaris in northwestern South America.
Econ. Bot. 47:408-423.

FAO. 2006. FAOSTAT [Online] Available at http://faostat.fao.org (verified July 23, 2008).

Freyre, R., P.W. Skroch, V. Geffroy, 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 an integrated linkage map of
common bean. 4. Development of a core linkage map and alignment of RFLP maps.
Theor. Appl. Genet. 97: 847-856.

Gaitan-Solis, M.C. Duque, K.J. Edwards, and J. Tohme. 2002. Microsatellite repeats in
common bean (Phaseolus vulgaris): Isolation, characterization, and cross-species
amplification in Phaseolus spp. Crop Sci. 42: 2128-2136.

Geffroy, V., D. Sicard, J .C. F. de Oliveira, M. Sevignac, S. Cohen, P. Gepts, C. Neema, T.
Langin, and M. Dron. 1999. Identification of an ancestral resistance gene cluster
involved in the coevolution process between Phaseolus vulgaris and its fungal
pathogen Colletotrichum lindemuthianum. Mol. Plant-Microbe Interact. 12: 774-784.

26

Gepts. P. 2001. [Online] Phaseolus vulgaris (Beans). Available at
http://www.agronomy.ucdavis.edu/gepts/a1749.pdf (verified July 23, 2008).

Gepts, P. 2005. Management of germplasm and prebreeding in common bean. Anais /
CONAF E, VIII Congresso Nacional de Pesquisa de F eijd'o, Goia‘nia, GO,
18 a 20 de outubro de 2005. - Santa Antonio de Goids .' Embrapa Arroz e F eijfio,
2005.

Gepts, P., and RA. Bliss. 1986. Phaseolin variability among wild and cultivated common
beans (Phaseolus vulgaris L) from Colombia, Econ. Bot. 40:469-478.

Gepts, P., F .J .L. Aragao, E. Barros, M.W. Blair, R. Brondani, W. Broughton, I. Galasso, G.
Hernandez, J. Kami, P. Lariguet, P. McClean, M. Melotto, P. Miklas, P. Pauls, A.
Pedrosa-Harand, T. Porch, F. Sanchez, F. Sparvoli, and K. Yu. 2008. Genomics of
Phaseolus beans, a major source of dietary protein and micronutrients in the tropics.
In: P.H. Moore, R. Ming (eds.), Genomics of Tropical Crop Plants, 113-143.
Springer.

Gepts, P., and D. Debouck. 1991. Origin, domestication, and evolution of common bean
(Phaseolus vulgaris L.). In: A. Van Schoonhoven and O. Voysest (eds.) Common
beans: Research for crop improvement, 7-53. CABI, Wallingford, UK, and CIAT,
Cali, Colombia.

Ghaderi, A., G.L. Hosfield, M.W. Adams, and M.A. Uebersax. 1984. Variability in culinary
quality, component interrelationships, and breeding implications in navy and pinto
beans. J. Amer. Soc. Hort. Sci. 109:85-90.

Gomez, 0. 2004. Evaluation of Nicaraguan common bean (Phaseolus vulgaris L.)
landraces. Doctoral diss. Dept. of Ecology and Crop Production Science, SLU. Acta
Universitatis Agriculturae Sueciae. Agraria vol. 476.

Gomez, 0. , M.W. Blair, B.E. F rankow-Lindberg, and U. Gullberg. 2004. Molecular and
phenotypic diversity of common bean landraces from Nicaragua. Crop Sci. le412-
1418.

Goncalves-Vidigal, M.C., G. F. Lacanallo and P. S. Vidigal Filho. 2008. A new gene conferring
resistance to anthracnose in Andean common bean (Phaseon vulgaris L.) cultivar Jalo
Vermelho. Plant Breed. doi: 10.1 1 11/j.1439-0523.2008.01530.x.

Goncalves-Vidigal, M.C., P.S. Vidigal F ilho, A.F Medeiros, and M.A. Pastor-Corrales. 2009.
Common bean landrace J alo Listras Pretas is the source of a new Andean anthracnose

resistance gene. Crop Science. In Press.

Guerra-Sanz, J.M. 2004. New SSR markers of Phaseolus vulgaris from sequence databases.
Plant Breed. 123: 87-89.

Gur, A. and D. Zamir. 2004. Unused natural variation can lift yield barriers in plant breeding.
PLOS Biol. 2: e245.

27

Guzman-Maldonado, SH, 0. Martinez, J. A. Acosta-Gallegos, F. Guevara-Lara, and O.
Paredes-Lopez. 2003. Putative quantitative trait loci for physical and chemical
components of common bean. Crop Sci. 43: 1029-1035.

Hanai, L.R., T. de Campos, L.E.A. Camargo, L.L. Benchimol, A.P. Souza, M. Melotta,
S.A.M. Carbonell, A.F. Chioratto, L. Consoli, E.F. Formighieri, M.V.B.M. Siqueira,
S.M. Tsai, and M.L.C. Vieira. 2007. Development, characterization, and comparative

analysis of polymorphism at common bean SSR loci isolated fi'om genic and genomic
sources. Genome 50: 266-277.

Hang, A.N. 2004-06. [Online] National Cooperative Dry Bean Nursery. Available at:
http://www.prosser.wsu.edu/faculty/hang.html (verified July 23, 2008).

Hangen, L., and MR. Bennink. 2003. Consumption of black beans and navy beans
(Phaseolus vulgaris) reduced azoxymethane-induced colon cancer in rats. Nutr.
Cancer 44:60-65.

Harter, L.L., C.F. Andrus, and W.J. Zaumeyer. 1935. Studies on bean rust caused by
Uromyces phaseoli typical. J. Agric. Res. 50: 737-759.

Hosfield, G.L. 1991. Genetic control of production and food quality factors in dry bean.
Food Tech. 45: 98-103.

Hosfield, G.L., A. Ghaderi, and M.A. Uebersax. 1984. A factor analysis of yield and sensory
and physico-chemical data from tests used to measure culinary quality in dry edible
beans. Can. J. Plant Sci. 64:285-293.

Hosfield, G.L. & M.A. Uebersax. 1991. The definition and measurement of culinary quality
in dry bean. Mich. Dry Bean Digest 14: 20-22, 27.

Jung, G., D.Coyne, P. Skroch, J. Nienhuis, E. Amaud-Santana, J. Bokosi, H. Ariyarathne, J.
Steadman, J. Beaver, and S. Kaeppler. 1996. Molecular markers associated with
plant architecture and resistance to common blight, web blight and rust in common
beans. J. Amer. Soc. Hort. Sci. 121:794-803.

Jung, G., P. Skroch, D. Coyne, J. Nienhuis, H. Ariyarathne, S. Kaeppler, and M. Bassett.
1997. Molecular-marker-based genetic analysis of tepary-bean-derived common

bacterial blight resistance in different developmental stages of common bean. J.
Amer. Soc. Hort. Sci. 122:329-337.

Kami, J ., B. Becerra Velasquez, D.G. Debouck and P. Gepts. 1995. Identification of
presumed ancestral DNA sequences of phaseolin in Phaseolus vulgaris. Proc. Natl.

Acad. Sci. 92:1101-1104.

Kaplan, L. and T.F. Lynch. 1999. Phaseolus (Fabaceae) in archaeology: AMS radiocarbon
dates and their significance for pre-Colombian agriculture. Econ. Bot. 53:261-272.

28

Kelly, J .D. 1995. Use of Random Amplified Polymorphic DNA markers in breeding for
major gene resistance to plant pathogens. HortScience 30:461-465.

Kelly, J .D. 2000. Remaking bean plant architecture for efficient production. Adv. Agron.
71:109-143.

Kelly, J .D. 2004. Advances in common bean improvement: some case histories with broader
applications. Acta Hort. 637:99-122.

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.

Kelly, J ., J. Kolkrnan, and K. Schneider. 1998. Breeding for yield in dry bean (Phaseolus
vulgaris L.). Euphytica 102:343-356.

Kelly, J .D. and VA. Vallejo. 2004. A comprehensive review of the major genes conditioning
resistance to anthracnose in common bean. HortScience 39:1196-1207.

Kelly, J .D., and RA. Young. 1996. Proposed symbols for anthracnose resistance genes.
Annu. Rep. Bean Improv. Coop. 39:20-24.

Kelly, J .D., J .R. Stavely, and RN. Miklas. 1996. Proposed symbols for rust resistance
genes. Annu. Rep. Bean Improv. Coop. 39:25-31.

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.

Komegay, J ., C. Cardona, and CE. Posso. 1993. Inheritance of resistance to Mexican bean

weevil in common bean, determined by bioassay and biochemical tests. Crop Sci.
33:589-594.

Kosambi, DD. 1944. The estimation of map distances from recombination values. Ann.
Eugen. 12:172—175.

Li, G. and CF. Quiros. 2001. Sequence-related amplified polymorphism (SRAP), a new
marker system based on a simple PCR reaction: its application to mapping and gene
tagging in Brassica. Theor. Appl. Genet. 103:455-461.

Lioi, L., A.R. Piergiovanni, D. Pignone, S. Puglisi, M. Santantonia, G. Sonnante. 2005.

Genetic Diversity of some surviving on-farm Italian common bean (Phaseolus
vulgaris L.) landraces. Plant Breed. 124: 576-581.

Lippman, Z., and SD. Tanksley. 2001. Dissecting the genetic pathway to extreme fruit size

in tomato using a cross between the small-fruited wild species Lycopersicon
pimpinellifolium and L. esculentum var. Giant Heirloom. Genetics 158:413-422.

29

Logozzo, G., R. Donnoli, L. Macaluso, R. Papa, H. Knupffer, and RS. Zeuli. 2007.
Analysis of the contribution of Mesoamerican and Andean gene pools to European
common bean (Phaseolus vulgaris L.) germplasm and strategies to establish a core
collection. Genet. Resour. Crop Evol. 54:1763-1779.

Lu, W., and KC. Chang. 1996. Correlations between chemical composition and canning
quality attributes of navy bean (Phaseolus vulgaris L. ). Cereal Chem. 73:785-787.

Ma, Y. and F .A. Bliss. 1978. Seed proteins of common bean. Crop Sci. 17:431—437.

McClean, P.E., R.K. Lee, C. Otto, P. Gepts, and M.J. Bassett. 2002. Molecular and
phenotypic mapping of genes controlling seed coat pattern and color in common bean
(Phaseolus vulgaris L.) J. Hered. 93: 148-152.

Melotto, M. and Kelly, J .D. 2000. An allelic series at the Co-I locus conditioning resistance
to anthracnose in common bean of Andean origin. Euphytica 116: 143-149.

Melotto, M., Balardin, RS, and JD. Kelly. 2000. Host-pathogen interaction and variability
of Colletotrichum lindemuthianum. In: Prusky, D., S. Freeman, and MB. Dickrnan
(eds.), Colletotrichum Host Specificity, Pathology, and Host-Pathogen Interaction
346-361. APS Press, St. Paul, MN.

Miklas, P.N., V. Stone, M.J. Daly, J .R. Stavely, J .R. Steadman, M.J. Bassett, R. Delorme,
and J .8. Beaver. 2000. Bacterial, fungal, and viral disease resistance loci mapped in

a recombinant inbred common bean population ('Dorado'/XAN 176). J. Amer. Soc.
Hort. Sci. 125: 476-481.

Miklas, RN. 2006. Molecular Advances with Common Bean Rust Resistance. Annu. Rep.
Bean Improv. Coop. 49: 47-48.

Miklas, P.N., J .D. Kelly, S.E. Beebe, and M.W. Blair. 2006a. Common bean breeding for
resistance against biotic and abiotic stresses: From classical to MAS breeding.
Euphytica 147: 105-131.

Miklas, P.N., J. Hu, N.J. Grfinwald, and KM. Larsen. 2006b. Potential application of TRAP
(targeted region amplified polymorphism) markers for mapping and tagging disease
resistance traits in common bean. Crop Sci. 46:910—916.

Miklas, P.N., and SP. Singh. 2007. Common Bean. In: C. Kole (ed). Pulses, Sugar and
Tuber Crops Vol 3. Genome mapping and molecular breeding in plants, 1-31.
Springer-Verlag, Berlin.

Miklas P.N., K.M. Larsen, K.A. Terpstra, D.C. Hauf, K.F. Grafton, and J .D. Kelly. 2007.

QTL analysis of ICA Bunsi-derived resistance to white mold in a pinto x navy bean
cross. Crop Sci 47:174-179.

Mmbaga, M., J .R. Steadman, and J .R. Stavely. 1996. The use of host resistance in disease
management of rust in common bean. IPM Rev. 1: 191-200.

30

Mumba, LE, and N.W. Galwey. 1999. Compatibility between wild and cultivated common
bean (Phaseolus vulgaris L.) genotypes of the Mesoamerican and Andean gene
pools: Evidence from the inheritance of quantitative characters. Euphytica 108: 105-
119.

Nodari R.O., S.M. Tsai, RL. Gilbertson, and P. Gepts. 1993. Towards an integrated linkage
map of common bean. II. Development of an RFLP-based linkage map. Theor. Appl.
Genet. 85: 513-520.

Papa, R., E]. Acosta, E.A. Delgado-Salinas, and P. Gepts. 2005. A genome-wide analysis of
differentiation between wild and domesticated Phaseolus vulgaris fiom
Mesoamerica. Theor. Appl. Genet. 111: 1147-1158.

Papa, R., Bellucci, E., Rossi, M., Leonardi, S., Rau, D., Gepts, P., Nanni, L., and G. Attene.
2007. Tagging the signatures of domestication in common bean (Phaseolus vulgaris)
by means of pooled DNA samples. Ann. Bot. 100: 1039-105 1 .

Park, 8.0., D.P. Coyne, G. Jung, P.W. Skroch, E. Amaud-Santan, J .R. Steadman, H.M.
Ariyaranthe, and J. Nienhaus. 2000. Mapping of QTL for seed size and shape traits
in common bean. J. Amer. Soc. Hort. Sci. 125:466-475.

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

Pastor-Corrales, M.A. 1991. Estandarizacion de variedades diferenciales y de designacion de
razas de Colletotrichum lindemuthianum. Phytopathology 81:694.

Pastor-Corrales, M.A. 2004. Review of Coevolution Studies between pathogens and their
common bean hosts: Implications for the development of disease resistant beans.
Annu. Rep. Bean Improv. Coop. 47: 67-68.

Posa—Macalincag, M.T., G.L. Hosfield, K.F. Grafton, M.A. Uebersax, and J .D. Kelly. 2002.
Quantitative trait loci (QTL) analysis of canning quality traits in kidney bean
(Phaseolus vulgaris L.). J. Amer. Soc. Hort. Sci. 127:608-615.

Ruengsakulrach, S., N. Srisuma, M.A. Uebersax, G.L. Hosfield, and LG. Occena. Early
generation screening of navy bean breeding lines by canning quality assessment and
pasting characteristics of bean flour. 1994. J. Food Qual. 17:321-333.

Sax, K. 1923. Association of size differences with seed-coat pattern and pigmentation in
Phaseolus vulgaris. Genetics 8: 552-560.

Schneider, K.A., M.E. Brothers, and J .D. Kelly. 1997. Marker assisted selection to improve
drought resistance in common bean. Crop Sci. 37: 51-60.

Shellie, KC, and G.L. Hosfield. 1991. Genotype X environmental effects on food quality of

common bean: resource-efficient testing procedures. J. Amer. Soc. Hort. Sci.
116:732-736.

31

Singh, S.P., P. Gepts, and D.G. Debouck. 1991a. Races of common bean (Phaseolus
vulgaris, Fabaceae). Econ. Bot. 45:379-396.

Singh, S., A. Gutierrez, A. Molina, C. Urrea, and P. Gepts. 1991b. Genetic diversity in
cultivated common bean. II. Marker-based analysis on morphological and agronomic
traits. Crop Sci. 31 :23—29.

Singh, S.P., A. Molina, and P. Gepts. 1995. Potential of wild common bean for seed yield
improvement of cultivars in the tropics. Can. J. Plant Sci. 75:807-813.

Singh, SP. 1999. Production and utilization. In: Singh SP (ed), Common Bean Improvement
in the Twenty-First Century, 1-24. Kluwer, Dordrecht, the Netherlands.

Singh, SP. 2001. Broadening the genetic base of common beans cultivars: A review. Crop
Sci. 41:1659—1675.

Singh, SP. 2005. Common bean (Phaseolus vulgaris L.). In: Singh, RP and PP. Jauhar
(eds.), Genetic Resources, Chromosome Engineering, and Crop Improvement: Grain
Legumes 11-47. CRC Press, Boca Raton, FL.

Skroch, P., G. Jung, J. Niehuis, and D. Coyne. 1996. Integration of RAPD marker linkage
maps and comparative mapping of QTL for disease resistance in common bean.
Annu. Rep. Bean Improv. Coop 39:48-49.

Sonnante, G., T. Stockton, R.O. Nodari, V.L. Beccera Velasquez, and P. Gepts. 1994.
Evolution of genetic diversity during the domestication of common bean (Phaseolus
vulgaris L.). Theor. Appl. Genet. 89: 629-635.

Stavely, J .R. 1998. Recombination of two major dominant rust resistance genes that are
tightly linked in repulsion. Annu. Rep. Bean Improv. Coop. 41 :17—18.

Stavely, J .R., G.F.Freytag, J .R.Steadman, and H.F.Schwartz. 1983. The 1983 bean rust
workshop. Annu. Rep. Bean Improv. Coop. 26:iv-vi.

Stavely, J .R., J.D. Kelly, and KR Grafton. 1994. BelMiDak-rust resistant navy dry beans
germplasm lines. HortScience 29:709-711.

Steadman, J .R., M.A. Pastor-Corrales and J .S. Beaver. 2002b. An overview of the 3rd bean
rust and 2nd bean common bacterial blight international workshops, March 4-8,
2002, Pieterrnaritzburg, South Africa. Annu. Rpt. Bean Improv. Coop. 45:120-124.

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

Tar’an, B., T.E. Michaels, and KP. Pauls. 2002. Genetic mapping of agronomic traits in
common bean. Crop Sci. 42:544-556.

Thoday, J.M. 1961. Location of polygenes. Nature 191:368-370.

32

Tiranti, V. and V. Negri. Selective microenvironmental effects play a role in shaping genetic
diversity and structure in a Phaseolus vulgaris L. landrace: implications for on-farm
conservation. Mol. Ecol. 16:4942-4955.

Tohme, J ., D. Orlando-Gonzales, S. Beebe, and M.C. Duque. 1996. AFLP analysis of gene
pools of a wild bean core collection. Crop Sci. 36:1375-1384.

Tsai, S.M., R.O. Nodari, D.H. Moon, L.E.A. Camargo, R. Vencovsky, and P. Gepts. 1998.
QTL mapping for nodule number and common bacterial blight in Phaseolus vulgaris
L. Plant Soil 204: 135-145.

USDA-ERS. 2008. [Online] Vegetables and Melons Outlook/VGS-325/February 20, 2008.
Available at http://www.ers.usda.gov/briefing/DryBeans/PDFs/DBnOutlook.pdf
(verified July 23, 2008).

Vallejos, E.C., N.S. Sakiyama, and CD. Chase. 1992. A molecular marker based linkage
map of Phaseolus vulgaris L. Genetics 131: 733-740.

Vavilov, NJ. 1940. The new systematics of cultivated plants. In: J. Huxley (ed.) The New
Systematics, 549-566. Clarendon, Oxford.

Vos, P. R. Hogers, M. Bleeker, M. Reijans, T. Van de Lee, M. Homes, et a1. 1995. AFLP: a
new technique for DNA fingerprinting . Nucleic Acids Res. 23:4407-4414.

Voysest, 0., M.C. Valencia, and M.C. Amezquita. 1994. Genetic diversity among Latin

American Andean and Mesoamerican common bean cultivars. Crop Sci. 34:1100-
] 110.

Walters, K.J., G.L. Hosfield, M.A. Uebersax, and J .D. Kelly. 1997. Navy bean canning
quality: Correlations, heritability estimates, and random amplified polymorphic DNA
markers associated with component traits. J. Amer. Soc. Hort. Sci. 122:338-343.

Wassimi, N.N., T.G. Isleib, and G.L. Hosfield. 1986. Fixed effect genetic analysis of a diallel
cross in dry beans (Phaseolus vulgaris L.). Theor. Appl. Genet. 72:449-454.

Welch R.M., W.A. House, S. Beebe, and Z. Cheng. 2000. Genetic selection for enhanced
bioavailable levels of iron in bean (Phaseolus vulgaris L.) seeds. J. Agric. Food
Chem. 48: 3576—3580.

Welsh, J. and M. McClelland. 1990. Fingerprinting genomes using PCR with arbitrary
primers. Nucleic Acids Res. 18:7213-7218.

Weeden, NE. 1984. Distinguishing among white seeded bean cultivars by means of
allozyme genotypes. Euphytica 33:199-208.

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

Winham, D.M., A.M. Hutchins, and OS. Johnston. 2007. Pinto bean consumption reduces
biomarkers for heart disease risk. J. Amer. Coll. Nutr. 26:243-249.

33

Yaish, M.W.F and M. Perez de la Vega. 2003. Isolation of (GA)n Microsatellite Sequences
and Description of a Predicted MADS-box Sequence Isolated from Common Bean
(Phaseolus vulgaris L.). Genet. and Mol. Bio. 26: 337-342.

Yan, X., H. Liao, S.E. Beebe, M.W. Blair, and J .P. Lynch. 2005. Molecular mapping of
QTLs associated with root hairs and acid exudation as related to phosphorus uptake
in common bean. Plant Soil 265: 17—29.

Young, ND. 1996. QTL mapping and quantitative disease resistance in plants. Annu. Rev.
Phytopathol. 34:479-501.

Yu, K., 8.]. Park, and V. Poysa. 1999. Abundance and variation of microsatellite DNA
sequences in beans (Phaseolus and Vigna). Genome 42: 27-34.

Yu, K., 8.]. Park, V. Poysa, and P. Gepts. 2000a. Integration of simple sequence repeat

(SSR) markers into a molecular linkage map of common bean (Phaseolus vulgaris
L.). J. Heredity 91: 429-434.

Yu, K., 8.]. Park, and V. Poysa. 2000b. Marker-assisted selection of common beans for

resistance to common bacterial blight: efficacy and economics. Plant Breed. 119:
411-415.

34

Chapter 2: Identification of QTL for agronomic and canning quality traits in a black

bean population

Abstract

Quantitative trait loci (QTL) analysis was used to identify QTL for agronomic
performance and canning quality in a recombinant inbred line (RIL) population from the
cross of the black bean cultivar ‘Jaguar’ and the breeding line 115M. A total of 96 RILs
were evaluated for seven agronomic and four canning quality traits in replicated trials at
one location over four years (2004-2007) in Michigan. SSR, TRAP, SRAP, and
phenotypic markers were used to create a genetic map of the population consisting of 119
loci including a locus associated with resistance to a new race of bean rust isolated in
Michigan. The map consisted of 15 linkage groups spanning 460cM (38%) of the bean
genome. Composite interval mapping analysis identified a total of 20 QTL for 10 traits
averaged across environments, while an additional 18 QTL were identified in one or
more individual environments. QTL were identified on 10 linkage groups. A major QTL
for seed yield was identified on B10 while B05 contained the greatest number of
independent loci. A total of 7 QTL for yield, seed size, plant height, and canned bean
texture showed positive alleles from 115M. Several QTL co-localized with regions

identified in previous studies while others, particularly for canning quality, were unique.

35

Introduction

All breeding programs share a common goal to develop high yielding cultivars with
desirable agronomic and quality traits (Brick and Grafton, 1999). However, genetic
improvement of yield potential in bean cultivars has been lower than in other crops (N ienhuis
and Singh, 1988; Kelly et al., 1998). Improving yield requires breeding for the interrelated
effects of growth habit, seed size, maturity, and gene pool (Komegay et al., 1992), while the
end result, improved cultivars, must fit the constraints of a particular production system to be

accepted by the marketplace (Kelly, 2001).

Bean breeders have used varied selection strategies to adapt basic breeding methods
for increasing yield. Recurrent selection was successful in generating an upright type II pinto
cultivar (Kelly and Adams, 1987) and generally increased yield potential in other studies
(Ranalli et al., 1991; Ramalho et al., 2005). Selection based on an individual yield
component was not successful (N ienhuis and Singh, 1988), likely restricted by yield
component compensation (Adams, 1967). Early generation yield testing (EGT) can be
effective (Singh et al., 1990), although the resources necessary for implementing EGT for
yield may limit its application in most breeding programs (Kelly et al., 1998). Singh (1994)
proposed gamete selection for the simultaneous selection of multiple traits, although the
single seed descent (SSD) method resulted in more lines with more desirable combinations of
traits (Singh, 1997).

Other approaches to breeding for yield in bean were conceptualized and implemented
successfully. Adams (1973) proposed breeding for an ideotype. Wallace et al. (1993)
proposed breeding for physiological efficiency by simultaneously selecting for the
interrelated traits biomass, harvest index, and maturity. Singh (1992) suggested specific

combining abilities should guide breeding decisions. Beebe et al. (2008) showed that

36

breeding for abiotic stress tolerance also improved harvest index and increased yield in
favorable environments.

In addition to manipulating the diversity present in the cultivated gene pool, breeders
have targeted wild germplasm as a source of additional favorable alleles. Effective
utilization of wild germplasm requires the efficient introgression of genetic diversity into an
adapted growth habit that can be grown and evaluated across a range of environments (Kelly,
2000). Historically, breeders have struggled to uncover useful alleles masked by undesirable
traits such as climbing growth habit and photoperiod sensitivity in wild beans. Knowledge of
the domestication syndrome, a relatively small group of loci controlling a large proportion of
the differences in growth habit, seed/pod traits, and photoperiod sensitivity between wild and
cultivated bean (Koinange et al., 1996), has led to renewed hope for capturing favorable
variation from the wild. The proliferation of molecular markers linked to domestication traits
has also made the process more attainable in recent years.

Breeding techniques such as the inbred-backcross line method (IBL) (Bliss, 1993)
have been especially well suited for transferring genetic diversity from wild beans into an
adapted background. The Centro Intemacional de Agricultura Tropical (CIAT, Cali,
Colombia) has extensively utilized IBL to introgress diversity from wild beans of Colombian
origin into domesticated germplasm (Beebe et al., 2004). Molecular methods were used to
identify wild beans that were unique from the Middle American, Andean, or northern Andean
groups of origin (Tohme et al., 1996). One of these wild beans, G24423, was crossed with
the Mexican cultivar ‘Tacana’ (Lopez-Salinas et al., 1997) to establish an inbred backcross
BC; population that closely resembled the cultivated parent while lacking the undesirable
characteristics of the wild parent. G24423 possesses a type IV, indeterminate-climbing
grth habit and small seed size (9g per 100 seed) typical of wild beans (Beebe et al., 2001).

The IBL population that resulted from these crosses at CIAT was first evaluated in Mexico.

37

One BC2F4;7 line yielded 5790 kg/ha in Michigan, which surpassed all local checks
and exceeded the yield of the domesticated recurrent parent by 27% (Beaver et al., 2003;
Kelly, 2004). The same high yielding line, 115M, continued to perform well in the 2004-06
national cooperative dry bean nursery trials, producing the top mean yield in the black bean
market class across 12 locations (Hang, 2004; 2005; 2006). These results encouraged filrther
research to dissect the genetic architecture underlying the increased performance of progeny
derived from wild by domesticated crosses.

Quantitative trait loci (QTL) analysis is a useful tool for exploring the genetic control
and variation of complex traits. QTL are defined as regions of the genome statistically
associated with phenotypic variation of a quantitative trait (Doerge, 2002). Identification of
QTL requires the construction of a genetic linkage map in a population segregating for traits
of interest along with phenotypic data for the traits. Identification and mapping of QTL
provides a starting point for marker assisted selection (MAS), which can be utilized to
improve traits with low heritabilities or those that are difficult or expensive to measure using
direct phenotypic selection (Collard et al., 2005). QTL analysis can also provide a method to
locate genes of interest for future fine mapping, validation, or map-based cloning studies (Li

et al., 2006; Liu et al., 2008).

Past QTL studies in bean have examined diseases, insects, and abiotic stresses that
limit yield. To date, QTL analysis for disease resistance has been the focus of extensive
research, and MAS based on these QTL studies has been widely implemented in breeding for
resistance to bean golden mosaic virus and common bacterial blight (Miklas et al., 2006).
Several studies have examined agronomic traits contributing directly to yield potential, either
by measuring total yield or its individual components such as plant height, seeds per pod, pod

number, or seed size (Beattie et al., 2003; Tar’an et al, 2002; Blair et al., 2006). Although

38

these studies have identified some similar QTL, many of the results have been unique to
specific populations or environments. These results underscore the difficulty of defining the
genetic elements contributing to complex traits such as yield and suggest that further studies
are warranted.

In addition to high yield, acceptable canning quality in cultivars is a trait valued by
consumers, bean processors and plant breeders (Posa-Macalincag et al., 2002). Consumers
desire canned beans that are visually appealing, with a color and texture that are pleasing to
the palate following processing. Processors seek to provide bean products that satisfy these
requirements, but are also concerned with the logistics of efficiently processing beans.
Therefore they desire beans with a durable seed coat that will hydrate efficiently and
uniformly during blanching and have a high water holding capacity that increases processor
yield (Walters et al., 1997). Plant breeders are faced with the challenge of providing bean
cultivars that address quality standards from both of these perspectives.

Components of canning quality are quantitatively inherited and exhibit a continuous
range of phenotypes (Hosfield et al., 1984; Walters et al., 1997; Posa-Macalincag et al.,
2002). Consequently, developing cultivars that possess a balance of these components that
collectively contribute to acceptable or superior canning quality requires constant evaluation
at all stages of the breeding process. Typically, breeders invest 3 or more years in early
generation line development before initial quality evaluations are made and inferior lines can
be eliminated. This delay not only adds time and cost to developing new cultivars, but it also
limits the number of lines that can be reasonably evaluated. Due to the difficulty and
expense associated with selecting for canning quality, breeders would benefit fi'om
alternative selection methods that can be used confidently to select for superior canning

quality (Walters et al., 1997).

39

A number of selection strategies have been suggested or implemented based on a
limited amount of previous research in this area. Wassimi et al. (1990) suggested recurrent
selection might be the most effective means of combining desirable characteristics, while
Walters et a1. (1997) and Posa-Macalincag et al. (2002) revealed potential for MAS of QTL
in guiding selection of some components. While these studies illustrate the potential for
MAS, they also suggested some limitations due to population or gene pool specificity of
some markers. Indirect phenotypic selection methods have been considered for traits
correlated to canning quality based on studies by Bushey and Hosfield (2007), Lu et al.
(1996), Ruengsakulrach et al. (1991) and Shellie and Hosfield (1991). These methods
evaluated various physical or chemical characteristics of a small sample of seed and
correlated the results with those of traditional canning protocols. In practice, these early
generation selection methods have not been widely utilized, suggesting the need for
continued research in this economically important area of bean breeding.

The objectives of the current study were: 1) Develop a linkage map utilizing a
population of 96 F45 RIL individuals derived fiom the cross ‘Jaguar’ by 115M. 2) Collect
phenotypic field data over four seasons to conduct QTL analysis of yield and other
agronomic traits. 3) Measure color, texture, visual appearance, and washed-drained weight to
conduct QTL analysis of canning quality traits in the same RIL population.

Materials and Methods
Plant Material
‘Jaguar’ (Kelly et al., 2001) and the breeding line 115M (CIAT) were used as parents
to develop 96 F 45 recombinant inbred lines (RILs). The initial cross was made in 2001 and
advanced to the F2 generation in the Michigan State University greenhouse. The F 2 family
was planted at the Saginaw Valley Bean and Beet Research Farm in 2002, where 96 plants

were randomly chosen to establish the RIL population through single seed descent. ‘Jaguar’

40

is a black bean cultivar adapted to Michigan growing conditions. 115M was selected for its
high yield potential from an inbred backcross line (IBL) population developed at the
International Center for Tropical Agriculture (CIAT) in Cali, Colombia (Kelly, 2004; Acosta-
Gallegos et al., 2008).

115M was developed by the inbred-backcross technique by crossing ‘Tacana’
(Lopez-Salinas et al., 1997) and the wild bean G24423, followed by two backcrosses to the
recurrent parent ‘Tacana’. This IBL population was phenotypically very similar to the
recurrent parent, and 46 of the IBL were evaluated in Michigan in 2000. From this field trial,
four lines, including 115M, were selected that significantly exceeded the yields of ‘Tacana’,
as well as local checks in Michigan (Kelly, 2004). ‘Tacana’ is a black bean cultivar from
Mexico. G24423 is a wild bean fiom Colombia originally selected for its unique molecular
marker pattern (Tohme et al., 1996).

Field Trials

To investigate the agronomic potential of the 96 RILs, the population, along with
‘Jaguar’, ‘Tacana’, 115M, and the commercial check cultivar ‘T-39’, were evaluated at the
Saginaw Valley Bean and Beet Research Farm from 2004 to 2007. Plots consisted of four
rows 6.4M in length, with 0.5M row spacing. They were organized in a 10 x 10 lattice with
three replications. Standard agronomic practices were followed to ensure adequate crop
growth and development. Data were collected for days to flower, plant height, lodging, days
to maturity, and overall agronomic desirability. Yield and 100-seed weight data,
standardized to 18% moisture, were collected by direct harvesting 4.6m of the middle two

rows of each plot.

41

Canned Bean Evaluation

The population was also evaluated for canned bean color, texture, visual appearance,
and washed-drained weight. Color measurements were recorded as a luminosity (L) value on
the Hunter LAB scale using a LabScanXE (Hunter Laboratory, Reston, VA) where 1=black
and 100=white. Texture measurements were made with a Kramer Shear Press (Food
Technology Corp., Sterling, VA). For each genotype, two 100g samples taken fiom a single
can of thermally processed beans were tested using the bean processing methodology posted
on the Bean Improvement Cooperative website (BIC, 2008). Visual appearance was
subjectively rated on a 1=undesirable to 7=desirable scale by a group of panelists. Washed-
drained weight was determined as the weight of the entire canned bean sample rinsed under
cold water and allowed to drip dry for 2 minutes on a standard number 8 (2.36m) sieve.
Data was collected from beans that were grown and canned during the years 2005-2007.

DNA Isolation and Molecular Marker Analysis

The RIL population and parents were grown in the greenhouse and DNA was
extracted from young trifoliate leaf tissue bulked from three to four individual plants per
genotype using a modified CTAB method (Haley et al., 1994). DNA concentrations were
determined with a fluorometer (Hoeffer DyNA Quant 200, San Francisco, CA) according to
the manufacturer’s procedure and adjusted to 40 ng pl" for use in PCR. Molecular markers
screened for polymorphisms between the parents 115M and ‘Jaguar’ included 444 SSR, 64
SRAP, 220 TRAP, and 7 SCAR markers. Those that were polymorphic between parents
were used to genotype the population.

SSR Markers

Amplification reactions were performed with 1 ul of DNA diluted to 40 ng pl", 1.0ul

of (2mM) primer, 0.2ul (1U) of Taq polymerase, 0.6111 (50mM) MgCl2 , 2.0111 (10x) PCR

42

buffer, 0.8ul of a 5mM mix of dNTPs, and 14.4ul sterile distilled water. PCR was
conducted in a 96 well PT C-100 Programmable Thermal Controller (MJ Research, Inc.,
Waltham, MA) programmed for 1 cycle of 5 minutes at 94°C, followed by 30 cycles of 1
minute at 94°C, 1 minute at 47°C, and 1 minute at 72°C, and a final extension step at 72°C
for 5 minutes. Prior to loading on gels, 8p] of forrnamide loading buffer was added to each
sample, which was then denatured for 5 min. at 94° C. PCR products were separated on 6%
denaturing polyacrylamide gels in 0.5x TBE buffer, electrophoresed on Sequi-Gen GT
Sequencing Cells (Bio-Rad Laboratories, Hercules, CA) at a constant power of 1800W for
approximately three hours, and silver stained with a Silver Sequence kit (Promega, Madison,
WI) according to the manufacturer’s procedure for viewing.

SRAP and TRAP Markers

Amplification and electrophoresis on agarose gels was performed as described by
Terpstra et al. (2006) for most SRAP and all TRAP markers. The remainder of the SRAP
markers were electrophoresed and viewed as described above using polyacrylamide gels.

Phenotypic Markers

Segregation for resistance to race 73 of Colletotrichum lindemuthianum, the causal
agent of anthracnose, and race 3:22 of Uromyces appendiculatus were assayed in the RIL
population following the methods of Kelly et al. (1994) and Stavely (1983).

Data Analysis, Linkage Map Construction, and QTL Analysis

Analysis of variance for all traits in a given year and a combined analysis as a
randomized complete block design (RCBD) across years were performed with Proc GLM in
the Statistical Analysis System (SAS) (SAS Institute Inc., Cary, NC). The mean values for

each trait across years were used to calculate Pearson correlation coefficients among traits.

43

Linkage analysis was performed on genotypic data using JoinMap 3.0 (Van Ooijen
and Voorrips, 2001). The Kosambi mapping function was used, which assumes the existence
of interference that is negatively related to recombination frequency. A minimum logarithm
of odds (LOD) threshold of 3.0 and recombination frequency smaller than 0.300 was used to
divide the 181 markers into linkage groups, determine marker order, and calculate relative
map positions. LOD scores are = log (Ll/L0), where L1 is the likelihood for the alternative
hypothesis and L0 is the likelihood of the null hypothesis. A LOD score of 3 means the
alternative hypothesis is 1000 times more likely than the null hypothesis. Linkage groups
were identified and named according to the core reference map (F reyre et al., 1998) based on
microsatellite map locations previously assigned in Blair et al. (2003a) and Grisi et al.
(2007). Remaining linkage groups were anchored by mapping one or more markers in a
subset of the BAT93/JaloEEP558 RIL population.

QTL analysis was performed for the combined environment using the mean for each
of the 11 traits across the four seasons for each line, and separately for each individual
environment using the mean for each line in the respective year. Windows QTL cartographer
version 2.5 (Wang et al., 2007) was used to identify QTL for days to flower, plant height,
lodging, maturity, overall desirability, seed yield and 100-seed weight, in addition tocanned
bean color, texture, visual appearance, and washed-drained weight. The Composite Interval
Mapping (CIM) function set to a window size of 100M, 5 background markers, 2cM walk
speed, and a forward and backward regression model was used to identify QTL. Significant
QTL for individual traits were determined by the location of the peak LOD score at a genome
wide empirical threshold of p=0.05 after 1000 permutation tests (Churchill and Doerge,

1994). Linkage maps and QTL were displayed using Mapchart v2.2 (V oorips, 2002).

44

Results
Field Trials

Mean squares for genotype and environmental differences were significant
(p<0.0001) for all seven agronomic traits measured (Table 2.1). Significant genotype by
environment (GxE) interaction was detected for all traits except days to flowering. With the
exception of seed weight, all other agronomic traits were significantly correlated with three
or more other traits (Table 2.2). Yield was positively correlated with days to flowering,
lodging score, and plant height. Desirability score was inversely correlated with days to
flowering, maturity, lodging, and height. Days to flowering, maturity, lodging score, and
plant height were all positively correlated with each other.

Although significant differences were observed in the combined analysis across all
four years on an entry mean basis for all seven traits across the population (Table 2.3), the
means for the parents were significantly different only for lOO-seed weight, maturity, lodging
score, and desirability. All traits showed transgressive segregation and nearly normal
distributions (Figure C. 1). There were two significant differences between 115M and its
recurrent parent ‘Tacana’, 115M yielded 435 kg/ha more and flowered one day later. 115M
yielded significantly more, had increased plant height and desirability score, and decreased
lodging score when compared to the check, ‘T-39’. ‘Jaguar’ possessed significantly smaller
seed size, lodging score, and increased plant height when compared to the check cultivar ‘T-
39’.

Nine lines yielded in the top 10% of the yield trial two or more years, and two lines
appeared in this group all four years (Tables 2.4, 2.7). Similarly, five lines ranked among the
bottom 5% of the trial for two or more years. In contrast to the consistently high yielding
lines, the same line consistently yielded the least in all four years. The combined average

yield for the 96 RILs was 3058 kg/ha, with a range of 2249 to 3654 kg/ha (Table 2.3).

45

Within individual years, yields ranged from a low of 1214 kg/ha in 2004 to a high of 4261
kg/ha in 2006, a range of 3047 kg/ha (Table 2.4). These extremes were observed in the driest
and wettest years, respectively. In 2004, total precipitation from June 1 to September 30
measured 195mm, while 328mm was recorded during the same time period in 2006 (Figure
2.1).

Seed size, recorded as 100-seed—weight, varied significantly by as much as 20%
(Table 2.4). With the exception of 2005, where mean seed size for the population increased
significantly to 22.8g during the second driest season, average seed size varied between 19.0
to 20.4g. Both low and high yielding lines exhibited a range of seed sizes (Table 2.4), which
agreed with the lack of correlation between yield and seed weight (Table 2.2).

Days to flower and maturity remained relatively constant fi'om one year to the next
(Table 2.8). However, flowering was delayed by 8-10 days in 2007, and maturity was
similarly delayed. Lodging scores were the lowest in 2004, highest in 2006, and were
equivalent between the high and low yielding groups. Similarly, plant height remained
consistent from 2004-2007. Average desirability scores were similar between 2004 and
2005, but decreased in 2006 and 2007.

Canning Traits

Significant differences among lines in the population were observed for canned bean
color, texture, and washed-drained weight. Mean squares for genotype, environment, and the
genotype by environment interaction were significant at p<0.0001, both averaged over 2005-
2007 and for each individual year, (Table 2.10). Significant differences were also observed
among seasons for all population mean values of each seed quality trait (Table 2.11).

The population showed a range of variation for canned bean color, texture, visual
appearance, and washed drained weight. The distributions for color, visual appearance, and

washed-drained weight were normally distributed (Figure C. 1) between 115M and ‘Jaguar’.

46

The parents showed the greatest difference for texture (23.1 kg-force), and the population
followed a bimodal distribution. Although a majority of the lines were normally distributed,
there was a second group that more closely resembled the firmer texture of 115M. Washed-
drained-weight was normally distributed.

Beans retained the most color in 2006 with a mean color (luminosity) of 14.2 and
exhibited the greatest color loss in 2005 with a mean color of 17.6 (Table 2.11). The softest
textures were measured in 2007 with a mean of 53.5kg-force needed to compress a 100g
sample of beans to the point of catastrophic bean failure. Conversely, beans from 2005 had
the firmest texture of the three seasons with a mean of 63.9kg-force. Washed-drained weight
was negatively correlated to color and texture, and positively correlated with visual
appearance, with a low value of 243g in 2005 and a high of 254g in 2007 (Tables 2.2, 2.11).
Color was inversely related to visual appearance and washed-drained-weight.

Markers and Linkage Map

A total of 182 loci were included in the linkage analysis which resulted in 119
markers placed on the linkage map divided among 15 linkage groups for a total map distance
of 460cM (Figure 2.2; Figure C2). The number of markers per linkage group varied fiom 2
to 24. Markers clustered on BS, B6, and B 10, while their distribution was more uniform for
the remaining linkage groups. Three linkage groups consisting of a total of seven markers
were not successfully anchored to one of the 11 bean linkage groups (Freyre et al., 1998), and
B9 was the only linkage group not represented in the current map.

Polymorphism levels observed with the molecular markers used in this study ranged
from low to moderate depending on marker type. A total of 444 SSR, 220 TRAP, 64 SRAP,
and 7 SCAR markers were screened. Fifty six SSR markers (12.6%) amplified polymorphic
fragments between the parent lines. Twenty one SRAP (32.8%) primer pairs amplified 42

clearly scorable fragments, while 55 TRAP (25%) produced 81 scorable fi'agments.

47

Two phenotypic markers for disease resistance were also placed on the linkage map.
‘Jaguar’ possesses the Co-I gene that resides on B1 and conditions resistance to race 73 of
Colletotrichum lindemuthianum (Kelly et al., 2001; Vallejo and Kelly, 2008), the causal
agent of anthracnose, but was susceptible to race 3:22 of Uromyces appendiculatus, which
causes common bean rust. Conversely, 115M was susceptible to anthracnose, and possessed
an unknown gene that conditioned resistance to rust race 3:22. Forty eight lines in the
population were resistant to race 73 of anthracnose, while 27 were susceptible and 21 lines
segregated for resistance reactions (Table C.7). These results did not fit the expected 1:1
segregation ratio for a single resistance gene in a RIL population (p=0.0153). Sixty three
lines were resistant to race 3:22 of common bean rust, while 15 were susceptible and 18 lines
segregated for both reactions. These results deviated significantly (p<0.0001) from a 1:1
ratio in favor of the resistant (115M) allele (Table 3.3).

The phenotypic marker for anthracnose resistance was used to anchor a linkage group
to B1, which allowed the SSR IAC28 to be mapped to Bl for the first time (Figure 2.2). The
phenotypic marker for rust resistance was mapped to B4, in the same region as the rust
resistance genes Ur-Dorado-108 and Ur-5 (Miklas et al., 2000). All markers on this linkage
group exhibited a skewed segregation toward the 115M allele, which was consistent with the
skewed phenotypic marker distribution. Mapping rust resistance is discussed further in
Chapter 3.

QTL Analysis

Composite interval mapping identified 20 QTL associated with 10 traits in 13 marker
intervals on 10 linkage groups when data was combined for all four environments (Table
2.12; Figure 2.2). QTL per linkage group ranged from 1 to 4, with clusters of 2 or more QTL
occurring on 4 linkage groups. Individual QTL explained 7 to 22% of the phenotypic

variation, and total phenotypic variation explained for a trait varied from 14% for plant

43

height and canned bean visual appearance to 46% for canned bean color (Table 2.12). The
number of QTL per trait ranged fiom 1 to 4. Individual environments varied for both the
total number of QTL and the number of traits for which those loci were detected (Table
2.13). In 2006, 16 QTL were detected for 10 traits, while in 2004, 11 QTL for 5 traits were
identified. The total number of QTL for 2005 and 2007 were intermediate between these
results, with 13 and 12 QTL identified for these environments, respectively. Eighteen
additional QTL were detected in one or more single environment that were not present in the
combined environment.

Yield

A single major QTL for yield that originated in 115M was identified on B10 with
R2=0.19 and an additive effect of 127 kg/ha (Table 2.12). No other significant QTL for yield
were detected in the combined environment. Linkage groups B3, B5, B10, and B1 1
possessed significant QTL in one or more environments fi'om 2004-2006, while none were
detected in the 2007 (Table 2.13). The R2 values ranged fiom 0.08 to 0.28, and additive
effects varied from 41-192kg/ha. The only QTL detected from ‘Jaguar’ was located on B3,
increased yield by 168kg/ha in the 2004 environment, and was located 20cM from a QTL
detected in 2006 Item 115M (Figure 2.2).

Seed Size

The alleles from 115M at loci on B6 and B11 each increased seed size by 0.3g per
100 seed and had R2 values of 0.08 and 0.11, respectively, in the combined environment
(Table 2.12). QTL identified in one or more environments were located on BS, B6, B8, and
B11, controlled relatively small proportions of the variation in seed size (R2 = 0.09-0.15), and
had additive effects of 0.4-0.5g (Table 2.13). In contrast to seed yield where both parents
contributed alleles with an additive effect in some environments, only alleles from 115M

contributed to increased seed size.

49

Days to Flowering

QTL from Jaguar on B11 and from 115M on LG2 delayed flowering by 0.3d each in
the combined environment (Table 2.12). No additional loci influenced this trait in any
individual environment but the two QTL consistently showed equivalent effects on days to
flowering in 2004 and 2006 (Table 2.13).

Maturity

In the combined environment, two alleles from 115M delayed maturity by 0.5d each
(Table 2.12). These QTL resided on B5 and LG2, and both had R2 values of 0.19. In one or
more individual environments, additional QTL from ‘Jaguar’ on Bl, B3, and B7 also delayed
maturity by less than one day (Table 2.13). In 2006 and 2007, three loci accounted for 50%
and 36% of the total variation in maturity.

Lodging

Two loci increased lodging score in both the combined and individual environments
(Tables 2.12, 2.13). These QTL with R2 values of 0.13 and 0.15 were associated with the
115M allele on linkage groups B4b and B6 and increased lodging score minimally by 0.2
points each. Rust resistance in 115M also mapped in the same region as the lodging QTL on
the lower end of linkage group B4b, and was the marker most tightly linked to the lodging
QTL in 2007.

Height

A single QTL that slightly increased plant height was detected on B5 in the combined
environment and was associated with the 115M allele (Table 2.12). Additional QTL on
linkage groups BB, B6, and 311 were detected in one or more environments and associated
with the ‘Jaguar’ allele (Table 2.13). An additional QTL from 115M was detected on B6 in

2004 at a distance of 18cM fi'om the QTL detected in ‘Jaguar’ in 2006 (Figure 2.2).

50

Desirability

Two QTL were detected for desirability score on linkage group B5 and B6 (Table
2.12). Increased desirability was associated with the ‘Jaguar’ allele and each locus had an
additive effect of 0.2. No additional QTL were detected in individual environments, but the
locus on B6 accounted for 20% of the variation for this trait in 2004 (Table 2.13).

Canned Bean Color

QTL influencing canned bean color retention in the combined environment resided
on linkage groups B3, B5, B8, and B11 and collectively accounted for 46% of the variation
for color. Each locus on B3, B5, and B8 decreased black color by 0.4 (increased L-value)
and originated in 115M while the locus on B11 decreased color by 0.3 and originated in
‘Jaguar’ (Table 2.12). An additional QTL was detected on B1 in 2007, for a total of four
QTL from 115M that decreased color in one or more environments (Table 2.13).

Texture

Variation for canned bean texture was associated with two regions of linkage group
B1 and one region of B6 in the combined environment (Table 2.12). Together, the three
QTL accounted for 42% of the variation for texture and at each locus the 115M allele
increased texture by 2.0-3.6kg force. An additional QTL was identified on B11 in 2005 that
increased texture by 2.9kg force (Table 2.13).

Visual Appearance

For the combined environment, a single QTL associated with the ‘Jaguar’ allele on
linkage group B8 slightly increased visual appearance by 0.1 (Table 2.12). However, in
2006, this region was associated with the 115M allele and increased visual appearance by
0.4, and an additional QTL for this trait was also detected on B5 and associated with the

‘Jaguar’ allele in 2005.

51

Washed-Drained Weight

Washed-drained weight represented the only trait for which no stable QTL across
environments were identified (Table 2.12). In 2006, three QTL were detected on linkage
groups B3 and BIO (Table 2.13). On B3 and the upper end of B10, the ‘Jaguar’ allele
increased washed-drained weight by 1.65g while on the lower end of B10 the 115M allele
resulted in a similar increase (Figure 2.2).

Co-localized QTL

QTL were detected at four locations that co-localized in the genome for the combined
environment. On linkage group B6 QTL that co-localized for lodging and agronomic
desirability were detected. QTL for canned bean color and visual appearance resided on B8.
A region of the LG2 possessed QTL for both days to flowering and maturity. A complex
cluster of 4 QTL was identified on B5 for maturity, plant height, overall agronomic
desirability, and canned bean color. Within this cluster, QTL for maturity and plant height
were detected adjacent to each other. Agronomic desirability and canned bean color QTL
co-located with each other, as well as both maturity and plant height. This cluster represents
the only location where a seed quality QTL co-located with QTL for agronomic traits. Only
one QTL was detected on each of the remaining linkage groups (Figure 2.2), with the
exception of three QTL distributed across B11.

QTL 1: Environment Interactions

Significant environmental interactions (QxE) were observed for one or more QTL
detected for each trait except days to flowering (Table 2.13). The proportion of QTL for a
particular trait that showed an environmental interaction varied. QxE was frequently
detected for yield, seed size, desirability, and canned bean color, while fewer QTL for
maturity, lodging, height, texture, visual appearance, and washed-drained weight showed an

environmental interaction.

52

Discussion

Field Trials

The evaluation of the recombinant inbred line population over four years provided the
opportunity to observe these genotypes in the field under a range of environmental
conditions. The four growing seasons represented a range of conditions from dry and hot
environments that limited yields to years with more moderate temperatures and adequate
precipitation that maximized yield potential (Figure 2.1). In 2004 and 2005, adequate soil
moisture was present at planting, but was followed by below normal precipitation throughout
the growing season. In 2006, growing conditions were average early in the season; July
rainfall was more than twice the 30-year average, which led to increased vegetative growth,
while August had a 20-day period without rain. June and July were dryer than normal in
2007, followed by above average rainfall in August.

From a breeding standpoint, these conditions provided a challenge to identify stable
lines that consistently performed despite different environmental conditions. Breeding line
BO4431 with a 4 year mean of 3654kg/ha, significantly exceeded the mean yield of the
population, ‘Jaguar’, and the check cultivar ‘T-39’, but failed to yield significantly more than
the high yielding parent 115M. Several other lines also consistently produced yields above
the test mean and in the top 10% of the population (Table 2.4, 2.7). Similarly, there were
lines that ranged in yield depending on the year (Table 2.7), and those that consistently
produced poor yields (Table 2.4). 304442 with a mean of 2249 kg/ha was consistently the
lowest yielding line in the population every year.

The normal distributions and transgressive segregation provided an opportunity to
identify lines that exceeded the average yield of 115M by up to 298kg/ha (Table 2.9),
although this difference was statistically insignificant (LSD=321kg/ha). Although modest in

comparison to 115M, which has produced record high yields, the top yielding lines in the

53

population represent a much larger yield advantage compared to many of the other elite
breeding lines trialed during the same seasons. Conversely, eleven lines yielded significantly
less than ‘Jaguar’, up to 799kg/ha less. These results illustrate the difficulty of generating
progeny with increased yield potential, and the relative ease of recovering lines with inferior
performance compared to the agronomically desirable parents.

Although significant genotype by environment interactions were present in all years,
the accumulated data on these lines provided adequate information to select the best lines for
use in future crosses with other elite lines. Those lines with stable yield potential across
diverse environments represent useful germplasm that will be utilized to improve the yield
potential of future breeding lines.

Compared to the other black bean yield trials conducted at the same site during the
same years, the mean yield of the population was higher in all seasons but 2005. This
difference may reflect the extended dry conditions during most of the growing season (Figure
2.1), which appeared to limit the overall yield potential of this population more than the
genetically diverse lines in the standard breeding trials. In other studies, ‘Tacana’, the
recurrent parent of 115M, has shown less tolerance for drought stress compared with other
black bean cultivars despite its superior yield potential under improved growing conditions
(Beebe et al., 2008). 115M appears to exhibit the same characteristic that limited the yields
of the population during the dry conditions of July and August 2005. Similarly, yields for the
population were also substantially less in 2004, the overall driest of the four seasons (Figure
2.1).

Correlations among traits generally agreed with those of other studies. Seed size was
the only trait not correlated to any other traits, which agreed with results in a navy bean
population (Tar’an et al., 2002). The same population showed positive correlations among

days to flower, maturity, and plant height, but no correlation between lodging and either days

54

to flower or maturity was detected. The inverse relationship of desirability with days to
flower, maturity, lodging, and plant height was expected since later maturity and increased
lodging decrease desirability rating.

Canning Traits

The variation observed among years for processed bean characteristics of the
population agrees with previous studies that have shown seed quality traits vary widely based
on environmental conditions among growing seasons (Posa-Macalincag et al., 2002; Walters
et al., 1997). Differences in weather patterns such as air temperature or available moisture at
critical times during the development of the bean crop have been implicated as contributing
to the large range in bean quality traits from one season to the next. However, genetic
differences still account for a significant portion of this variation in both the current
population and previous studies (Shellie and Hosfield, 1991; Posa-Macalincag et al., 2002).

The inverse relationship observed between texture and washed drained weight agrees
with studies in both navy and black beans (Wassimi et al., 1990), and in three navy bean
populations (Walters et al., 1997). This relationship is logical, since higher washed-drained
weight results from more water entrainment in the bean, which makes the bean easier to
break apart, or exhibit a softer texture.

‘Jaguar’ and 115M differed substantially in canning quality traits, particularly for
texture. These contrasts led to a distribution of progeny, and made this population especially
suitable for detecting QTL for canning quality (Figure 2.2). Therefore more variation was
explained by the analysis of canning quality traits than for that of the agronomic traits.

Despite the variation for these traits, most lines appeared brown and washed out with
considerable loss of bean integrity after canning. Unfortunately, some of the highest yielding
lines were among the least desirable based on canning traits. Due to the importance of

canned bean visual appearance to commercial processors and consumers, these lines failed to

55

meet minimum quality standards when compared visually with check cultivars. The
generally undesirable canned bean appearance of members of the population that closely
resembled that of 115M was unexpected at the beginning of the study since ‘Jaguar’
possesses acceptable canning quality traits.

Although every effort was made to treat samples the same in all years, any minor
changes to the canning procedure likely influenced the quality attributes measured in this
study. The only intentional modification in procedures was made between 2005 and 2006
during the soak prior to canning, so differences in color retention between those years may
reflect the change in processing protocol. This change was made to reduce the amount of
color lost from the beans, so the darker color values measured in the later years suggest this
modification was effective. No other adjustments to the canning protocol were made fi'om
year to year.

Linkage Map

Common bean has 11 linkage groups, which correspond to the genome’s 11
chromosomes that are estimated to cover a total genetic distance of 1200cM (F reyre et al.,
1998). The current map consisting of 15 linkage groups spans 460cM representing 38% of
the estimated genome size. Ten of the 11 linkage groups of the bean consensus map were
anchored based on the placement of SSR markers (Blair et al., 2003), or mapping of SRAP
and TRAP markers in the BAT93 x JaloEEP558 (BJ) core map population. Linkage group
B9 was absent fi'om the current map, although three small linkage groups remain unanchored
and one may represent a small portion of B9. Linkage groups B1 and B4 were each
represented by two un-joined linkage groups.

Low polymorphism levels are often observed in narrow crosses within a gene pool,
race, or market class. A recent study by Blair et al. (2006b) examined polymorphism levels

of 129 SSR markers in 44 common bean genotypes from both Middle American and Andean
56

gene pools. A higher inter gene pool polymorphism level of 59.6% was observed, while the
intra gene pool level was 37.9%. Comparisons between two races or closely related cultivars
were less polymorphic, suggesting comparisons between two black beans would result in an
additional reduction of informative markers.

The low number of polymorphic SSR markers available in this study that resulted in
only 38% coverage of the genome suggests the need for continued marker development to
make this marker system widely applicable to variety development. Although SSR markers
have been successfully used to detect numerous alleles at a locus in genetic diversity studies
such as Blair et al. (2006b) or Gomez et al. (2004), the reality is that many loci will be fixed
for the same allele within elite breeding germplasm. To overcome this limitation, breeders
need access to a larger group of markers so that despite a lower polymorphism rate in closely
related populations, upwards of 120 markers would still be informative. More of these
molecular tools are currently available in other crops such as soybean (USDA, 2008), and
help to provide improved coverage in linkage mapping studies.

Although developing new markers will require an investment of resources,
substantial progress has been realized in other crops. Over 1000 SSRs have been placed on
the consensus map of soybean (USDA, 2008). In wheat, over 500 SSRs have been developed
and more than 300 placed on the consensus map (Song et al., 2005). Yu et al. (1999)
concluded microsatellite sequences are abundant in the common bean genome. However, to
date less than 200 of the 500 SSR markers developed have been mapped (BIC, 2008).

Recently, Buso et al. (2006); Benchimol et al. (2007); Campos et al. (2007); Grisi et
al. (2007); and Hanai et al. (2007) have developed a large group of new SSR markers, more
than doubling the number available at the beginning of the present study. A number of these
markers were integrated into the current linkage map (Figure 2.2). However, if 1000 or more

SSR markers were available, one could construct a map of a similar population with a single

57

co-dominant marker system, as is routinely done in other legumes such as soybean. This
could lead to more uniform genome coverage by selecting evenly distributed markers rather
than relying on non-species specific, dominant marker systems such as SRAP or TRAP that
tend to cluster (Miklas et al., 2006).

Few published maps of bean have utilized TRAP markers with the exception of
Miklas et al. (2006b), and no literature is available for SRAP markers in bean. Unpublished
data indicate the polymorphism rate of SRAP markers was three times greater than either
RAPD or AF LP markers, and TRAP markers were twice as polymorphic as those marker
systems (V. Vallejo, personal comm.) In the present study, SRAP markers possessed three
times the polymorphism rate of the SSR markers, which was equivalent to the rate observed
with SSRs for intra gene pool comparisons (Blair et al., 2006). Similarly, TRAP markers
were about twice as polymorphic as the SSRs, and the 1.5 markers generated per primer pair
agreed closely with the results of Miklas et al. (2006b) who observed an average of 1.3
markers per primer combination within a race of the Mesoamerican gene pool in a ‘Dorado’
x XAN176 RIL population.

Segregation distortion

Linkage group B4b (Figure 2.2) contained six markers including the phenotypic
marker for rust resistance. These markers all showed severe segregation distortion that
favored the 115M allele. Although markers on other linkage groups in the current map
differed from the expected 1:1 ratio, the observed differences were much less than those
markers on B4b. These data were particularly interesting since they occurred near a known
cluster of resistance genes (Miklas et al., 2006), suggesting this region of the 115M genome
was favored throughout the population development process. Similarly, Blair et al. (2003b)
found significant distortion on the same region of B4 in a cultivated by wild population. In

that study, the cultivated allele was always favored, and the region was associated with the

58

architecture of the recurrent cultivated parent. Cichy et al. (2009) reported that a genomic
region related to the determinate growth habit on B1 was favored over indeterminate plant
types in a population derived from a detenninate/indeterminate cross. In the present study,
the distorted linkage group was associated with lodging, suggesting an association to plant
architecture but the reason for the distortion remains unclear, unless unconscious selections
were made for upright plant types during population development.

QTL

Twenty QTL were identified for ten traits in thirteen marker intervals across the
genome when data was combined across the four environments (Table 2.12). Several
additional QTL were identified in one or more single environments for some traits, while for
other traits no QTL were detected in some years (Table 2.13). These results support the
value of using a RIL population to conduct the experiment across a wide range of
environments in order to determine which genomic regions consistently control the largest
portions of the variation for each trait.

Although more agronomic traits were considered, fewer QTL and a lower percentage
of total variability were explained per trait than detected for the seed quality traits (Table
2.13). The firm texture and poor color retention of 115M along with the softer texture and
higher color retention of ‘Jaguar’ resulted in a wide distribution of lines in the population that
facilitated more efficient detection of loci associated with quality characteristics (Table B. 1 ,
Table 2.13). In contrast, the two parents differed significantly only for a few agronomic traits
including seed size, maturity, lodging, and desirability score, which resulted in a narrow
distribution for these traits within the RIL population.

Yield

A single region of linkage group B10 was associated with 19% of the variation for

yield in the combined environment. The allele from 115M had an additive effect of

59

127kg/ha. The detection of this QTL in only three of four individual environments was
surprising, based on the high LOD score in the combined environment. In addition, the lack
of a significant yield QTL in 2007 was unexpected. However, these varied results agree with
those of Tar’an et al. (2002), who found only 25% (5 of 20) of the QTL detected across
environments were detected in single environments. The study also found additive effects
and approximate location in the genome varied from one environment to another, which
agrees with the results of the current study.

The detection of QTL on B3, B5, and B11 in one or more years but not consistently
in all years suggests that several genomic regions with relatively small effects are influencing
yield in this population and their effect varies depending on the environmental conditions
present that season. These results were not unexpected due to the variation in precipitation
and other weather patterns among the four growing seasons in the study. In addition, the
QTL detected on B3 was associated with the ‘Jaguar’ allele in 2004 but in 2006 a region
20cM from the same QTL was associated with the 115M allele. Previous studies identified
QTL for yield on B3 and B5 in a navy bean breeding line (Beattie et al., 2003), and QTL on
BS and B10 were identified in the navy cultivar ‘OAC 95-4’ (Tar’an et al., 2002). Blair et al.
(2006a) also identified two QTL on B3, one associated with a wild bean (G24404) and one -
with the Andean cultivar ‘Cerinza’. The wild bean was collected in the same region of
Colombia as G24423, the wild parent of 115M. This information suggests one or more
regions of linkage group B3 are associated with enhanced yield in a range of both Middle
American and Andean beans from diverse genetic backgrounds. These data also support the
complex genetic nature of yield potential described by previous studies and suggest that
limited improvements in yield are possible by selecting for any one QTL alone. A breeder
would need to transfer positive alleles at several loci into a single cultivar to significantly

improve yield and ensure stable increased yield potential across varied environments.

60

Introgression of all these QTL into other breeding lines may be challenging, but these minor
QTL still represent a source of positive variation for yield.

In contrast, QTL with more significant effects, such as the B10 QTL that explained
19% of the variation in yield across four environments, represent loci that could have a larger
individual influence on yield. The BIG QTL represents a region that could be targeted for
MAS in black beans or for introgression into other classes of common bean. The cost of
performing MAS for this region may influence whether genotypic or phenotypic selection is
used to introgress this QTL into other lines, but RILs possessing this QTL certainly should be
crossed with other elite breeding lines.

Seed Size

All QTL identified for seed size were associated with the 115M allele. These regions
located on linkage groups B6 and B11 were identified in the combined environment and each
increased seed size by 0.3 g. These results were interesting as they suggest no negative effect
of the small seeded wild bean, G24423, on seed size in 115M or the population. The average
seed size of 115M was slightly larger than that of the recurrent parent ‘Tacana’ as well as
‘Jaguar’, which was not associated with any QTL for this trait. In every individual
environment, one or more regions were associated with an increase in seed size, with
additional QTL on B5 and B8 identified in one environment each. No QTL accounted for
more than 15% of the variation for seed size, suggesting control of this trait resides at many
genomic locations each with small effects. Blair et al. (2006a) identified a QTL for seed size
on B6 associated with G24404, as well as on B8 and B11 associated with ‘Cerinza’. Perez-
Vega et al. (2008) located QTL for seed size on B6 and B8 in an Andean by Middle
American population. Tar’an et al. (2002) also detected a QTL on El] in a navy bean
population, whereas Park et al. (2000) identified similar seed size QTL associated with

Andean cultivar ‘PC-50’ on BS, B6, and B8.

61

Days to Flowering

QTL detected for days to flowering on B11 and LG2 were consistent among the
combined and individual environments, although they were not detected in all single
environments. The ‘Jaguar’ allele on B11 was associated with a slight increase in days to
flower, while the 115M allele on LG2 had a similar effect. The occurrence of QTL for both
days to flowering and maturity at the same location on the unanchored LG2 supports the
results of Tar’an et a1 (2002) for navy bean and Blair et al. (2006a) for the wild bean G24404.
Both of these studies showed co-localized QTL for days to flowering and maturity in
populations derived fi'om similar genetic backgrounds as the ‘Jaguar’ll 15M RIL population.
This information suggests LG2 could correspond to the same region of B9 where co-
localized QTL for these traits were previously identified, but attempts to anchor this 15cM
linkage group to the core map were unsuccessful, and B9 was not mapped in the current
study.

Maturity

Linkage groups BS and LG2 carried QTL associated with the 115M allele that
delayed maturity in the combined environment. The QTL on LG2 was interesting as it co-
localized with a QTL for days to flowering. As mentioned above, previous studies identified
a region of B9 that controlled both of these traits, and although speculative, these results
suggest LG2 could represent a portion of B9. Due to the absence of sufficient polymorphic
markers on B9, we were unable to verify an association with LG2. Additional QTL were
detected on B l , B3 and B7 and in each case the ‘Jaguar’ allele delayed maturity. These
additional regions were each specific to a single environment. Blair et al. (2006a) associated
similar regions of BS and B7 with maturity in the Andean cultivar ‘Cerinza’. In contrast,
Beattie et al. (2003) and Tar’an et al. (2002) did not identify any similar regions associated

with maturity in Middle American beans.

62

Lodging

Two QTL on B4 and B6 that increased lodging score were associated with the 115M
allele. The effect of each of these regions on lodging was relatively small, although together
they accounted for 28% of the variation in lodging score. Unlike most traits studied where
various regions influenced a trait depending on the year, lodging was consistently associated
with these regions in both individual years and the combined environments. Beattie et al.
(2003) associated a similar region of B4 with lodging in a navy bean population while the
QTL on B6 has not been identified in previous studies. The location of the B4b QTL was
also interesting in that rust resistance in 115M mapped to the same region, and a higher than
expected frequency of resistant lines was observed. Since all markers in this linkage group
also showed a distorted segregation in favor of the 115M allele, the B4b QTL provides an
explanation why the population more closely resembles 115M than ‘Jaguar’ in regard to

lodging.

Height

Increased plant height was associated with the 115M allele in a region of linkage
group B5 in the combined environment. Additional QTL associated withthe ‘Jaguar’ allele
in regions of B3, B6, and B11 were detected in one or more environments but not in the
combined analysis, supporting the hypothesis that plant height is largely influenced by
environmental conditions. In addition, a QTL from 115M was detected on linkage group B6
in 2004 at a distance of 18cM from the B6 QTL contributed by ‘Jaguar’ in 2006, suggesting
that multiple alleles influencing plant height reside in close proximity to each other on
linkage group B6. Similar QTL associated with increased plant height were reported by
Checa and Blair (2008) on linkage groups B3 and B11 in an indeterminate Middle American

climbing bean. Blair et al. (2006a) identified similar QTL on B6 that were derived from both

63

an Andean cultivar and a wild bean accession, while Tar’an et al. (2002) located a QTL for
height in a similar region of B6 in a navy bean population. Both Tar’an et al. (2002) and
Beattie et al. (2003) identified QTL for plant height in similar regions of B3 in different
Middle American cultivars, suggesting that plant height is controlled by this region in a

number of different genetic backgrounds.

Desirability

Increased desirability was associated with two ‘Jaguar’ alleles located on B5 and B6.
Each locus had an equivalent effect on desirability score and no additional QTL were
detected in any individual environment. No QTL for desirability were associated with 115M.
This result was not unexpected due to the less desirable architecture of 115M compared with
‘Jaguar’, which has a more compact, upright grth habit. As increased desirability score
reflects the sum of other phenotypic traits, such as early maturity, lodging resistance, and
increased plant height, so QTL on BS and B6 are likely associated with regions that control
these traits.

Canning Traits

Eight QTL for quality traits were located at seven unique locations across the
genome, suggesting that the contrast between 1 15M and ‘Jaguar’ for seed quality
characteristics allowed for the efficient detection of regions influencing these traits. Except
for the co-localization of QTL for color and visual canning score on B8, all QTL occurred in
separate regions of the genome, supporting the complex, quantitative nature of these traits as
established by previous studies (Hosfield et al., 2004). However, direct comparisons with
previous studies were not possible due to unanchored linkage groups reported in previous
QTL analyses. Six of the eight QTL detected in the current study were contributed by 115M.

These results were interesting since this line was never selected for canning traits, but three

64

of the six QTL associated with 115M had a positive effect on canning quality. The lack of
QTL detected from ‘Jaguar’ was surprising, but reasonable based on the resemblance of
many of the lines in the population to 1 15M when canned.

Canned Bean Color

Although four regions were associated with color retention in the combined
environment, within a single year one to three loci were detected, suggesting that
environmental conditions in a given year largely influenced this trait. Previous studies have
also implicated environmental factors as contributing to large differences in the results of
canned bean quality evaluations, and suggested that results of quality evaluations are largely
location and population specific (Walters et al., 1997; Posa-Macalincag et al., 2002). The
QTL identified on B3, B5, and BS each decreased color retention by 0.4 points and originated
in 115M. The QTL on B11 decreased color retention similarly, but was associated with the
‘Jaguar’ allele. The QTL from 115M were not surprising based on the poor canning
characteristics of that line, while the QTL from ‘Jaguar’ was not expected based on the
acceptable canned bean color of that cultivar. Individually the four QTL accounted for 7-
15% of the variation in bean color retention, but collectively they accounted for 46% of the
variation in color. Posa-Macalincag et al. (2002) identified a similar region of B3 that was
associated with improved canning quality in the Andean kidney bean cultivar ‘Montcalm’,
but they did not detect any other QTL identified in the current study.

Texture

Together, the three QTL detected in two regions of linkage group B1 and on B6
accounted for 42% of the variation for texture. At each locus the 115M allele had a positive
effect on texture ranging from 2.0-3.6kg force. The increase in texture influenced by the
115M allele was surprising based on the poor visual canning characteristics of that parent.

However, similar increases in texture have been recorded for pinto beans with poor visual

65

appearance following canning. The total R2 for all QTL detected for canned bean texture in
this population was greater than in any of three populations examined by Walters et a1.
(1997) or two populations studied by Posa-Macalincag et al. (2002).

Visual Appearance

In the combined environment, a single QTL associated with the ‘Jaguar’ allele on
linkage group B8 slightly increased visual appearance. However, in 2006, this region was
associated with the 115M allele and increased visual appearance by 0.4, which suggests the
115M allele influenced visual appearance more than ‘Jaguar’ in that environment. An
additional QTL influencing this trait was detected on linkage group B5 in 2005 and
associated with the ‘Jaguar’ allele. Neither of these loci explained a large percentage of the
variation for visual appearance, suggesting that QTL analysis for this trait was not as
effective as it was for canned bean color or texture. These contrasting results could reflect
the difference between the objective measures of color and texture and the subjective
evaluation of visual appearance. Walters et al. (1997) also explained a lower percentage of
the variation for visual appearance explained compared with other canning quality traits.

Washed-Drained Weight

QTL for washed-drained weight were identified only in 2006 where three QTL were
detected on linkage groups B3 and B10. The ‘Jaguar’ allele increased washed-drained
weight by 1.6g each on B3 and the upper end of B 10, while on the lower end of B10 the
115M allele resulted in a similar increase. Together, these loci accounted for 33% of the
variation for the trait. The reason these loci were detected only in a single year remains
uncertain, as this was the only trait where QTL were inconsistent across multiple years.

These results underscored the large effect environmental conditions had on this trait.

66

Co-localized QTL

QTL that co-localized at four locations in the genome were detected for the combined
environment. Co-localized QTL often indicate the location of tightly linked loci, or a single
locus with pleiotropic effects (Hittahnani et al., 2002). QTL co-localized on B6 for lodging
and agronomic desirability, on B8 for canned bean color and visual appearance, and on LG2
for days to flowering and maturity. Since lodging score was a component of the desirability
score, canned bean color contributed to visual appearance, and days to flowering influences
maturity, these QTL detected in the same regions likely indicate a single locus controlling
multiple traits. In contrast, the cluster of four QTL for maturity, plant height, desirability,
and canned bean color on linkage group B5 represented the only instance of a seed quality
QTL co-located with loci controlling agronomic traits. While maturity, plant height, and
desirability were correlated with each other and are likely controlled by the same QTL, seed
color was not correlated with any of those traits, suggesting it is influenced by another locus

that is adjacent but distinct.

Combined and Individual Environments

In addition to the 20 QTL identified in the combined analysis, 18 other QTL were
identified in one or more single environments. No additional QTL for days to flowering,
lodging, or desirability were detected in any of the four environments considered. Only QTL
for seed size and color were identified in every environment, while there were environments
where no QTL for other traits were detected. Chaib et al. (2006) reported similar variation in
a study comparing stability of quality QTL over years, generations, and genetic backgrounds
using multiple QTL introgressed into various population structures and genetic backgrounds
of tomato. Their results showed large differences in the number as well as magnitude and

direction of individual QTL detected depending on environment, even when phenotype was

67

determined in a closely controlled greenhouse. Together, these results suggest there is value
in conducting QTL studies over a number of diverse environments to detect as many of the
different genomic regions influencing a trait as possible so as to identify stable QTL over
years.

Conclusions

The QTL analysis for agronomic and seed quality traits across four contrasting
environments identified desirable alleles from 115M that enhanced yield, seed size, plant
height and canned bean texture. A single QTL accounted for 19% of the variation for yield
across four environments, while in a single environment up to three QTL were identified that
controlled 34% of the variation for yield. Likewise, 19%, 16%, and 42% of the variation for
seed size, plant height, and canned bean texture, respectively, were accounted for in the
combined environment. However, alleles with undesirable effects for days to flowering,
maturity, lodging, overall desirability, as well as canned bean color and visual appeal were
also detected.

The analysis was particularly useful for dissecting the genetics of canned bean color
and texture. These two traits were controlled by loci on at least six chromosomes, suggesting
that accumulating favorable alleles at all loci in a single line will remain difficult. Although
the 115M allele was generally associated with an undesirable effect on canning quality, a few
positive effects were noted. In the combined environment, the 115M allele for a QTL on B11
improved canned bean color, while in 2006, the 115M allele at a QTL on B8 improved visual
appearance. The positive effects of the three 115M QTL associated with texture were also
unexpected based on the undesirable visual appearance of 115M following canning. These
loci demonstrate the potential of inferior parents to contribute positive alleles that result in

desirable transgressive segregants.

68

One or more QTL were identified for all agronomic traits examined including yield,
seed size, days to flower, maturity, lodging, plant height, and overall agronomic desirability
score. Additional QTL were found for seed processing quality traits including canned bean
color, texture, and visual appearance. Washed-drained weight was the only trait considered
where no stable QTL across combined environments were identified. The total phenotypic
variation explained for visual appearance exceeded that of four out of five previously studied
populations. A total of 42% of the variation in texture was also explained by 3 QTL. A
complex cluster of 4 QTL was identified in the middle of linkage group B5, while pairs of
QTL for different traits co-localized on groups B6, B8 and LG2.

This group of black bean lines reflected the yield potential of 115M, and showed
transgressive segregation for both high and low yield. Several lines that consistently
exceeded the yield of 115M should be considered for use as parent material to enhance the
yield and seed size of elite germplasm. Alleles that improve canned bean texture could be
separated fi'om those that confer color loss, based on the independence of these loci. These
results support the use of TRAP markers for mapping and tagging QTL in common bean.
However, conversion of closely linked TRAP or SRAP markers to more robust SCAR
markers prior to implementing MAS would likely improve the efficiency of the selection
process by facilitating the multiplexing of markers.

Continued research would provide additional details regarding the true breeding value
of this germplasm. Although useful alleles were identified across diverse environments for
key traits, the current study provides no information about the combining ability of these
alleles with different genetic backgrounds. Crosses with a subset of elite lines from this
population will provide additional insight into how these lines will combine with other

breeding materials to improve yield of common bean.

69

Table 2.1. ANOVA table showing mean squares (p50.0001) for yield, 100 seed weight,
days to flowering, plant height, lodging score, maturity, and agronomic desirability score
for 96 recombinant inbred lines in the Jaguar/115M population combined across four
environments (2004-2007) in Michigan.

 

Trait YLD SW FLWR HT LDG MTR DS
Genotype (G) 139723.] 10.6 4.2 3.9 0.9 7.9 1.7
Environment (E) 26612095 755.9 3872.1 1307 44.8 2349 76

GxE 25018.5 1.6 0.89ns 2.3 0.3 2.1 0.5

 

ns= not significant at p50.05
YLD=Yield, SW=100-Seed Weight, FLWR=Days to Flowering, HT=Plant Height,
LDG=Lodging Score, MTR=Maturity, DS=Agronomic Desirability

70

EBB 855-333.1939? season? a=a>u<> 3&8. "be. .360 5.5 Baaoumao 5592239 2888?qu
.beeazugz .aoom mewefinon: awe: amine: $55: a amalgam .awaa commeeuam Earle.»

 

 

ScodWa:
~o.o.v.**
modw...
$1.36 1.5de * ad- ofic E.G- h—dn wed- 26. mod mod- E;
m _ .o azmmd- mod 2 .o- mod m _ .o- wed mod vod <>
cod- E.G- mod- *omd h — .o- bod- v_ .O mod EXP
cod- *avmd 56 m _ .o 26 Ed 3.,de “AU
iavwd- ..:.:..o~..o.. «aaowd- *twhvdu vfior ofiou mn—
eabmd *eammd sliced mad- aszd P:
fiend .12..de 22¢- ***wm.o ”VA—1—
***om.o c—d ofio MPH:
wed .Ibmd guf—
ofio- 3m
<> EXP MAD mm— H: 991— Mat—Z gs:— Bm 94>

 

$8-38 macaw .3322 5 Saw sea 2m: 3 teams.
a Eofi come—gov 8:: 355 2859882 ea Soc mash bzmzc 38 can 015.8%“ Lou 356508 coca—2.80 cow—mom .NN 28¢.

71

Table 2.3. Phenotypic means and ranges for yield, 100 seed weight, days to flowering,
maturity, lodging score, plant height, and desirability score; canned bean color, texture,
visual appeal, and washed drained weight for 96 recombinant inbred lines in the Jaguar/115M
population combined across four environments (2004-2007) in Michigan.

 

 

 

 

 

 

flaring Recombinant inbred lines C_hegk_s
Trait Jaguar 1 15M Mean Range LSD.05 Tacana T-39
Agronomic Traits
Yield (kg/ha) 3050 3350 3058 2249-3654 321 2915 2999
100 Seed Weight (g) 19.5 21.6 20.4 18.1-22.7 1.1 21.1 20.8
Days to Flower 47 47.5 47.4 45.8-50.1 1.1 46.3 47.2
Days to Maturity 95 96.5 96.3 94.5-99.4 1.4 95.8 95.3
Lodging Score 1.2 1.9 1.6 1.0-2.6 0.5 1.7 3.6
Plant Height (cm) 48.4 49.3 49.1 47.3-50.7 1.5 48.7 42.5
Desirability Score 5.3 3.9 4.4 3.1-5.4 0.7 4.25 2.9
Canning Traits
Color (L) 14.9 17.2 16.0 13.7-20.0 1.3 16.3 14.6
Texture (kg-force) 48.4 71.5 59.3 44.6-79.0 9.6 84.1 48.5
Visual Appeal 3.9 2.3 2.4 1.6-3.2 0.6 2.4 3.3

Washed-Drained Weight (g) 253 246.1 248.6 2349-260.5 6.8 245.8 254.7

 

72

Table 2.4. Four year average (2004-2007) seed yield and 100-seed weight of the top ten and
bottom five recombinant inbred lines in the Jaguar/115M population ranked by seed yield.

 

 

Rank based on Yield
lOO-Seed
Line Yield Weight 2004 2005 2006 2007
kg/ha g
B04431 3654 22.3 5 6 2 2
B04404 3601 20.5 1 4 8 6
B04391 3553 22.6 39 16 1 1
B04444 3539 18.6 7 3 6 12
B04445 3527 21.1 2 11 7 8
B0441 1 3482 21.2 22 14 3 5
B043 84 3474 19.8 6 9 46 3
B04429 3463 21.1 8 18 10 7
B04412 3393 20.2 31 2 17 27
B04443 3387 18.4 15 25 11 11
B04434 2642 22.2 91 80 74 98
B04392 2625 21.2 97 98 81 57
B04381 2532 20.8 77 99 97 99
B04425 2502 21.8 99 97 72 94
B04442 2249 22.7 100 100 100 100
Jaguar 3050 19.5 60 38 71 51
115M 3350 21.6 11 23 25 28
Tacana 2915 21.1 78 64 75 55
T-39 2999 20.8 10 77 80 87
Test
Mean(100) 3058 20.4
LSD(.05) 321 1.1

 

73

Table 2.5. Flowering day, maturity, lodging score, plant height, and agronomic desirability
score of the top ten and bottom five yielding recombinant inbred lines in the Jaguar/115M
population ranked by average seed yield from 2004-2007.

 

 

Flowering Maturig Lodging Height Desirabilig
Line Mean Rang Mean Range Mean‘l' Rflge Mean Range MeanI Range_
--------------- days--------------- -------cm----

B04431 48 45-55 97 96-107 1.6 1.5-3.0 51 49-55 4.4 3.0-4.0
B04404 48 43-55 96 92-101 2.2 1.0-2.0 50 47-53 3 .8 4.0-6.0
B04391 48 46-56 99 95-101 1.9 1.0-3.0 50 43-53 3.1 3.0-5.0
B04444 48 44-54 96 94-102 1.6 1.0-2.0 50 47-54 4.4 3.0-6.0
304445 48 42-54 98 96-102 1.8 1.0-2.9 50 48-53 4.3 3 .0—5.0
B044] 1 48 43-55 97 94-102 2.0 1.0-2.5 50 48-52 4.4 3.5-5.0
804384 47 43-53 97 92-101 1.4 1.0-2.6 50 48—53 5.0 3.0-4.5
B04429 47 43-55 96 93-106 1.6 1.0-2.4 50 47-54 4.1 3.5-5.5
BO4412 47 43-53 95 92-101 1.1 1.0-2.0 49 46-52 5.3 3.5-5.5
804443 47 45-55 96 95-101 1.5 1.0-3.0 50 48-54 4.5 3.0-5.0

B04392 46 42-54 96 91-100 1.3 1.0-1 .5 49 45-52 4.3 4.0-6.0
B04381 47 42-54 95 95-102 1.2 1.0-2.0 48 47-54 4.3 4.5-5.5
B04425 49 44-54 99 96-100 1.8 1.0-2.0 50 49-54 3.3 3.0-5.0
B04442 47 45-54 96 96-102 1.3 1.0-2.6 48 49-54 4.6 3.5-5.0

Jaguar 47 43-54 95 92-100 1.2 1.0-1.5 48 45-52 5 .3 5.0-6.0
115M 48 44-54 97 94-102 1.9 1.5-2.5 49 46-54 3.9 3.5-4.5
Tacana 46 42-52 96 94-99 1 .7 1.0-2.0 49 46-52 4.3 4.0-4.5
T-39 47 43-55 95 93-100 3.6 2-4.1 43 35-47 2.9 1.5-5.0
Test Mean(100)

47 96 1.6 49 4.4
LSD(.05L 1.1 1.4 0.5 1.5 0.7

 

'1' Lodging rated 1=erect to 5=prostrate
I Desirability rated 1=undesirable to 7=desirable

74

ozfifiovnh 8 28:82?“ 688 88.83% 835 A.

 

 

 

 

 

 

we 3 ed 2 $953
38 am 8% 32 28:8»: 8...
88.38 5.8 5-3 mm 23.3. 3:. 8.2-8.2 8.2 8%
5.8-58 38 3-8.. am 32.: 3.» «some: m2 «828
$8.08 38 3-: mm 32.8 n: 32 .2 N: 22 _
$8.88. 38 313 mm c.8884 3:. 02-8.2 a: 8:8.
38-38 ~88 8.28 E 32.: Se 4.2-82 o2 N348
4.98.38 38 48-2 3 0.2.8.8 m8 98-5.2 3: 8448
8.83.48 38 3-: ed 8.8-3.. «.2 3:62 q: 82.8
5886.8 3.8 3-2 3 33.8 28 33.: .2 89.8
382.48 38 3-3 2 88.9%. New ”.24.: “.2 838
4.2.3.8 2.8 4.28 E 33.8 OS 5.2-8.2 8.2 $38
8.3.3.38 28 2-2 _.N 8.8-08 v.8 3:2 a: £43m
38-8.38 28 «8-2 am c.8864 «:8 8.2.8.: 3. 83.8
282.98 3.8 8.2 _.m 32.8 2: 928.2 _.: 838
4,884.88 28 4.3.. 3 8.8.2:. «.2. 8.2-4.: ”.2 _ :48
8.8288 38 3-3 3 239. gm 8.2-3. 4.: $38
38-0.88 28 4.38 an 93.3.. «.8 8.2.32 m2 343m
38-38 38 4.38 cm 8.8-5.8 8.8 :32 4.2 33.8
28.38 5.8 3-8.. 2H 33.2 28 4.2.8.2 8.2 335m
“2.48.38 28 m 38 am 13.38 E 8.2-8.: 3: 23.8

w my. I--I-o=8>-.._ .......
owcmm +802 ow8m 8o: amp—8m .822 0:5

3 flag. 3.8

 

.nooméoom 80¢ 22» team 0888 3 88:8.— aouflaom 22 C883. 2: E 8:: .855 “8:588?—
wEEoE gm Eaton E8 :8 no“ 2: mo £303 388863.83 c8 608.83% 8:2.» 62sz .830 .83 3580 .oN 038,—.

Table 2.7. Number of years recombinant inbred lines in the Jaguar/115M population ranked
in the top 10% or bottom 5% based on seed yield.

 

 

Top 10%
4 Years 3 Years 2 Years
B04431 BO43 84 B04366
B04404 B04429 B04391
B04444 B04411
........... B. 03545_-__-_-_.
Bottom 5%
B04442 804381 B04392
B04408

 

Table 2.8. Yearly trait means for 2004-2007 and corresponding least significant differences
for the ‘Jaguar’/ 1 15M recombinant inbred line pgmlation grown in Michigan.

 

 

Year YLD sowr FLWR MTR LDG'l' HT 08;:
kg/ha ---g---- ---d--- --d--- --cm-

2004 2226 20.4 46.5 93.7 1 46.9 5.1

2005 3508 22.8 43.6 96 1.8 47.7 4.7

2006 3474 19 45 94.3 2.2 48.8 3.7

2007 3047 19.6 54.4 101.4 1.5 52.8 4

LSD(.OS) 67 0.1 0.2 0.3 0.1 0.3 0.2

 

YLD=Yield, SW=100-Seed Weight, FLWR=Days to Flowering, HT=Plant Height,
LDG=Lodging Score, MTR=Maturity, DS=Agronomic Desirability

‘1' Lodging rated 1=erect to 5=prostrate

I Desirability rated 1=undesirable to 7=desirable

76

Table 2.9 Four year mean (2004-2007) yields for 96 recombinant inbred lines in the
Jaguar/115M population, parents, and checks ranked by descending seed yield.

 

 

Yield Line Yield Line Yield
kg/ha kg/ha kg/ha

80443 1 3648 804396 3 196 Tacana 2915
804404 3601 804452 3196 804419 2892
804391 3553 804407 3190 804436 2892
804444 3539 804422 3173 804377 2889
804445 3527 804385 3170 804417 2889
80441 1 3482 804361 3159 804401 2884
8043 84 3474 804449 3 151 804432 2853
804429 3463 804450 3 128 804426 2842
804412 3393 804369 3126 804379 2833
804443 3387 804409 3 1 17 804437 2825

115M 3350 804372 3117 804388 2822
804394 3350 804447 31 14 804415 2822
804423 3345 804440 3109 804399 2819
804414 3322 804454 3 106 804359 2799
804387 33 17 804441 3106 804371 2797
804360 33 1 1 804382 3103 804393 2797
804451 3308 804420 3100 804389 2797
804370 3297 804397 3100 804416 2780
804366 3286 804406 3089 804390 2763
804410 3274 804421 3089 804438 2749
804376 3244 804418 3083 804395 2743
804446 3241 804403 3081 804448 2726 a
804386 3230 804402 3075 804424 2718 3
804453 3224 804364 3066 804375 2710 a
804383 3215 804367 3061 804427 2676 a
804400 3215 804363 3050 804380 2667 3
804398 3215 Jaguar 3050 804408 2645 a
804374 3210 804362 3019 804434 2642 a
804433 3207 804373 3007 804392 2625 a
804439 3207 804368 3007 804381 2532 a
804435 3207 T-39 2999 804425 2502 a
804413 3204 804428 2991 804442 2249 3
804405 3 199 804378 2951
804365 3196 804430 2915

 

a= Significantly lower yield than ‘Jaguar’ (LSD(.05)=321kg/ha).

77

Table 2.10. Analysis of variance for canning quality traits of a ‘Jaguar’ by 115M RIL
population including canned bean color, texture, washed-drained weight, and visual
appearance.

 

 

Trait Color Texture Washed—drained weight Visual Apgarance
Genotype (G) 6.7 279.4 50.9 0.4
Environment (E) 566.5 5682.4 2823.7 32.5
G x E 1.3 72.5
Mean 16 59.3 248.6 2.4
LSD 0.8 2.5 6.7 0.6

 

Table 2.11. Mean values by year for canned bean color, texture, washed-drained weight, and
visual appearance for 2005-2007.

 

 

YEAR CLR TXT VA WDWT
2005 17.6 63.9 2.2 243.3
2006 14.2 60.8 3.1 248.9
2007 16.2 53.5 2.0 253.9
LSD(.05) 0.4 2.5 0.1 1.5

 

CLR=Canned Bean Color, TXT= Texture, VA=Visual Appearance, WDWT=Washed-
drained Weight

78

.885... 89a 0383: :8 22. 89a 20:8 0808:. 83? 03:8: 85:8 8 5:98: 0:: :5... 20:8 03:8 8 wag—£533 mo «ootm _r
.958 mama—8:. _8>..8:_ 3.89:8 38: D04 82— 8 APO 05 .3 3:8.on 88...?
2.528;: 05 we :08..an 93.: n a 8 8388.50: 82 mo Bop—mob: 83.880 023 2:88 2... @388 .o 82: 8.88 004 :8 £25 no we: .QO... H

 

 

 

 

 

 

 

 

 

 

 

 

58.85.. a2......
58 soeaoeo.e=8.=ma.86.§§a... 8 8...... 88.8... 5.8 5.0 a... no 8.58 a. .888. 88.8... .3: 5.... as... a. 8.888.... 85.... ..

:5- z 5 Z Z 5545:: .8 8528...... .22....»

.88 85 .5 58.3.. 2:
:8 .55 58 55m. 55.22 2.
5.8 N. .5 Z 55. 888?. .m 6.5.3

85. S5 55 8.5.8... . .m
2.5 . .5 N... 554.55.... .8
5.8.5 :5 .5 58.5.5... 2.
$5 25 2 3 52.55.. mm 3.5

38.; wfiEfiD

25- m . 5 m... 598?. 2.

25- N. 5 Z .5 52555... 2. 8.32.8:
85 :5 2 am 2: 2. ...fl
25 :5 5... 2.9... a...

E 5 m . 5 .8 .... 4.5.5.. 2. 8.5.5..
545 5. 5 N... 58.4.82 8:
2.5 5. 5 55 .5. 5m . 555.... 8 been:
.2. . .5 5... 53.4.8: 8..

85- m . 5 m5 5 59.55:... :m 83.25.... e an:
85 . .5 55 585:5. . .m
:5 .55 .8 .8 8.8.5.8: 2. 8.5 .85

R. 5. 5 55 45 828... 5.: 22>
22.82% 9.3... sou—.8:— 95.3
.5583. 5 2.0 a: :3 30.. 5.8282 85...: :2... 6.5868...

 

558-458 8...... 88.8.2 a. 8.8.8.6 5.... 86.6 22 :88?

8 :8... 3:230: 8:: “85:. 5:858:58. 9.. Set 80:89.35 3:588 65 E .8582: 888 .983 :08 :8 2:55.88 .8 4:0 0385:.— .N. .N 038...

79

 

m. cm... m... .mm mgm 09..th .n. ..m. 2.3

 

 

 

 

 

a. . 3... w . .0 fim outfimms. NO.
0.. m5..- 5.... Wm 9m 09.5.5... . .m 33
NN 3.... 0.... N... 05.539). mm...
5.. Rd- N . .o 5N .mm 09..an .n. . .m SEN
m.. .m... . .6 ad chivkmz NO.
Nd mm..- m . .0 m...” Em 85.3.5 .n. . .m 3:580“.
$8530.... 0. 9.5..
0... 3... m . .0 9m 2.03%.... 0..
M... cm... mo... Wm 9N cm 33.0.... mm. Sam
0..” am... m ..0 5m QM 0033...”. . .m. 2.:
NM 2.... N... N». 2.0.5.5 . .m
o.m 9.... 0.... .mm ..m 303mm... cm. 33
5.. ow... ...0 EN 03.6. .mF mm
.mm .m... N... 5N 9m 389.9. cm. 3...”.
m.m 5m... . .0 9m 30.5.3. . .m.
9N .m... no... ad ad mnmfimmz om. 3.....800
55% 3m
m. 0.. 0.... v... www.mmfi. o...
n. . R. we... 5m EN 395?. mm. 2.3
0... mm. mm... 58 82.3.5... o.m.
5m .5. m... ed 0... 09.3.5... mm mag
5.. .v N... m.m Qmmz ..m
NM 3.. ...o ~.m .w.~.m>n. o...
m. 3.- ...0 5m ~.m 809.”... mm. 2.3
0N .N. 0.... QN .mm 3.99.9. 2m. 3.....800
Eu;
Hum—PO 2.0 2053...... 9.00m .5388 9.0%
.80.. 3.2.6.... .9. 00.. .00.. .2282 «55...... 5...... 5.52.85...

 

$03-33 wag... 8583.2 :. 3.53.5.6 .85 80.5 22:18.3... 5 80... 300.05.. 3.... 3...... 8555803..
cm 80... 380880.33 .8836... .85 38383 0... a. 35.3... 3.5... 8.5.... 38 .85 0.808085 .0. APO 03.58.. .m..~ 0.95...

80

 

 

 

 

 

 

 

5N hmd :d w.m mé dwméfidmm mm Eda
Wm mmd- 2d Wm 5N SOS cm 33
2 mod. n _ d 9m dmvdmim _ a
a; de 2d 5m 83%;; cm
md 3d- m _ d ”N dd 03.3 5:2 mm 33
dd cmd 3d EN Wm 2.."— mm REE—.50
2%:
d._ 2d :d _.m and 3m
_.N 3d 2d Em wd amt/m om Eda
v; Rd E d 9N QM v Em>m om eddm
w._ mmd odd 5N dwmdfifi E nvm
m; 3d 2d 0:6. m.m Ede/m cm 33
M: o _ d m _ d ed 805 3m
9m 3d 2d ad 3“ Emm>m om 62.3.5.0
ugh—5w—
o._ 3d _ _ d dd dhvdknm—z ~04
5N dvd 3 d WV cm 3383 mm
wd 3d- _ .d 1N EN Mame/n— _m 33
m._ :d Ed d6 www.mRm—E N01—
w._ mud NNd mé om _ .deE mm
N; wmd- :d wd ”mm SOS—mg mm eddu
s; end. 2 d On dd. dmcdma um 33
5.. Ed d _ d «é duvdkm—z N01—
md ovd a _ d QN _.m dm _ .wmdz mm 62.3980
hat—5a:
max—BO 5:0 22.82; 28m hog—.5:— m=ew|
3o: F§é=6< 3m :3 ES tssoz $535 :9; 252.23

 

.3258 SN 2an

81

 

 

 

 

 

 

«N 85. 3o 3. 3322 _ E
3 and _ 3 an 832E ”m
3 one 25 3 3 RN. EE 5 33
2 33 m3 2 «a 2:. ENE mm 8..”
mm 85 Ed 3 33%: mm
mm :5 o; nm 803 mm
3 m; 85 3 3 c2. 35 8 m2.”
3 go. 8.0 mm 332$ _ .m
3 one :5 3 2:3: 3
2 a? m; 3 32%: E
S. as 25 2 mm on. 2% mm 3:355
.530
zaflfiafiu
9m .3- _ 3 Z 3 355 cm 33
3 3o- ad 3 9m Emmi cm 2.3
”a 2.? m3 3 8mm: em
3 2.? N; Z 3 figs: 8 3.....an
$58.8:
aka—Emu 5:0 238...; 989w gov—ha:— macaw
33 3352.4 a E :3 .33 t8§z «33:5 :29 252.23

 

35:0an m _.N 2me

82

530885 2885.35 3 cabocow a do 3:035 23 .8: 208 GO: w

..B:w£. E2.“ ozfiwo: was Em: 89¢ 20:“ 8865 mos—g 02:8: .8505 3 E83 one :5: 20.? o_w:_m a war—5:33 mo “outm— F
ASE—0v win—9w:— EZBE 0:82:50 wEm: DOA xdua 3 APO 2: 3 35298 02893 £90821 05 mo .5anon m

ddd n a “a mcowfiascoa 82 mo 20525 13$an 023 082% 05 3088 3 82: 388 :04 =m £35 :0 we: ”DOA H

«3.33 325.5 Mmm gum—\n—Dfiomfiswo.:mE.mmo.>233\\x—t:

as 25:0 88.38 Mmm d0 98 ”.0 333. E 958 on :8 £819: 33: was mém how cog—€85 36F:

 

 

 

 

 

 

 

NE 3.. 3d .d 35.95: d5
m._ 8.7 :d Nd muddy—mm Sm
Nd 3.7 Ed 9N dd 3095: mm 2.3
2.sz 3.35-1233
ed hdd Ed ad _.v 30023. mm eddu
wd End- Ed EN EN 305 mm mddu
dd Ed- 3 d 5N 5N ddvdma mm 3:35.00
852.8%; .55;
_.v dmd Ed 9N wd dmmNMwm om wddm
vd mad :d Nam dd dmvden— :m mddN
Ed wed mad :8 £05 Em
m; Edd wdd dd ddE.d\vm§ om
dd 3d Ed nd ed mdmmm>m Em 62.3.20
QL—uuNQH.
WEE—.0 3:0 2239:; 0.80m haw—.5:— 9.9%
33 3.363. as. :3 3o.— t8§z $3.51. :9; 252.23

 

.3288 m _.N 033.

83

i T

‘ 160 , ‘

‘ 140 '
‘ 120

I 2004 l

'00 ' 2005

“ 80 I 2006 I

60 . E 2007 i

40 ‘ 30 Year Mean ‘

‘ i

1 20 I

» i

1 0 ' ~ ‘

     

 

July August September ~

Figure 2.1. Monthly precipitation (mm) measured from June to September
2004-2007 at the Saginaw Valley Bean and Beet Research Farm, Saginaw, MI.

84

81 81b 83

 

 

 

 

 

 

 

 

 

 

o.o F13R2.250 o.o mcza 0.0 “T "150-550 E E

2.3 ammo 6.3 film “an a 0 g

7.2 W107 g a

" 12.: —~— F9R1.1so E I;
10.7 _T._ 11253.32: g
“'9 P11311233 g a g 19.9 AN1'73 21.2 ——— Pvuooo
. E
.. 29.3 —e- m

28.8 F3R1.235

 

 

 

Figure 2.2. Linkage map of ‘Jaguar’/1 15M RIL population and QTL locations for seed yield
(YLD), seed size (SDWT), days to flowering (FLWR), plant height (HT), days to maturity
(MTR), lodging (LDG), agronomic desirability (DS), and canned bean visual appearance
(VA), color (CLR), texture (TXT), and washed-drained weight (WDWT). QTL are fiirther
identified by the last two digits of year (04-07) and QTL with no year specified were detected
in the 4-year combined environment.

85

B4b

0.0 IACBG

F7R1 .150

Rust

F1 “(10.580 °'
"$6.320

7.1
10.1

1 3.4
16.3

F8R5.650

 

22.1

Figure 2.2 (Cont’d.)

5.

0.0
2.5
l' 3.7
g 7.3
9.0
13.0
19.2
20.9
21.9
22.0
22.9
25.9
27.0
20.2
23.0
29.3
30.1
31.4
32.9
34.1
42.3

85

M18
114151.375
F22R1.400
111133.175
Budzo
F10R8.150
F8R10.200
FJ16
F12R7.250
norm»
F20R4.2so
mcae
$1119.35
PvBR61 _<
F1 $5.230 i r-
F10R5.275 8
F10R5.265
F7R2.800

F1 3125.220
513115.175
3mm

86

90810

0.0
a;
1.5

ID

/PVBR5
~Pvanzo
\PVBRM

 

10.8

FBR1 .540

 

14.2

F8R2.350

 

F1 3R5.420
PVBR163

 

22.6

274\+_

31 .0
35.6
37.0

m
40.2

41.3 ‘1
43.8 ‘1

I

I TIIII

/ “187

”456.300
"385.325

/ F21 R1350

/ "755.590

/ "453.1200

,, M047
’ F3R9.875

901MOS

 

47.7 /
49.0

“.0 NH
50.5 “C

 

II

WLMOS

111571100
\ 710119.350
\ 11355375 I

LOJMOS

L; 137112

 

L‘ “750.140

Figure 2.2. (Cont’d.).

EH

[.0901

NIH

001.14

87

H H
up

1X1

87

 

0.0

2.9 N4
5.0 \—
7.1 ”‘

F22R4.550
/ M185

/ 3111210

\ PVBR67

 

11.0

F1 R9350

 

14.7

PVBR209

 

 

24.3

 

F1 R2335

i

0.0

12.8

22.0

29.5

45.0
49.8

 

 

 

 

 

 

 

 

5141151000

522112.325

5353.200

FJ28

5159.400 3 o
i-‘l :5

7155.340 2 8

Figure 2.2. (Cont’d.).

810
VA

[.0810
90VA

B10

 

0.0 PVBR181

10.4 F17R10.320
1 8.5 F2R2.57 5
25.8 “‘ MR185
37.3 F1 5R7.285
40.4 M1 3!:
42.0 F5R8.410
43.2 F1 7R3.580
43.8 F1 8R8.1 1 00
45.0 F18R3.400
45.8 F8R1 .1 25
45.7 M1 3.
47.2 F5R8.550
47.8 F13R3.1 35
48.3 F17R1 0.380
49.2 F1 7R4.225
49.5 F1 R1 .880
50.0 F5R5.890
51 .3 F1 9R5.790
52.2 F1 7R4.275
52.7 F1 7R4.285
53.0 F1 8R2.775
53.9 F1 9R8.1 200
55.8 F1 2R10.280

L

I

 

 

88

E
ii '5- l6
5

B11 L62

 

 

 

 

 

 

 

 

0.0 5753.020 "’ g a, 0.0—5—11353545
3.0 5750.000 IE IS lg .n g 1.
a 3 " .3 3 " 3
7,7 514501200 3 5 8 9. 5
2
. 517123.420 3
12 6 P! " ,1' i‘ .n 15.2 —u— 113541.470
17.2 5254.200 5 g S a 3‘ g
-< 2 3 °'
,_
26.6 141353.130 .2. o O
5 S
3
35.2 55510475

 

 

Figure 2.2. (Cont’d.).

89

Literature Cited

Acosta-Gallegos, J .A., J .D. Kelly, and P. Gepts. 2007. Pre-breeding and genetic diversity in
common bean (Phaseolus vulgaris). Proc. International Plant Breeding Symposium,
Ciudad de Mexico. Crop Sci. 47(83): 844—859.

Adams, M.W., 1967. Basis of yield component compensation in crop plants with special
reference for field beans, Phaseolus vulgaris. Crop Sci 7: 505-510.

Adams, M.W., 1973. Plant architecture and physiological efficiency. In: Potential of Field
Beans and other Food Legumes in Latin America, 226-278. CIAT, Cali, Colombia.

Beattie, A.D, J. Larsen, T.E. Michaels, and KP. Pauls. 2003. Mapping quantitative trait loci
for a common bean (Phaseolus vulgaris L.) ideotype. Genome 46:411-422.

Beebe, S., J. Tohme, J. Nienhuis, F. Pedraza, J. Rengifo, E. Tovar, and A. Islam. Studies in
Phaseolus germplasm diversity: a review of work at CIAT. 2004. Annu. Rep. Bean
Improv. Coop. 47: 33-34.

Beebe, S.E., I.M. Rao, C. Cajiao, and M. Grajales. 2008. Selection for drought resistance in
common bean also improves yield in phosphorous limited and favorable
environments. Crop Sci. 48: 582-592.

Benchimol, L.L., T. Campos, S.A.M. Carbonell, C.A. Colombo, A.F. Chioratto, E.F.
F ormighieri, L.R.L. Gouvea, and AP. Souza. 2007. Structure of genetic diversity
among common bean (Phaseon vulgaris L.) varieties of Mesoamerican and Andean

origins using new developed microsatellite markers. Genet. Resour. Crop Evol. 54:
1747-1762.

BIC. 2008. [Online] Bean Processing. Available at: http://www.css.msu.edu/bic/ (verified
July 23, 2008).

Blair, M.W., F. Pedraza, HF. Buendia, E. Gaitan-Solis, S.E. Beebe, P. Gepts, and J. Tohme.
2003a. Development of a genome-wide anchored microsatellite map for common
bean (Phaseolus vulgaris L.). Theor. Appl. Genet. 107:1362-1374.

Blair, M.W., G. Iriarte, and S. Beebe. 2003b. QTL analysis of an Andean advanced
backcross population for yield traits derived from wild P. vulgaris. Annu. Rep. Bean
Improv. Coop. 46:7-8.

Blair, M.W., G. Iriarte, and S. Beebe. 2006a. QTL analysis of yield traits in an advanced
backcross population derived from a cultivated Andean x wild common bean
(Phaseolus vulgaris L.) cross. Theor. Appl. Genet. 112:1149-1163.

Blair, M.W., M.C. Giraldo, H.F. Buendia, E. Tovar, M.C. Duque, and SE. Beebe. 2006b.

Microsatellite marker diversity in common bean (Phaseolus vulgaris L.). Theor.
Appl. Genet. 113: 100-109.

90

Bliss, F.A., 1993. Breeding common bean for improved nitrogen fixation. Plant Soil 152: 71-
79.

Brick, M.A., and K.F. Grafion. 1999. Improvement of medium-seeded race Durango
cultivars. In: S. Singh (ed.) Common bean improvement in the twenty-first century,
223—253. Kluwer, Dordrecht, the Netherlands.

Bushey, SM. and G.L. Hosfield. 2007. Absorbance of the soak water to predict canning
quality. Annu. Rep. Bean Improv. Coop. 50: 39—40.

Buso, G.S.C., Z.P.S. Amaral, R.P.V. Brondani, and M.E. Ferreira. 2006. Microsatellite
markers for the common bean Phaseolus vulgaris. Mol. Ecol. Notes 6: 252-254.

Campos, T., L.L. Benchimol, S.A.M. Carbonell, A.F. Chioratto, E.F. Formighieri, and A.
Pereira de Souza. 2007. Microsatellites for genetic studies and breeding programs in
common bean. Pesq. agropec. bras. 42(4): 589-592.

Chaib, J ., L. Lecomte, M. Buret, and M. Causse. 2006. Stability over genetic backgrounds,
generations and years of quantitative trait loci (QTLs) for organoleptic quality in
tomato. Theor. Appl. Genet. 112: 934-944.

Checa, OE. and M.W. Blair. 2008. Mapping QTL for climbing ability and component traits

in common bean (Phaseolus vulgaris L.). Molecular Breeding: DOI:
10.1007/sl 1032-008-9167-5.

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

Cichy, K.A., M. W. Blair, C. H. Galeno-Mendoza, S. S. Snapp, and J. D. Kelly. 2009.
QTL Analysis of Root Architecture Traits and Low Phosphorus Tolerance in an
Andean Bean Population. Crop Sci. 48:(in press).

Collard, B.C.Y., M.Z.Z. Jahufer, J.B. Brouwer, and B.C.K. Pang. 2005. An introduction to
markers, quantitative trait loci (QTL) mapping and marker-assisted selection for crop
improvement: The basic concepts. Euphytica 142: 169-196.

Doerge, R.W. 2002. Mapping and analysis of quantitative trait loci in experimental
populations. Nature Rev. Genet. 3: 43-52.

Freyre, R., P.W. Skroch, V. Geffroy, 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 an integrated linkage map of

common bean. 4. Development of a core linkage map and alignment of RFLP maps.
Theor. Appl. Genet. 97: 847-856.

91

Gomez, O.J., M.W. Blair, B.E. Frankow-Lindberg, and U. Gullberg. 2004. Molecular and
phenotypic diversity of common bean landraces from Nicaragua. Crop Sci. 44: 1412-
1418.

Grisi, M.C.M., M.W. Blair, P. Gepts, C. Brondani, P.A.A. Periera, and R.P.V. Brondani.
2007. Genetic mapping of a new set of microsatellite markers in a reference common
bean (Phaseolus vulgaris) population BAT93 x Jalo EEP558. Genet. Mol. Res. 6:
691-706.

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

Hanai, L.R., T. Campos, L.E.A. Camargo, L.L. Benchimol, A.P. Souza, M. Melotto, S.A.M.
Carbonell, A.F. Chioratto, L. Consoli, E.F. Formighieri, M.V.B.M. Siqueira, S.M.
Tsai, and ML. C. Vieira. 2007. Development, characterization, and comparative

analysis of polymorphism at common bean SSR loci isolated from genic and genomic
sources. Genome 50: 266-277.

Hittalmani, S., H.E. Shashidhar, P.G. Bagali, N. Huang, J .S. Sidhu, V.P. Singh, and GS.
Khush. 2002. Molecular mapping of quantitative trait loci for plant growth, yield
and yield related traits across three diverse locations in a doubled haploid rice
population. Euphytica 125: 207-214.

Hosfield, G.L., M.A. Uebersax, and T.G. Isleib. 1984. Seasonal and genotypic effects on

yield and physical-chemical seed characteristics related to food quality in dry, edible
beans. J. Amer. Soc. Hort. Sci. 109:182-189.

Hosfield, G.L., G.V. Vamer, M.A. Uebersax, and J .D. Kelly. 2004. Registration of
‘Merlot’ small red bean. Crop Sci. 44: 351-352.

Kelly, J .D. 2000. Remaking bean plant architecture for efficient production. Adv. Agron.
71 :109-143.

Kelly, J .D. 2004. Advances in Common Bean Improvement: Some Case Histories with
Broader Applications. Acta Hort. (ISHS) 637:99-122.

Kelly, J .D. and Adams, M.W. 1987. Phenotypic recurrent selection in ideotype breeding of
pinto beans. Euphytica 36:69-80.

Kelly J .D., L. Afanador and LS. Cameron. 1994. New races of Colletotrichum
lindemuthianum in Michigan and implications in dry bean resistance breeding.
Plant Dis. 78:892-894.

Kelly, J .D., J. Kolkman, and K. Schneider. 1998. Breeding for yield in dry bean (Phaseolus
vulgaris L.). Euphytica 102:343-356.

92

Kelly, J .D., G.L. Hosfield, G.V. Vamer, M.A. Uebersax, and J. Taylor. 2001. Registration of
‘Jaguar’ black bean. Crop Sci. 41:1647

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.

Komegay, J ., J .W. White & O. Ortiz de la Cruz, 1992. Grth habit and gene pool effects on
inheritance of yield in common bean. Euphytica 62: 171-180.

Li, C.B., A.L. Zhou, and T. Sang. 2006. Rice domestication by reducing shattering. Science
311:1936-1939.

Liu, S., K. Yu, and SJ. Park. 2008. Development of STS markers and QTL validation for
common bacterial blight resistance in common bean. Plant Breed. 127:62-68.

Lopez-Salinas, E., J .A. Acosta-Gallegos, E.N. Becerra-Leor, G. Frayre-Vazquez, S.H.
Orozco, and SE. Beebe. 1997. Registration of Negro Tacana common bean. Crop
Sci. 37:1022.

Lu, W., K.C. Chang, K.F. Grafion, and RB. Schwarz. 1996. Correlations between physical

properties and canning quality attributes of navy bean (Phaseon vulgaris L.).
Cereal Chem. 73: 788-790.

Miklas, P.N., J .D. Kelly, S.E. Beebe, and M.W. Blair. 2006a. Common bean breeding for
resistance against biotic and abiotic stresses: From classical to MAS breeding.
Euphytica 147:105-131.

Miklas, P.N., J. Hu, N.J. Grunwald, and KM. Larsen. 2006b. Potential application of TRAP
(targeted region amplified polymorphism) markers for mapping and tagging disease
resistance traits in common bean. Crop Sci. 46: 910-916.

Nienhuis, J. and SP. Singh. 1988. Genetics of seed yield and its components in common
bean (Phaseolus vulgaris L.) of Middle-american origin. 11: Genetic variance,
heritability and expected response from selection. Plant Breed. 101: 155-163.

Park, 8.0., D.P. Coyne, G. Jung, P.W. Skroch, E. Amaud-Santan, J .R. Steadman, H.M.
Ariyaranthe, and J. Nienhuis. 2000. Mapping of QTL for seed size and shape traits
in common bean. J. Amer. Soc. Hort. Sci. 125:466-475.

Perez-Vega, E., A. Campa, R. Giraldez, and J .J. Ferreira. 2008. Mapping of QTL involved
in the genetic control of seed traits in common bean. Annu. Rep. Bean Improv.
Coop. 51:116-117.

Posa-Macalincag, M.C.T., G.L. Hosfield, K.F. Grafton, M.A. Uebersax, and J .D. Kelly.

2002. Quantitative trait loci (QTL) analysis of canning quality traits in kidney bean
(Phaseolus vulgaris L.). J. Amer. Soc. Hort. Sci. 127:608—615.

93

Ramalho, M.A.P., A.F.B. Abreu, and J .B. Santos. 2005. Genetic progress after four cycles
of recurrent selection for yield and grain traits in common bean. Euphytica 144: 23-
29.

Ranalli, P., G. Ruaro and P. Re. 1991. Response to selection for seed yield in bean
(Phaseolus vulgaris). Euphytica 57: 117—123.

Ruengsakulrach, S., N. Srisuma, M.A. Uebersax, G.L. Hosfield, and LG. Occena. Early
generation screening of navy bean breeding lines by canning quality assessment and
pasting characteristics of bean flour. 1994. J. Food Qual. 17:321-333.

SAS Institute, Inc. 2000. SAS version 8. SAS Institute Inc., Cary, NC.

Shellie, KC. and G.L. Hosfield. 1991. Genotype x environmental effects on food quality of

common bean: resource-efficient testing procedures. J. Amer. Soc. Hort. Sci. 116:
732-736.

Singh, SP. 1994. Gamete selection for simultaneous improvement of multiple traits in
common bean. Crop Sci. 34: 352-355.

Singh, SP. 1997. Single seed descent versus gamete selection for breeding beans for
resistance to insects. Agron. Abstracts p. 71.

Singh, S.P., R. Lepiz, J .A. Gutierrez, C. Urrea, A. Molina & H. Teran, 1990. Yield testing of
early generation populations of common bean. Crop Sci. 30: 874-878.

Song, Q.J., J.R. Shi, S. Singh, E.W. Fickus, J .M. Costa, J. Lewis, B.S. Gill, R. Ward, and
RB. Cregan. 2005. Development and mapping of microsatellite (SSR) markers in
wheat. Theor. Appl. Genet. 110: 550-560.

Stavely, J .R. 1983. A rapid technique for Inoculation of Phaseolus vulgaris with multiple
pathotypes of Uromyces phaseoli. Phytopathology 73: 676-679.

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

Tar’an B., T.E. Michaels, and KP. Pauls. 2002. Genetic mapping of agronomic traits in
common bean. Crop Sci. 42: 544—556.

Terpstra, K.A., E.M. Wright, and J .D. Kelly. 2006. Protocol for visualizing sequence related
amplified polymorphism (SRAP) and target region amplified polymorphism (TRAP)
markers on agarose gels. Annu. Rep. Bean Improv. Coop. 49:191-192.

Tohme, J ., D.O. Gonzalez, S. Beebe, and M.C. Duque. 1996. AF LP analysis of gene pools
of a wild bean core collection. Crop Sci. 36: 1375-1384.

USDA. 2008. [Online] Soymap. Available at:
http://bldg6.arsusda.gov/cregan/soy_mapl .html (verified July 23, 2008).

94

Vallejo, V.A. and J .D. Kelly. 2008. Molecular tagging and genetic characterization of
alleles at the Co-l anthracnose resistance locus in common bean. Icfai Univ. J. Genet.
Evol. 1: 7-20.

Van Ooijen, J .W. and RE. Voorrips. 2001. JoinMap Version 3.0: Software for the
Calculation of Genetic Linkage Maps, CPRODLO Plant Research International,
Wageningen, The Netherlands.

Voorips RE. 2002. Mapchart version 2.2: Windows software for the graphical
presentation of linkage maps and QTLs. J. Hered. 93:77-78.

Wallace, D.H., J .P. Baudoin, J .8. Beaver, D.P. Coyne, D.E. Halseth, P.N. Masaya, H.M.
Munger, J .R. Myers, M. Silbemagel, K.S. Yourstone and R.W. Zobel. 1993.

Improving efficiency of breeding for higher crop yield. Theor. Appl. Genet. 86: 27-
40.

Walters, K.J., G.L. Hosfield, M.A. Uebersax and J .D. Kelly. 1997. Navy bean canning
quality: correlations, heritability estimates and randomly amplified polymorphic
DNA markers associated with component traits. J. Amer. Soc. Hort. Sci. 122: 338-
343.

Wang S., C. J. Basten, and Z.B. Zeng. 2007. Windows QTL Cartographer 2.5. Department
of Statistics, North Carolina State University, Raleigh, NC.

Wassimi, N.N., G.L. Hosfield, and M.A. Uebersax. 1990. Inheritance of physico-chemical
seed characters related to culinary quality in dry bean. J. Amer. Soc. Hort. Sci.
115:492—499.

Xu, Y. and J .H. Crouch. 2008. Marker-assisted selection in plant breeding: From
publications to practice. Crop Sci. 48: 391-407.

Yu, K., 8.]. Park, and V. Poysa. 1999. Abundance and variation of microsatellite DNA
sequences in beans (Phaseolus and Vigna). Genome 42: 27-34.

95

Chapter 3: Use of TRAP markers to map resistance to a new race of common bean rust

in Michigan

Abstract

A recombinant inbred line (RIL) population from the cross of the black bean
cultivar ‘Jaguar’ and the breeding line 115M was used to map resistance to a new race of
common bean rust (Uromyces appendiculatus). The pathogen was isolated during 2007
from a cultivar possessing Ur-3, a gene which previously conditioned resistance to all
indigenous races of bean rust in Michigan. A differential series was used to characterize
the isolate and a total of 96 RILs were inoculated in the greenhouse and their reaction to
the pathogen was evaluated based on pustule size. SSR, TRAP, SRAP, and phenotypic
markers were used to create a genetic map of the population, including a locus that
conditioned rust resistance. The isolate was characterized as race 3:22 and its virulence
against Ur-3 was confirmed by the susceptible reaction of the differential cultivar
‘Aurora’. In the RIL population, rust resistance was associated with 115M and mapped
to linkage group B4. The locus was flanked by two TRAP markers, both at an
approximate distance of 3 cM. These results support the utility of TRAP markers to tag
disease resistance loci and provide a valuable source of rust resistance in a black bean

adapted to Michigan.

96

Introduction

Common bean rust caused by the hypervariable fungal pathogen Uromyces
appendiculatus (Pers.:Pers.) Unger severely limits common bean (Phaseolus vulgaris L.)
production worldwide (Stavely and Pastor-Corrales, 1989). Bean rust has the potential to
reduce yields by 25 to nearly 100 percent in susceptible cultivars (Mmbaga et al., 1996b),
severely limiting dietary protein and nutrients in developing countries and causing economic
losses in developed areas (Broughton et al., 2003). Strategies to manage this pathogen
include crop rotation, residue management, adjustment of planting date, use of fungicides,
and host plant resistance (Mmbaga et al., 1996b). However, due to the varied effectiveness
of cultural practices and expense or inaccessibility of fungicides, genetic resistance to rust
remains the preferred management strategy to prevent crop losses (Steadman et al., 2002).

Virulence diversity in U. appendiculatus was first described by Harter et al. (1935, as
cited by Stavely et al., 1994). Since that time, over 300 races have been reported worldwide
(Mmbaga et al., 1996a). Following the adoption of an international differential series
(Stavely et al., 1983), 90 unique races have been verified and catalogued (Stavely, 2000).
Currently, a revised differential series of twelve bean cultivars, six each from the two bean
gene pools, is used to classify the variability of the rust pathogen (Steadman et al., 2002).
Each cultivar in this series corresponds to a binary value, and the binary values of the
differentials susceptible to a particular isolate are summed to describe its virulence. This
classification system more accurately reflects the gene pool differences of rust isolates and
resistance genes than the previous differential series used to describe pathogenic variation of
the rust pathogen.

Specific races of rust exhibit patterns of virulence that mirror the division between the

Andean and Middle American bean gene pools, suggesting a history of co-evolution between

97

the rust pathogen and its host (Pastor-Corrales, 2004; Acevedo et al., 2008). Molecular
analysis of the pathogen has also confirmed this relationship (Araya et al., 2004).

Breeding for genetic resistance to rust is complicated by the variability of rust
populations and the rapid breakdown of major resistance genes deployed in cultivars.
Pyramiding multiple resistance genes originating in both the Andean and Middle American
gene pools into a single cultivar provides the broadest and most durable resistance to this
pathogen (Stavely, 2000). Incorporation of varied resistance characteristics (specific, slow
rusting, reduced pustule size, age-dependent resistance, and pubescence) also decreases the
likelihood that the rust pathogen will defeat a resistance gene (Miklas et al., 2005). Despite
the development of adapted germplasm which possesses two resistance genes from each gene
pool with complementary resistance spectrums, many commercial cultivars carry only single
resistance genes (Pastor-Corrales et al., 2005).

Nine named resistance genes and four unnamed genes have been characterized,
tagged with RAPD or SCAR markers, and mapped to five linkage groups of common bean
(reviewed by Miklas et al., 2006a). Pastor-Corrales et a1. (2008) mapped and tagged an
additional unnamed gene from P1260418 that confers resistance to all but one known race of
U. appendiculatus. At least one rust resistance gene (Ur-5) is inherited as a cluster of five
tightly linked loci (Stavely, 1984), and others such as Ur-3 may consist of similar complex
clusters of tightly linked genes (Miklas et al., 2006a). To date, all rust resistance genes
characterized are dominantly inherited.

Previously, the Ur-3 gene conditioned rust resistance to all rust races found in the
state of Michigan (Pastor-Corrales et al., 2007). However, during 2007, rust was observed on
the leaves and stems of the cultivars ‘Jaguar’, ‘Merlot’, and ‘Vista’ which possess the Ur-3
gene. The objectives of the present study were to 1) characterize the new isolate of rust

collected in 2007 using the rust differential series 2) validate the reaction of current bean

98

cultivars including ‘Jaguar’ and 115M, the parents of a RIL mapping population, to the new
rust isolate 3) map the source of rust resistance in 115M using the RIL population and 4)
identify TRAP markers associated with rust resistance.

Materials and Methods

Samples of infected leaves with sporulating pustules were collected from the cultivars
‘Jaguar’, ‘Merlot’, and ‘Vista’ by G.V. Vamer in Tuscola county, MI in the fall of 2007. A
spore suspension was prepared from these samples and used to inoculate the same three
cultivars, along with the susceptible cultivar ‘Othello’ and the breeding line 115M in the
MSU greenhouse according to the methods of Stavely et al. (1983). Spores collected from
‘Jaguar’, ‘Merlot’, and ‘Vista’ were used to inoculate ‘Othello’to obtain additional inoculum
to facilitate screening the differential cultivars for rust reaction.

Preliminary characterization of this unknown race of common bean rust was
accomplished by inoculating the differential series of twelve cultivars proposed by Steadman
et al. (2002). This series included six Andean and six Middle American genotypes that
together possessed the characterized rust resistance genes Ur-3 thru Ur-I3 (Table 3.1). The
rust evaluation scale proposed by Stavely et al. (1983) was used to rate the reaction of
cultivars to the rust pathogen on a scale of 1=immune to 6=very susceptible. If more than
one pustule size was observed, the most common pustule size is reported first, followed by
the less frequent pustule size. The binary values associated with each susceptible
(predominant presence of grade 4 or greater pustule size) differential were summed for both
the Andean and Middle American differentials to determine the race of the isolate as
described by Steadman et al. (2002).

A population of 96 RILs derived from the cross ‘Jaguar’ (susceptible) by 115M
(resistant) was also inoculated to enable mapping of the rust resistance segregating in this

black bean population. A minimum of four plants were inoculated per RIL. The genetic map

99

constructed with JoinMap 3.0 for this population consisted of 119 loci including 62 TRAP,
19 SRAP, 36 SSR, and 2 phenotypic markers spanning 460cM of the common bean genome.
Results

Reaction of cultivars

This isolate of common bean rust produced a susceptible, large pustule reaction in the
cultivars ‘Jaguar’, ‘Merlot’, and ‘Vista’ (Table 3.2). These cultivars carry the Ur-3 resistance
gene and have previously been resistant to all known races of rust in the state of Michigan.
‘Aurora’, the rust differential that possesses Ur-3, and the susceptible check ‘Othello’ both
showed a similar large pustule reaction following inoculation with this isolate (Table 3.1,
3.2). These results confirmed the virulence of this isolate on cultivars with resistance
conditioned by Ur-3, representing the first report of the breakdown of this resistance source
in the state of Michigan. ‘Tacana’, the recurrent parent of 115M, exhibited a resistant small
pustule reaction to this isolate. Additional cultivars screened for rust resistance were
susceptible except the black bean cultivar ‘Shania’ which was heterogeneous and exhibited
both resistant and susceptible plants (Table 3.2).

Characterization

When the complete differential series proposed by Steadman et al. (2002) was
inoculated with this isolate, five of the twelve cultivars displayed a susceptible reaction
(pustule size 2 4; Table 3.1). The Middle American resistance genes Ur-3 and Ur-7
conditioned susceptible and moderately susceptible reactions, respectively. The Andean
resistance genes Ur-6 and Ur-13 also produced moderately susceptible or susceptible
reactions when challenged with this isolate. ‘Montcalm’, an important dark red kidney
variety grown in Michigan, was also susceptible. Conversely, Ur-5 and Ur-I I were immune
to this isolate. The remaining cultivars in the series were moderately resistant with varying

frequencies of small pustule reaction. The isolate was classified as race 3:22 based on the

100

summation of the binary values associated with the two Middle American and three Andean
susceptible differentials.

Mapping and Tagging

Race 3:22 conditioned a small pustule resistant reaction on 115M and a highly
susceptible large pustule reaction on ‘Jaguar’. The RIL population (‘Jaguar’/1 15M) and
genetic map provided an opportunity to map the resistance present in 115M. 96 RILs were
evaluated; 63 were resistant, 18 were heterogeneous with both resistant and susceptible plants
and 15 were susceptible (Table 3.3). These data significantly differed (p<.0001) from the
expected 1:1 resistant to susceptible ratio for a single gene trait.

Using the data (Table 3.3) as a phenotypic marker, resistance to race 3:22 was
mapped to linkage group B4b, which corresponds to the lower end of linkage group B4 of the
bean consensus map (F reyre et al., 1998). Rust resistance was flanked by two TRAP
markers, 3cM from F7Rl.150 and 3.3cM from F 15R10.58O (Table 3.3, Figure 3.1). F7R1
amplified a 150bp fragment that co-segregated in coupling phase with rust resistance while
Fl 5R10 produced a 580bp fragment that co-segregated in repulsion phase. All markers on
this linkage group exhibited skewed segregation ratios that favored the 115M allele.
Discussion

Reaction of cultivars

The susceptible reaction of ‘Jaguar’, ‘Merlot’, and ‘Vista’ to the isolate of U.
appendiculatus confirmed the virulence of the rust pathogen has evolved fiom that previously
reported in Michigan. These cultivars possess Ur-3, a resistance gene that previously
conditioned resistance to all common bean rust found in the United States (Pastor-Corrales et
al., 2007). This discovery suggests commonly grown bean cultivars are vulnerable to rust
infection and significant yield reductions could occur in the future if environmental

conditions favor disease development. This isolate was collected from a single field late in

101

the 2007 season, so no conclusions can be drawn about the distribution of the race in the
state. However, no other sources have reported rust on cultivars possessing Ur-3, suggesting
the distribution of this race is limited. Collecting additional isolates from several fields in the
region in future years would provide additional information about the persistence and
distribution of this new race.

Characterization

The complete characterization of the rust isolate with a differential series revealed
five of the twelve cultivars were susceptible to race 3:22, but others including those
possessing Ur-5 or Ur-II were immune or resistant. These results supported the virulence of
this isolate against Ur-3, the resistance gene present in the differential cultivar ‘Aurora’. This
gene is used almost exclusively in rust resistant cultivars grown in Michigan, leaving the
bean crop vulnerable to losses from rust.

The immunity conferred by Ur-5 or Ur-I I was encouraging as these genes have been
widely effective against numerous rust races in previous studies (Stavely, 2000). The current
results indicate these were the most effective resistance genes when challenged with race
3:22. Ur-5 represents a tightly linked block of single dominant genes (Stavely, 1984) that
confers resistance to 70 of 90 rust isolates in the USDA-ARS rust collection (Stavely, 2000).
Ur-I I confers resistance to all but one isolate, race 108, and is linked to Ur-3 which
conditions resistance to 44 races including race 108 (Stavely, 2000). Ur-5 has been tagged
with RAPD (Haley et al., 1993) and SCAR (Melotto and Kelly, 1998) markers that are
effective for MAS in a range of genetic backgrounds while RAPD markers linked to Ur-ll
(Johnson et al., 1995) have been less reliable (Kelly and Miklas, 1999). Despite the
availability of reliable markers, Ur-5 has been underutilized in breeding for rust resistance
while Ur-I I has been more widely deployed in recent years (Kelly et al., 2003). Based on

their resistance to a broad range of rust races including race 3:22, bean breeders should

102

consider pyramiding one or both of these additional resistance genes into their breeding
programs to maintain complete resistance to all rust races in Michigan.

Mapping and Tagging

Resistance to race 3:22 was mapped to linkage group B4b in the ‘Jaguar’/l 15M RIL
population. Based on alignment of this linkage group with the bean consensus map, this
location corresponds to a region on the lower end of linkage group B4 where multiple disease
resistance genes have previously been identified (Miklas et al., 2006b; Pastor-Corrales et al.,
2008). This location suggests the Ur-gene conditioning resistance to race 3:22 was inherited
from 115M, since rust resistance in ‘Jaguar’ is conditioned by Ur-3 which resides on linkage
group B11 (Kelly et al., 2001; Miklas et al., 2006b).

Two rust resistance genes, Ur-5 and Ur-Dorado-108 reside in this region of B4
(Miklas et al., 2000) and could condition the resistance observed in 115M. However, the
immune reaction of the cultivar ‘Mexico309’ which carries only Ur-5 was not consistent with
the small pustule resistance of 115M. This suggests that Ur-5 may not condition the resistant
reaction of 115M to race 3:22. ‘Dorado’, the original source of Ur-Dorado-108 (Miklas et _
al., 2000), also proved susceptible to race 3:22 (Table 3.2). ‘Dorado’ appears in the pedigree
of ‘Tacana’ (Lopez-Salinas et al., 1997), the recurrent parent of 115M, suggesting Ur-
Dorado-108 could have been inherited by ‘Tacana’ and subsequently 115M. However, the
susceptible reaction of ‘Dorado’ suggests it does not confer the resistance observed in the
RIL population. Thus the small pustule resistance exhibited by both ‘Tacana’ and 115M
suggests rust resistance is conditioned by the same Ur-gene or genes in both cultivars, but
further work will be necessary to precisely identify this locus.

The skewed ratio of resistant to susceptible lines was unexpected. This ratio suggests
either the presence of a more complex, multi-locus resistance or an unintentional selection

bias that inadvertently favored resistant genotypes. Closer examination of the genotypic data

103

for each of the markers mapped to linkage group B4b revealed that all markers were skewed
in favor of the 115M allele and the segregation ratios were similar to that observed for rust
resistance. In contrast, markers residing on linkage group B4 of the ‘Jaguar’ll 15M map,
which corresponds to the upper end of linkage group B4 on the consensus map, are not
skewed significantly. The reason for the skewed segregation ratio at one end of the linkage
group but not the other remains unclear. A QTL for lodging co-located with rust resistance
(Table 2.13, Figure 2.2) raises further questions about the implications of the increased
frequency of the 115M allele. The location of that QTL for lodging in a genomic region
skewed toward 1 15M agrees with the increased frequency of RILs that lodge similarly to
115M.

The tagging of the resistance source present in 115M with flanking TRAP markers
supports the conclusion of Miklas et al. (2006b) who suggested the utility of TRAP markers
to tag disease resistance genes of common bean. These markers present a usefiil tool to use
for indirect selection of rust resistance. F 7R1 . 1 50, linked in coupling phase with rust
resistance at a distance of 3.1cM, would be the best marker to use for marker assisted
selection. Only three recombinants were observed when comparing the presence of
F7R1.150 with rust resistance. Six recombinants were observed between F 15R10.580 and
rust resistance. This suggests F15R10.580 is less tightly linked to the resistance loci,
although still valuable when used in addition to F7Rl .150 to flank the region surrounding the
resistance gene.

Since several disease resistance genes have been mapped to the same region of B4,
these markers may be useful in selecting for other resistance genes within the cluster if their
linkage can be verified. For example, F7Rl . 150 was also present in ‘Mexico309’, which
implies the marker is linked to Ur-5. The $119 SCAR marker linked to Ur-5 (Melotto and

Kelly, 2000) was present in 115M, but did not segregate in the RIL population, which

104

prevented the mapping of both F7R1.150 and 8119 markers in relation to each other.
Screening additional 115M plants with 8119 revealed the marker was not present in all cases,
suggesting 115M is heterogeneous at this marker locus. A previous study by Miklas et al.
(2000) attempted to map Ur-5 in relation to Ur-Dorado-108, but found that only the S119
marker and not the Ur-5 gene segregated in a ‘Dorado’ by XAN159 population. These
results underscore the difficulty in reconciling phenotypic and genotypic data at complex
resistance gene clusters. Additional markers mapped to this region in the future will help
verify the relationship between genes at this locus. Allelism tests between 115M, ‘Dorado’,
and ‘Mexico309’(Ur-5) will also be necessary to determine the precise relationship among
the resistance loci in these cultivars.
Conclusions

Breakdown of previously effective resistance to any pathogen presents a challenge to
breeders to identify alternative solutions. Although the Ur-3 gene was overcome in Michigan
bean fields during the 2007 growing season by a new rust race 3:22, the long-term
implications of this discovery remain uncertain. This knowledge should serve as a reminder
that pathogen populations are continually evolving, and maintaining successful genetic
resistance requires continual effort. Further work is needed to determine the precise identity
of the resistance gene conditioning resistance in 115M. Allelism tests with Ur-5 and Ur-
Dorado-108 should be performed in the future. Additionally, TRAP markers F7R1.150 and
F15R10.5 80 that flanked and co-segregated with rust resistance should be considered for use
in marker assisted selection to incorporate resistance to the new rust race 3:22 in future bean

cultivars for production in Michigan.

105

.Agovcfigotog 0.25.6 ESE—=0 03:38.5 53> 3308mm mo:_m> .933 we corms—82m

co woman BEEoE fl comm 636388 3 papacy. 2s v N 82.; 4:82.... .: c282: “5:ch amo— 332—8 .5382 “:26on 602m
.388 .3 Ho ~5vaon 9 @5808 03:383.»..an 8 ESmEBH— an .888 mcouoaomH

.383 .3 8 5:65on 8 wEEooom 83:3 REESE? E :8on 28m oogmmmoc “may:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3 Ease: m3 233: _ :-S a0 83:5
@ am at: G: 53 28 528 3 929$ Ga ozu
3 2.5 «-5 an cm on _ T: as am 852
hop 3. 5.555: € 55282 m +35 :3 m2 85:
@ 3. 2.2: Q 3255 €553. @ 3 rs E 505...
3 I: 3 55:3 ram @ mm 55 A: o: .20
mwcouomom +oco©é o=_w> 099056 S322. ficouomom +050;— o=_w> 090050
5.55 585 5855... 0:32:
A2 5:58

283% Sea 380:8 “we :39 .8888 mo mmnm 88 53» 682305 mags—:0 :89 5888 fizaeobmu Q Co couoaom ._.m 033.

106

 

Table 3.2. Reactions of selected bean cultivars to inoculation with U. appendiculatus race
3:22. '

 

 

Cultivar Reaction Class
Eclipse 5,6 Black
Jaguar 5,6 Black
Shania 3/5* Black
Zorro 5,6 Black
Vista 5 Navy
Matterhorn 5,4 Great Northern
Lapaz 5 ,6 Pinto
Othello 5,6 Pinto
Merlot 5,4 Red
Dorado 5,6 Red

 

*Heterogeneousz Four resistant plants and
two susceptible plants

107

Table 3.3. Reactions of 96 RILs from a ‘Jaguar’ by 115M population to U. appendiculatus
rust race 3:22 and presence or absence of two TRAP markers (F 7R1, F15R10).

 

F15R10

RUST F7Rl

F15R10 RIL

FISRIO RIL RUST F7R1

RIL RUST F7Rl

 

67

34
35
36
37
38
39
4O
41

68

69

7O
71

72
73

74
75

42

76
77
78

43

10
ll
12
13
14
15
16
17
18
19
20
21

44
45

79

46

80
81

47

48

82
83

49
50
51

84
85

52
53

86
87

54
55
56
57
58
59
60
61

88
89
9O
91

22

23

24
25

92
93

26
27
28

94
95

62
63

29
3O
31

96

115M R

A

Jaguar S

A

65

32
33

A

Tacana R

Heterogeneous

66

Absent, P=Present, R=Resistant, S

 

Susceptible, H=

108

B4b

 

 

0.0 IAC66
7.1 F7R1.150
10.1 Rust3:22

 

 

13.4 F15R10.580
16.3 M3E6.320

 

 

 

 

22.1 F8R5.650

Figure 3.1. Linkage group B4b of the ‘Jaguar’ by 115M recombinant inbred line population
which contains the resistance locus for U. appendiculatus race 3:22.

109

Literature Cited

Acevedo, M., J .R. Steadman, J .C. Rosas, and J. Venegas. 2008. Coevolution of the bean
rust pathogen Uromyces appendiculatus with its wild, weedy, and domesticated

hosts (Phaseolus spp.) at a center of diversity. Annu. Rep. Bean Improv. Coop.
5 1 : 22-23.

Araya, C.M., A.T. Alleyne, J .R. Steadman, K.M. Eskridge, and DP. Coyne. 2004.
Phenotypic and genotypic characterization of Uromyces appendiculatus fi'om
Phaseolus vulgaris in the Americas. Plant Dis. 88: 830-836.

Broughton, W.J., G. Hernandez, M. Blair, S. Beebe, P. Gepts, and J. Vanderleyden.
2003. Beans (Phaseolus spp.) -— model food legumes. Plant Soil 252: 55-128.

Haley, S.D., P.N. Miklas, J .R. Stavely, J. Byrum, and J .D. Kelly. 1993. Identification of
RAPD markers linked to a major rust resistance gene block in common bean.
Theor. Appl. Genet. 86:505-512.

Hosfield, G.L., G.V. Varner, M.A. Uebersax, and J .D. Kelly. 2004. Registration of
‘Merlot’ small red bean. Crop Sci. 44: 351-352.

Johnson, E., P.N. Miklas, J .R. Stavely, and J .C. Martinez-Cruzado. 1995. Coupling and
repulsion RAPD markers for marker-assisted selection of a rust resistance gene in
common bean. Theor. Appl. Genet 90: 659—664.

Kelly, J .D. and P.N. Miklas. 1999. Marker assisted selection. In: Singh SP (ed.), Common
Bean Improvement in the Twenty-First Century, 93-124. Kluwer, Dordrecht, the
Netherlands.

Kelly, J.D., G.L. Hosfield, G.V. Vamer, M.A. Uebersax, and J. Taylor. 2001.
Registration of ‘Jaguar’ black bean. Crop Sci. 41 :1647.

Melotto, M. and J.D. Kelly. 1998. SCAR markers linked to major disease resistance
genes in common bean. Annu. Rep. Bean Improv. Coop. 41 :64-65.

Miklas, P.N., J .D. Kelly, S.E. Beebe, and M.W. Blair. 2006a. Common bean breeding for
resistance against biotic and abiotic stresses: From classical to MAS breeding.
Euphytica 147:105-131.

Miklas, P.N., J. Hu, N.J. Grunwald, and KM. Larsen. 2006b. Potential application of

TRAP (Targeted Region Amplified Polymorphism) markers for mapping and
tagging disease resistance traits in common bean. Crop Sci. 46:910-916.

110

Miklas, P.N., R. Delorme, V. Stone, M.J. Daly, J.R. Stavely, J .R Steadman, M.J. Bassett,
J .S. Beaver. 2000. Bacterial, fimgal, and viral disease resistance loci mapped in a
recombinant inbred common bean population (‘Dorado’/XAN 176). J. Amer. Soc.
Hort. Sci. 125: 476-481.

Mmbaga, M.T., J .R. Steadman, and KM. Eskridge. 1996a. Virulence patterns of
Uromyces appendiculatus from different geographical areas and implications for

finding durable resistance to rust of common bean. Phytopathology 144: 533-
541.

Mmbaga, M.T., J .R. Steadman, and J .R. Stavely. 1996b. The use of host resistance in
disease management of rust in common bean. [PM Rev. 1: 191-200.

Pastor-Corrales, M.A. 2004. Review of coevolution studies between pathogens and their
common bean hosts: Implication for the development of disease-resistant beans.
Annu. Rep. Bean Improv. Coop. 47: 67-68.

Pastor-Corrales, M.A., A.C. Aime, and J .R. Steadman. 2005. Guiding the development
of common bean cultivars with durable rust resistance based on the genetic
diversities of the pathogen and its host. Proc. 1st Intl. Edible Legume Conf. April
17-25, 2005, Durban, S. Africa.

Pastor-Corrales, M.A., J .D. Kelly, J.R. Steadman, D.T. Lindgren, J .R. Stavely, and DP.
Coyne. 2007. Registration of six great northern bean germplasm lines with

enhanced resistance to rust and bean common mosaic and necrosis potyviruses. J.
Plant Reg. 1: 77-79.

Pastor-Corrales, M.A., P.A. Pereira, K. Lewers, R.V. Brondani, G.C. Buso, M.A.
Ferreira, and W.S. Martins. 2008. Identification of SSR markers linked to rust
resistance in Andean common bean P1260418. Annu. Rep. Bean Improv. Coop.
51: 46-47.

Stavely, J .R. 1984. Genetics of resistance to Uromyces phaseoli in a Phaseolus vulgaris
line resistant to most races of the pathogen. Phytopathology 74: 339-344.

Stavely, J .R., Freytag, G.F., Steadman, J .R., Schwartz, HF. 1983. The 1983 Bean Rust
Workshop. Annu. Rep. Bean Improv. Coop. 26:iv-vi.

Stavely, J.R., and M.A. Pastor-Corrales. 1989. Rust. In: HR Schwartz and M.A.
Pastor-Corrales (eds.), Bean production problems in the Tropics 159-194. Centro
Intemacional de Agricultura Tropical, Cali, Colombia.

Stavely J.R., J .D. Kelly, and K.F. Grafton. 1994. BelMiDak-rust-resistant navy dry bean
germplasm lines. HortScience 29:709—710.

111

Stavely, J .R. 2000. Pyramiding rust and viral resistance genes using traditional and
marker techniques in common bean. Annu. Rep. Bean Improv. Coop. 43:1-4.

Steadman, J .R., M.A. Pastor-Corrales, and J .S. Beaver. 2002. An overview of the 3rd
bean rust and 2nd bean common bacterial blight International workshops. March
4-8, 2002, Pietennaritzburg, South Africa. Annu. Rep. Bean Improv. Coop.
45:120-124.

112

Appendix A: Indirect screening for color loss in two black bean populations
Introduction

Black bean (Phaseolus vulgaris L.) is especially prone to loss of seed coat color
during the thermal processing prior to canning (Bushey and Hosfield, 2000; 2004). This
color leaching results in a canned bean product that appears brown or washed-out, and
visually unappealing to consumers. Prior work undertaken to better understand the
physiology and genetics associated with this leaching suggests the value of a rapid screen to
detect differences between black bean breeding lines at an early generation when seed
quantities are limited. Lu et al. (1996), Ruengsakulrach et al. (1991) and Shellie and
Hosfield (1991) proposed screening methods that evaluated various physical or chemical
characteristics of a small seed sample and correlated the results with those of traditional
canning protocols. To date, none of these early generation selection methods have been
widely implemented, suggesting the need for continued research in this economically
important aspect of black bean breeding.

Another indirect screening method, the soak water color test, was recently developed
by Bushey and Hosfield (2007). Their method requires ten seeds per line, along with
minimal lab facilities and time, to indirectly screen for color retention in black bean. The
objectives of the current study were twofold. The first was to purify and re-establish the
original populations used by Bushey and Hosfield (2007) to develop this screening technique
as a genetic resource to facilitate future study of black bean color retention. The second was
to verify the reproducibility of the technique in other germplasm for future use in breeding

for color retention in black beans.

113

Materials and Methods

Seed of two black bean populations previously established by G.L Hosfield was
obtained from USDA-ARS. Population 1 (‘Black Magic’ x ‘Shiny Crow’) consisted of 93
recombinant inbred lines (RILs), while population 2 (‘Black Magic’ x ‘Raven’) consisted of
106 RILs. ‘Black Magic’ and ‘Raven’ are black beans with a dull seed coat luster and ‘Shiny
Crow’ has a shiny seed coat. Several of the bulks in population 1 segregated for both shiny
and dull seed within a line, so a single seed descent purification process was immediately
undertaken for each line in the MSU greenhouse during spring 2007. At maturity, single
plant rows were established at the Saginaw Valley Bean and Beet Research Farm near
Saginaw, MI. Rows were harvested as bulks and data collected on each line included: total
seed weight, seed coat (dull or shiny), lOO-seed weight, and dry seed color (measured with
HunterLab LabScanXE, Reston, VA.). Two samples of 10 seeds each were then taken from
each line and tested using the soak water color test as described by Bushey and Hosfield
(2007). Soak water color was determined both as a luminosity (L) value using a HunterLab
UltraScanXE and as a visual rating from 1=clear to 5=very dark.

In addition, eleven of the lines in population 1 that were segregating for shiny and
dull seed coats were randomly chosen for use in creating a group of near isogenic lines
(NILs) differing in seed coat luster. The only differences in procedure from that described
above were three shiny and three dull seeds for each of the eleven lines were planted in the
greenhouse and then bulked prior to planting in the field.

Results and Discussion

The amount of seed obtained for each line within the populations varied from 53 to
963g, with most lines producing sufficient seed to facilitate future work in replicated field
plots. The few lines that produced little seed were the result of plant rows containing very

few plants. Seed size, measured as lOO-seed weight, ranged from 15.7 to 27 .0g. On average,

114

population 1 had larger seed size, with a mean of 21.2g, while population 2 was slightly
smaller with a mean of 19.7g (Table A. 1). As expected, population 1 segregated by line for
seed coat luster; 39 lines had shiny seed coats, while 54 were dull. All lines in population 2
had dull seed coats, as expected.

Dry seed color was equivalent between RILs in the two populations. However,
differences between shiny versus dull seed coats became apparent in the soak water color
test. In this test, a lower luminosity value for the soak water indicates more color loss from
the bean, thus a higher luminosity value is more desirable. Population 1, where 39 lines had
shiny seed coats, had an average luminosity of 75.6 and visual rating of 3.1 (Table A.1). In
contrast, population 2, with all dull seed coats, had an average luminosity of 61 .3 and visual
rating of 4.5. As shown in Table A.1 both populations had similarly low luminosity and
visual values, but population 1 had higher values reflecting the presence of lines with shiny
seed coats that did not leach as much color. As expected, the lines with dull seed coat luster
in population 1 lost more color than those with shiny seed coats. However they retained
more color than the lines in population 2 which all had dull seed coat luster. These data
suggest that selecting for dull seed coat luster from progeny derived from crosses between
dull and shiny black beans may improve color retention.

Similar trends were observed in seed color when comparing the 11 NILs differing
only in seed coat luster. Dry seed color was very similar between the two groups, while soak
water color was much lighter in the shiny group that leached less color (Table A2). The
range in soak water color resembled the range in values measured in population 1, suggesting
that the NILs reflect the range in color retention of the RIL population. Seed size varied from
an average of 22.2 to 25.6g/ lOO-seed for the shiny and dull groups of NILs, respectively.
This was unexpected, since the mean seed sizes for the dull and shiny groups of RILs derived

from population 1 were the same, and smaller than either group of NILs. One explanation for

115

these differences is that the initial lines used for development of NILs were chosen at random
from population 1, with no selection based on seed size. Since only eleven of 93 lines were
used, if several larger seeded lines were chosen, they would have easily increased the mean
of the NILs, whereas the average for the RILs represents the full variability of the entire
population. However, there is no apparent reason for the difference between the shiny and
dull groups of NILs, and the results for the larger group of RILs suggests there is not a
relationship between seed size and seed coat luster.

Conclusions

These data demonstrate that much of the variation for color loss originally present in
two populations was maintained throughout the process of purification and renewal. The
genetic variation for black bean color retention presents a unique opportunity for continued
study of this economically important trait. Individuals in population 1 that have a dull seed
coat facilitating water uptake, but a high luminosity value for soak water color would be
particularly interesting to breeders (Table A.1). These lines possessed improved color
retention when compared with the average color retention of population 2. These results
suggest that black beans with a shiny seed coat luster are useful for improving color retention
in black beans with dull seed coat luster and this strategy may provide an opportunity to
improve processing. In contrast, the luminosity and visual scores in population 2 underscore
the difficulty that breeders must confront in retaining processed seed color when crossing two
black bean lines with dull seed coats.

The group of NILs developed represent a useful genetic tool for studying other
changes associated with differences in seed coat luster. The range in luminosity among these
lines suggest the NILs reflect the variability for color retention present in population 1.
While it is evident that shiny or dull seed coats cause beans to take up water differently and

therefore influence their color retention, future analysis at the molecular level is needed to

116

elucidate additional genetic differences. Such studies will provide practical knowledge

useful for breeding future black bean cultivars with improved processing characteristics.

ll7

Agom .2050: can .35sz View .6on Hue—ou— chsfl 13$ >

 

 

 

 

 

5.33.2 Nua— w.wm-o.: N._~ 95.52 m. .N N._N va Emma? noomée
c.m-m.m m6 o.m-o.~ o.m o.m-m._ QM _.m 3.55.8200 53? xmom
934.94 m. E c.3436 Qmw wdqmév N43 9: Ad .200 .533 xmom
c.mm-~.£ a $34.12 ”3 v.mm-m.: 93 md— Ad 8.00 coom ED
owcmx :82 amend :82 owned :82 :82 tab.
53% x 232 £35 LEEm =30 =Eo>O
«1% 380 haw—m x omwsE :35 A Sufismom

 

.Lmoh 8200 .2855 xaom. 2: :0 women coup—82 823 .8 wcuawocwom mcotflaom oz: E 83.: 26 .59 mowcfi use manor: :8... ._.< 053..

18

1

Table A2 Trait means and ranges for five traits measured on 11 pairs of NILs selected from
population 1 (Black Magic x Shiny Crow) on basis of dull or shiny seed coat.

 

 

M! M
Trait Mean Range Mean Range
Dry Seed Color (L) 19.8 18.0-22.2 19.2 18.2-20.3
Soak Water Color (L) 85.5 73.8-98.5 61.9 51.8-78.7

Soak Water Color-Visual 1.9 1.0-3.5 4.5 3.0-5.0
lOO-Seed Weight (g) 22.2 17.8-26.1 25.6 22.2-31.0

 

Visual rating: 1=clear 5=very dark (Bushey and Hosfield, 2007)

119

Literature Cited

Bushey, S.M., G.L. Hosfield, and CW. Beninger. 2000. Water uptake and its relationship to
pigment leaching in black beans (Phaseolus vulgaris L.). Annu. Rep. Bean Improv.
Coop. 43: 104-105.

Bushey, S.M., Harris, L.S., and G.L. Hosfield. 2004. Development of an early generation
test for predicting color loss of black beans. Annu. Rep. Bean Improv. Coop. 47:
139-140.

Bushey, S.M., and G.L. Hosfleld. 2007. A test to predict color loss in black bean during
thermal processing. Annu. Rep. Bean Improv. Coop. 50: 41-42.

Lu, W., K.C. Chang, K.F. Grafton, and RB. Schwarz. 1996. Correlations between physical
properties and canning quality attn'butes of navy bean (Phaseolus vulgaris L.).
Cereal Chem. 73: 788-790.

Ruengsakulrach, S., N. Srisuma, M.A. Uebersax, G.L. Hosfield, and LG. Occena. Early
generation screening of navy bean breeding lines by canning quality assessment and
pasting characteristics of bean flour. 1994. J. Food Qual. 17: 321-333.

Shellie, K.C. and G.L. Hosfield. 1991. Genotype x environmental effects on food quality of

common bean: resource-efficient testing procedures. J. Amer. Soc. Hort. Sci. 116:
732-736.

120

Appendix B: Validation of the soak water color test in the ‘Jaguar’ll 15M RIL
population

Introduction

The soak water color test was developed by Bushey and Hosfield (2007) as an
indirect screening method to predict color loss from black beans that inevitably occurs during
the thermal processing associated with canning. Loss of black pigment during canning
results in processed black beans that appear brown and washed out, which makes them
unappealing to consumers, and therefore processors (Bushey et al., 2000). Due to the
importance of color retention, potential black bean cultivars with superior agronomic traits
will be discarded if they fail to produce an acceptable quality canned product (Posa-
Macalincag et al., 2002). The soak water color test provides a means to screen breeding lines
at an early generation before several years are invested to generate enough seed to perform
traditional canning evaluations (Bushey et al., 2004). This method uses as little as 10 seeds
(<3 g) and two hours of time, while at least 100g of seed are needed to evaluate a line by
traditional canning protocols that require more time, supplies, and specialized equipment.
Since the soak water color test had only been used in the two related populations where it
was initially developed, the objective of this study was to determine the reproducibility of the
method in different genetic backgrounds, and assess the correlation between this indirect
method and canning scores, both within a population and among a group of unrelated
breeding lines.
Materials and Methods

Seed from plots grown at the Saginaw Valley Bean and Beet Research Farm
(Saginaw, MI) in 2005 was evaluated using both the soak water color test as described by
Bushey and Hosfield (2007) and by the canning procedure described in detail at the BIC

website: (http://www.css.msu.edu/bic/PDF/Bean%20Processing.pdt). Seed from the same

121

plot was used for both methods and was free of splits or cracks in the seed coat. The 96 RILS
derived from the cross of ‘Jaguar’/ 1 15M were used to compare the methods within a
population. Similarly, 32 of the top yielding lines from the standard black yield trial grown
at the same location in 2006 were used to represent a group of genetically diverse black bean
breeding lines. The soak water color test was performed on two occasions with two
replications each, while visual appearance was based on the average score assigned to a
single can by a group of panelists rated on a scale of: l=undesirable to 7=desirable. Proc
GLM (SAS, 2000) was used to calculate significant differences in color loss due to genotype
and replication. Pearson correlation coefficients for luminosity measured by the soak water
color test and canning score were calculated with Proc Corr (SAS, 2000) using the average
luminosity (L-value) obtained from the two replications.
Results and Discussion

Both the RIL population and the group of breeding lines showed significant
(p<0.0001) variation for color loss and visual appearance, suggesting these groups were
suitable for evaluating a range in seed quality by both the soak water color test and canning
methodologies. Transgressive segregation was observed in the population using both
methods, although none of the RILs had higher visual appearance scores than ‘Jaguar’, even
though several had values lower than 115M (Table B. 1 , Table 3.2). In the group of breeding
lines, considerable variability was also noted, with some lines higher and others lower than
the parents of the RIL population (Table B.3). Unexpectedly, 115M had a higher visual
appearance score than ‘Jaguar’ in 2006. This result was unexpected since 115M generally
has very poor canned bean quality, although inconsistent results in canning quality have
occurred occasionally in other years (data not shown).

The correlation between the results of the soak water color test and the visual

appearance showed different relationships in the two groups. In the population, there was a

122

significant negative correlation between the two methods (Table B.2). This result seems
counterintuitive as it suggests a higher (more desirable) visual appearance score is associated
with a lower soak water luminosity value, and therefore more leaching of color in the soak
water. Intuitively, leaching more color when soaked in hot brine would be an indicator of
color loss that will be exacerbated by the increased temperature and pressure associated with
the canning process. These results support the hypothesis that different breeding lines have
different quantities of anthocyanin pigment in the seed coat, and therefore could leach more
color but still retain a darker seed color following processing.

Salinas-Moreno et al. (2005) compared dry seed luminosity with quantity of
anthocyanins in the seed coat for a diverse group of black beans. Their results indicated a
significant range in anthocyanin content ranging from 10.1 to 18.1 mg/g; however the
difference in dry seed luminosity for these two genotypes was insignificant (L=l7.9 and
18.1). Although only anthocyanins were measured, this study suggests that genotypes with
very similar luminosity values vary greatly in pigment content, so estimating color loss by
measuring the total amount of pigment that is leached into a standard volume of brine may
not be an appropriate measure of actual canned bean color. Further measurements of
variation for anthocyanins within seeds of the same genotype and luminosity of soaked seeds
in a similar type of study would be useful.

Color of the canned bean is only one component of the visual appearance score.
Other factors such as splitting, overall texture, or starchiness of the brine were all considered
by panelists when rating canned samples. These factors were not considered by the soak
water color test, and may explain the lack of a strong correlation with the visual appearance
score. A significant positive correlation between the luminosity values from the two separate
evaluations of the population suggests that this method reproducibly detects differences in

color of the soak water (Table 3.2).

123

In the group of diverse breeding lines from various genetic backgrounds, the results
of the two methods did not correlate significantly (p=0.24), although the relationship between
the two results was still inverse. This result supports the inverse relationship between
luminosity of the soak water and canning score observed in the population but it does not
disprove the hypothesis that pigment content varies by genotype. However, since the results
of the two methods do not significantly correlate, the soak water color test may not be
suitable for comparing genetically diverse germplasm.

In practice, comparing a wide range of unrelated germplasm would represent the
main application of this rapid screening method. However, results with a sample of unrelated
germplasm (Table B.3) suggest this screening method may not reliably predict the canning
quality under these conditions. One of the breeding lines (B04644) that leached the most
color (Table 8.3) has been used as a parent to improve canning quality, as the visual color of
the cooked beans is much blacker than other entries, despite leaching more color into the
soak water. This would suggest that certain lines possess higher pigment levels and appear to
leach more into the soak water, while retaining satisfactory cooked color. An objective
reading of cooked bean color is not possible as the meter gives erroneous results based on
light reflectance from the surface of moist cooked beans. Cooked bean color is rated visually
and panelists have noted the 'blacker' color of cooked samples of the B04644 breeding line
when compared with other black bean lines, based on higher score for visual appearance.
Other lines that leached less (higher L-values) have been discarded based on poor canning
quality as the cooked bean is brown in color and any black pigments in the seed coat have
been lost in the soak water.

Recently van der Merwe et al. (2006) suggested that canning evaluations in the
laboratory very closely predict performance under commercial canning procedures. Due to

the number of factors influencing the quality of canned beans, and the consequences of

124

bringing a new cultivar with inferior quality to the market, canning should be viewed as a

solid investment for breeding programs. The results of the current study indicate the soak
water color test was not appropriate for accurately comparing diverse black bean breeding
lines, and should not be considered for screening diverse germplasm for color loss during

processing.

Conclusions

The limited evaluation of the soak water color test showed this method detects
significant variation in color loss, both within and among genetic backgrounds of black bean.
This method may have some predictive value for evaluating lines within a population, and
the measured luminosity is inversely related to visual canning score. However, the moderate
strength of the correlation suggests caution should be exercised when interpreting the results
of the soak water color test, and canning score should still be considered more informative
since it encompasses all components of canning quality, not just color retention.

A weak and insignificant negative correlation was observed when a group of 32
diverse breeding lines were evaluated with both methods. Among diverse genetic
backgrounds, the soak water color test cannot be considered predictive of color loss during
canning. Many of the lines with above average canning characteristics lost the most color
when measured by this method, suggesting that they had increased levels of black seed coat
pigment. Overall, these results suggest canning should remain the preferred method of

evaluating color retention in black bean breeding programs.

125

Table B. 1. Phenotypic values for seed of the 96 ‘Jaguar’ll 15M recombinant inbred lines
evaluated by both the soak water color test and by visual evaluation of canned bean samples.

Mean Jaguar 115M Range

 

L-value 1 76.4 56.4 75.6 55.5-90.8
L-value 2 67.6 58.3 83.6 42.6-87.0
Canning Score 2.2 3.8 1.7 1.1-3.8

DrLSeed Color 15.9 15.5 16.3 14.8-18.9

Table B.2. Phenotypic correlations between visual canning score and Hunter L value of
leachate from the soak water color test in a population of 96 recombinant inbred lines
developed from the cross ‘Jaguar’/l 15M grown in Saginaw, MI in 2005.

L-value 1 L-value 2 Canning Score

 

L-Value 1

L-Value 2 0.86M

Canning Score -0.41** -O.33**

Dry Seed Color 0.19 0.14 0.01

 

I""‘Indicates significance at P<0.001.

126

Table 8.3. Phenotypic values for seed of the 32 diverse black bean genotypes evaluated by
the soak water color test and by visual evaluation of canned bean samples during 2006.
Black Bean Genotypes L-value Visual Appearance§

 

 

B04596 90.8 2.2
B05069 90.3 2.8
Raven'l' 90.3 NA
801793 89.6 3.0
B04585 89.3 3.2
Domino 88.4 2.3
805024 87.6 2.7
805051 86.9 2.6
B05065 85.0 2.9
B05066 85.0 2.8
B03622 84.8 3.0
B04607 84.6 3.6
804227 84.0 2.7
801741 84.0 2.8
B04610 83.4 3.1
Condor 82.9 4.3
B05070 82.4 3.8
B04561 82.2 2.3
804591 81.5 4.2
Zorro 81.1 4.0
805055 80.7 3.6
115M 80.6 3.8
B04587 79.2 2.8
Jaguar 78.1 3.2
B05054 78.1 3.0
B05041 77.9 3.6
Raven 77.8 2.7
B05040 77.5 3.9
804260 76.4 2.6
805039 76.3 3.3
80464412 76.3 4.1
T-39 75.4 3.1
804644 71.8 2.6
Eclipse 69.1 2.7

 

'1' Grown at Montcalm Research Farm
I Grown in Presque Isle, MI
§ Canning scores range from 1=undesirable to 4=neither undesirable nor desirable to 7=desirable

127

a; 8.9 @943. ON m _ .ho omwvom N.N QM: wmmvom

 

5. 8.8 888 3 8.8 838. 3 3.8 888.
3 8.8 23.8. 3 8.8 835m 3 8.8 83cm
3 8.8 888 E 8.8 83.8. 3 8.8 888.
3 8.8 888 3 8.8 83.8. 8.8 8.8 5.3m
8 8.8 833m 8 8.8 888 3 8.8 888
mm 2.8 8.3.8. 2 8.8 83.8 on 3.8 888
2 8.8 83.8 mm :8 38cm 3 88 83cm
8 8.8 835m 3. 2.8 888 3 8.8 88cm
3 2.8 888 3 8.8 83cm 8 8.8 88cm
3 2.8 888 I 8.8 888 3 8.8 83.8.
3 8.8 83.8 S 8.8 888. _.~ 8.8 888
mm 8.8 83.8. E 8.8 888 _.~ 8.8 8825
3 8.8 83.8 3 8. 8 83.8. E 8.8 888
2 8.8 23.8— 3 2.: 835m 3 8.8 882
.8 8.8 888 S 8. E 882 3 8.8 83.8
”a 8.8 888m 3 8.8 83.8. E 2.8 88cm
2” 8.8 83.8 3 8.: 83.8. S 8.8 88cm
3 8.8 83.8 2 8.: 83.8. 3 8. 8 :38
8.. 8.8 :38. _.~ 8.8 888 2 8.8 888
M: 8.8 83.8 S 8.8 883 8.. 8.8 888
2 8.8 835m 2 8.8 23.8 3 8.8 888.
2 8.8 89.8 E 8.8 888 E 8.8 888
<> ._ o>< ham _ <> ._ o>< ram _ <> 4 o>< Eam

 

.38 E 82958 :85 3:53
we 5:385 883 3 ES 88 8.8 883 888 05 .3 vegan—«>0 32 22:1??? 32 co 2.: mo :8 Eu mos—9» 2305.5 .vd 038,—.

128

oo§3a< _§m_>u<> Ad 029 8.8853 owfio>< n4 O><

 

 

88 8.8 8.88 8.8 8.8 3.3.3

8 8.8 83.8 .... 8.8 83.8.
.8 8.8 83.8. 8.. 8.8 888. m8 2.8 835m
.8 8.8 83.8 8 8.8 v8.8. .8 8.8 .825
8.8 8.8 83.8. 8 8.8 888 .8 ....8 888.
.8 8.8 83.8 8.8 8.8 888. .8 8.8 888.
Q». 8.8 888. ..m 8.8 888. .8 8.8 8.3.8
8.8 8.8 83.8. .8 8.8 888. n. 8.8 388.
.8 8.8 33.8 8.. 8.8 83.8 ... 8.8 888.
8 8.8 8...... o8 8.8 888. .8 2.8 83.8
o. 8.8 83.8. 88 5.8 :88 8.. 8.8 83.8.
<> 4 0>< Em. A <> 1. o>< ham _ <> 1. o>< Pam

 

28:8. ...m 288

9
2
1

Literature Cited

Bushey, S.M., G.L. Hosfield, and CW. Beninger. 2000. Water uptake and its relationship to
pigment leaching in black beans (Phaseolus vulgaris L.). Annu. Rep. Bean Improv.
Coop. 43:104-105.

Bushey, S.M., L.S. Harris, and G.L. Hosfield. 2004. Development of an early generation test
for predicting color loss of black beans. Annu. Rep. Bean Improv. Coop. 47:139-140.

Bushey, S.M., and G.L. Hosfield. 2007. A test to predict color loss in black bean during
thermal processing. Annu. Rep. Bean Improv. Coop. 50:41-42.

van der Merwe, D., G. Osthoff, and A]. Pretorius. 2006. Comparison of the canning quality
of small white beans (Phaseolus vulgaris L.) canned in tomato sauce by a small-scale
and an industrial method. J. Sci. Food Agric. 86:1046-1056.

Posa-Macalincag, M.T., G.L. Hosfield, K.F. Grafton, M.A. Uebersax, and J .D. Kelly. 2002.
Quantitative trait loci (QTL) analysis of canning quality traits in kidney bean
(Phaseolus vulgaris L.). J. Amer. Soc. Hort. Sci. 127:608-615.

Salinas-Moreno, Y., L. Rojas-Herrera, E. Sosa-Montes, and P. Perez-Herrera. 2005.
Anthocyanin composition in black bean (Phaseolus vulgaris L.) varieties grown in
Mexico. Agrociencia 39:385-394.

SAS Institute, Inc. 2000. SAS version 8. SAS Institute Inc., Cary, NC.

130

Appendix C: Supplemental Data Collected from the ‘Jaguar’ by 115M RIL population from 2004-
2007.
Table Q]. 2004 Agronomic and canning data for the ‘Jaguar’ by 115M RIL population.

 

 

Line Entry VAT YLD SDWT FLWR MTR LDG HT DS
kg/ha g days days cm
804404 46 2.6 3013.1 20.2 47.0 94.5 1.0 48.1 4.5
B04445 87 2.4 2990.6 20.1 47 .0 96.0 1.0 49.0 5.0
804407 49 2.5 2968.1 20.4 47.0 92.0 1.0 47.6 6.0
804422 64 2.4 2956.9 21.9 46.0 94.5 1.0 48.0 5.0
804431 73 2.7 2866.9 22.2 48.0 96.5 1.0 50.1 5.0
804384 26 2.1 2833.2 20.5 46.0 95.0 1.0 47.1 5.0
B04444 86 3.1 2833.2 17.9 47.0 94.0 1.0 48.5 5.5
804429 71 2.5 2822.0 21.9 46.0 93.5 1.0 48.5 5.0
B04374 16 2.5 2788.2 21.0 46.0 94.0 1.5 47.1 5.5
T-39 13 3.2 2777.0 19.9 47.0 93.0 2.0 47.0 5.0
1 15M 100 2.4 2754.5 22.2 47 .0 94.0 1.5 48.5 4.5
804394 36 3.4 2743.3 21.7 48.0 95.5 1.0 48.5 5.0
804383 25 2.7 2743.3 19.6 46.0 94.0 1.0 46.6 5.0
804403 45 2.3 2720.8 18.5 46.0 94.0 1.0 47.5 5.0
804443 85 2.5 2653.3 17.7 47.0 93.9 1.0 47.0 5.0
804452 94 3.5 2653.3 19.5 48.0 94.0 1.0 48.5 6.0
804439 81 3.5 2653.3 22.3 45.0 92.0 1.0 47.9 5.0
804433 75 3.4 2653.3 18.8 47.0 92.0 1.0 47.1 5.5
804387 29 2.9 2653.3 20.7 45.0 96.0 1.0 48.6 4.5
B04361 3 2.7 2642.1 19.0 48.0 94.5 1.0 47.6 5.0
804362 99 3.0 2630.8 19.0 45.0 92.5 1.0 46.0 5.5
804411 53 2.1 2630.8 20.6 47.0 95.0 1.0 48.5 6.0
804453 95 2.5 2619.6 21.8 46.0 93.5 1.0 47.0 5.5
804360 2 3.0 2585.9 19.8 49.0 95.5 1.0 48.6 4.5
804367 9 3.0 2563.4 20.9 47.0 94.0 1.0 47.0 5.0
B04370 12 2.5 2552.1 19.9 47 .O 95.0 1.0 49.1 5.5
804441 83 3.1 2540.9 20.7 47 .0 94.0 1.0 47 .0 5.5
804406 48 3 .5 2540.9 19.0 45.0 93.5 1.0 47.0 5.0
B04451 93 3 .0 2507.2 21.0 46.0 95.0 1.0 47 .0 5.0
804414 56 2.9 2495.9 22.3 47 .0 93.9 1.0 48.0 5.0
804412 54 2484.7 19.5 46.0 92.0 1.0 46.4 6.0
804385 27 2462.2 21.7 48.0 95.5 1.0 48.6 4.5
804399 41 2451.0 19.9 46.0 92.0 1.0 47.0 6.0
804418 60 2451.0 21.4 46.0 94.4 1.0 47.5 5.0
804376 97 2451.0 19.5 48.0 93.0 1.0 48.0 6.0
804420 62 2417.2 20.6 46.0 92.5 1.0 46.5 5.0
804410 52 2372.3 20.2 48.0 96.5 1.0 50.0 5.0

304435 77 2372.3 19.0 47.0 95.5 1.0 48.4 5.5

131

Table C.1 (cont’d.)

804391
804382
804409
804398
804369
804436
804359
804397
804365
804402
804373
804421
804450
804413
804363
804449
804400
804423
804447
804372
804405
Jaguar

804396
804368
804366
804454
804428
804393
804430
804446
804438
804415
804440
804378
804389
8043 86
804401
804408
804381
Tacana

804364
804419

33
24
51
40
ll
78
l
39
7
44
15
63
92
55
5
91
42
65
89
14
47
4
38
10
8
96
7O
35
72
88
80
57
82
20
31
28
43
50
23
18
6
61

4.4

2372.3
2372.3
2349.8
2349.8
2338.5
2327.3
2304.8
2304.8
2304.8
2293.6
2282.3
2271.1
2259.8
2259.8
2259.8
2248.6
2248.6
2226.1
2226.1
2214.9
2214.9
2169.9
2169.9
2158.6
2124.9
2124.9
21 13 .7
2057.4
2023.7
2012.5
1990.0
1990.0
1978.7
1967.5
1956.3
1956.3
1933.8
1933.8
1922.5
191 1.3
1900.0
1877.6

24.0
20.2
19.8
21.3
20.1
22.1
20.9
20.4
20.5
18.4
17.4
20.6
19.8
20.4
19.7
19.4
19.5
19.4
21 . 1
21.2
19.4
19.5
19.1
19.3
20.2
20.6
19.8
19.1
22.3
19.2
21.7
21.3
20.0
20.6
18.9
19.3
20.4
18.1
20.7
22.5
20.5
20.6

132

47.0
47.0
45.0
46.0
46.0
46.0
47.0
46.0
46.0
47.0
47.0
46.0
45.0
47.0
47.0
46.0
47.0
49.0
47.0
48.0
47.0
47.0
46.0
48.0
47.0
45.0
48.0
48.0
45.0
47.0
47.0
47.0
45.0
45.0
45.0
46.0
46.0
46.0
46.0
47.0
46.0
47.0

95.5
93.5
95.5
95.0
92.0
93.0
94.0
92.5
93.5
94.0
94.5
93.0
91.5
94.5
94.5
94.5
94.0
95.0
94.4
95.5
91.5
92.0
94.0
93.5
92.5
91.0
96.0
96.4
93.1
94.5
93.5
92.0
92.5
91.5
92.5
91.5
91.5
94.5
91.5
94.1
95.0
91.5

1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0

49.0
46.5
47.5
48.1
47.4
47.0
47.4
46.5
47.4
46.5
47.0
46.5
44.0
45.6
46.0
47.5
47.5
48.5
47.9
48.1
45.5
45.0
45.6
48.4
47.0
46.4
48.5
48.9
45.1
47.6
45.5
46.0
46.5
45.0
44.5
44.0
45.0
45.5
45.9
46.0
46.5
46.0

4.5
5.5
4.5
6.0
5.5
5.0
5.0
5.5
5.0
5.0
5.0
5.0
5.5
5.0
5.0
5.0
4.5
5.0
6.0
5.0
5.0
5.0
4.5
5.5
5.5
6.0
5.0
4.0
5.0
5.0
5.0
5.5
5.0
5.5
5.0
5.5
5.0
4.5
5.0
4.5
4.0
5.5

 

Table C.1 (cont’d.)

B04377 19 1866.3 19.4 47.0 92.5 1.0 45.0 5.0
B04432 74 1821.3 19.7 46.0 92.5 1.0 47.0 5.0
B04390 32 1810.1 21.6 48.0 94.0 1.0 46.1 4.5
B04375 17 1787.6 20.0 47.0 92.5 1.0 43.0 4.5
B04448 90 1776.4 21.6 45.0 95.5 1.0 47.9 5.0
B04426 68 1753.9 18.8 46.0 93.0 1.0 44.5 5.0
804371 98 1753.9 20.5 45.0 92.5 1.0 45.5 4.5
B04424 66 1742.6 20.7 45 .0 92.5 1.0 47.0 5 .0
B04379 21 1708.9 19.3 45.0 94.5 1.0 45.9 5.0
B04416 58 1686.4 21.6 45.0 91.0 1.0 43.0 4.0
B04434 76 1630.2 23 .4 49.0 96.0 1.5 49.0 4.0
B04417 59 1607.7 19.0 47.0 94.0 1.0 45.5 5.0
B04388 30 1551.5 19.0 47.0 93.0 1.0 46.4 5.0
B04437 79 1529.0 19.5 46.0 92.4 1.0 44.4 4.5
B04395 37 1472.8 22.1 47.0 91.0 1.0 44.5 6.0
B04427 69 1450.3 24.5 47.0 93.5 1.0 45.9 5.0
B04392 34 1439.1 20.6 45.0 93.5 1.0 45.9 5.0
BO43 80 22 1270.4 22.9 47.0 92.4 1.0 45.4 5.0
804425 67 1214.2 21.7 47.0 99.0 1.0 49.4 3.0
B04442 84 1214.2 24.0 46.0 94.0 1.0 46.9 5.0
MEANS 2237.3 20.4 46.5 93 .7 1.0 46.9 5.1
LSD (p=.05) 607.1 1.7 0.0 0.9 0.1 1.3 0.5
LSD (p=.01) 787.0 2.2 0.0 1.2 0.2 1.7 0.6

VA=Visua1 Appearance, YLD=Yield, SDWT=100-seed Weight, FLWR=Days to Flowering,
MTR=Maturity, LDG=Lodging Score, HT=Plant Height, DS=Agronomic Desirability
'1‘ Only a subset of the population was evaluated for visual appearance in 2004.

133

Table 02. 2005 Agronomic and canning data for the ‘Jaguar’ by 115M RIL population.

 

Line Entry VA YLD SDWT FLWR MT R LDG HT DS
kg/ha g days days cm
804366 8 2.3 4171.1 22.9 42.0 95.0 2.0 48.0 4.5
804412 54 2.7 4159.9 23.3 42.0 96.4 1.0 48.0 5.5
804444 86 2.5 4137.4 20.6 45.0 95.6 1.5 48.5 4.5
804404 46 2.4 4137 .4 23.6 43.0 96.0 2.5 47.5 3.5
804449 91 2.5 4092.4 20.9 45.0 95.0 2.0 47.5 5 .0
804431 73 2.3 4081.2 23.5 45.0 96.0 2.0 48.5 5.0
804410 52 1.9 4047.4 22.4 44.0 96.9 2.0 48.0 5.0
804386 28 2.4 4013.7 22.1 43.0 96.0 1.0 47.5 6.0
804384 26 1.7 3991.2 21.7 42.0 96.0 2.0 48.0 5.5
804365 7 1.3 3923.8 21.8 43.0 97.0 2.0 49.0 4.0
804445 87 2.4 3912.5 22.9 45.0 96.6 2.0 48.5 5.0
804433 75 1.9 3912.5 21.5 43.0 96.0 1.5 49.5 5.5
804394 36 2.3 3912.5 24.6 46.0 97.0 3.0 42.5 3.0
804411 53 1.6 3901.3 23.4 44.0 95.6 2.0 47.5 4.5
804396 38 2.1 3878.8 21.0 42.0 97.5 2.0 47.0 4.0
804391 33 1.5 3867.6 25.1 45.0 98.1 2.0 48.0 3.0
804423 65 2.5 3856.3 20.7 46.0 96.5 2.0 50.0 4.5
804429 71 1.9 3856.3 24.5 43.0 95.5 1.5 49.0 4.0
804446 88 2.2 3856.3 20.3 45.0 95.9 2.0 48.0 5.0
804360 2 1.1 3822.6 21.5 45.0 97.0 2.0 48.5 4.0
804363 5 2.1 3822.6 22.5 42.0 95.1 2.0 48.5 5.0
804451 93 2.2 3788.9 24.6 45.0 95.9 2.0 46.5 5.0
115M 100 1.7 3777.6 23.6 44.0 96.0 2.5 46.0 3.5
804376 97 1.4 3755.1 23.2 43.0 96.1 1.0 49.0 5.0
804443 85 1.6 3743 .9 20.2 43.0 96.0 1.5 48.0 5.0
804387 29 1.5 3743.9 24.2 43.0 95.9 2.5 44.5 4.0
804435 77 1.6 3732.6 20.6 45.0 95.9 2.0 48.5 4.0
804369 11 2.5 3732.6 23.7 42.0 95.0 2.0 47.5 5.5
804374 16 2.1 3721.4 24.4 43.0 96.9 3.0 46.0 4.0
804368 10 2.2 3721.4 21.1 46.0 95.5 2.0 47.0 4.0
804414 56 2.2 3710.2 21.8 43.0 96.0 1.0 49.0 4.0
804453 95 1.9 3687.7 23.8 44.0 96.1 2.0 48.5 4.5
804413 55 2.0 3687.7 24.3 44.0 97.0 1.0 48.5 5.0
804452 94 2.6 3687.7 20.9 45.0 96.0 2.0 49.0 4.5
804405 47 2.9 3676.4 21.8 42.0 95.0 1.0 46.9 6.0
804417 59 2.6 3676.4 22.0 43.0 95.0 1.5 46.5 5.0
804454 96 3.1 3676.4 24.1 42.0 95.5 1.5 47 .0 5.0
Jaguar 4 3.8 3665.2 21.5 43.0 95.5 1.5 48.5 6.0
804447 89 3.0 3642.7 23.5 44.0 96.4 2.0 46.5 4.5
804398 40 2.2 3631.5 23.4 44.0 96.0 2.0 47 .5 4.5

134

Table C.2 (cont’d.)

804409
804397
804370
804407
8043 83
804388
804372
804403
804400
804364
804420
804428
804450
804441
804373
804402
804385
804440
804419
804418
804379
804378
804439
Tacana
804432
804406
804361
804421
804362
8043 82
804367
804422
804427
804380
804377
804430
T-39

804359
804438
804434
804424
804416

51
39
12
49
25
30
14
45
42
6
62
70
92
83
15
44
27
82
61
60
21
20
81
18
74
48
3
63
99
24
9
64
69
22
19
72
13
1
8O
76
66
58

1.7
1.9
1.8
1.8
2.1
2.0
2.1
1.5
2.5
1.3
2.0
1.4
2.4
3.6
2.3
2.3
1.7
3.6
3.0
3.1
2.4
2.0
2.4
1.8
2.3
2.4
2.7
2.0
2.4
2.4
2.4
1.5
2.3
1.7
2.6
2.5
3.7
2.1
2.5
2.7
2.2
2.3

3631.5
3631.5
3631.5
3609.0
3597.7
3564.0
3564.0
3564.0
3541.5
3507.8
3507.8
3496.5
3496.5
3485.3
3485.3
3485.3
3485.3
3474.1
3474.1
3474.1
3462.8
3451.6
3451.6
3451.6
3384.1
3384.1
3361.6
3361.6
3350.4
3327.9
3316.7
3305.4
3282.9
3282.9
3271.7
3238.0
3226.7
3204.2
3181.7
3170.5
3159.3
3159.3

21.9
22.7
21.8
23.5
23.1
21.9
22.0
20.7
22.1
23.8
23.6
23.1
23.2
22.6
22.7
21.4
22.8
21.9
26.3
22.7
21.6
22.4
24.3
22.9
22.6
21.2
20.9
22.1
22.0
22.8
22.8
24.1
24.4
24.2
22.2
23.3
24.3
22.8
23.3
23.8
21.8
24.2

135

45.0
43.0
45.0
43.0
43.0
43.0
46.0
45.0
46.0
45.0
43.0
44.0
43.0
45.0
45.0
46.0
45.0
43.0
41.0
45.0
42.0
42.0
43.0
42.0
43 .0
43.0
44.0
43.0
42.0
45.0
45.0
45.0
46.0
43.0
44.0
43.0
43.0
43.0
43.0
45.0
44.0
44.0

96.1
96.0
96.1
95.0
96.5
95.5
96.0
95.5
97.5
96.0
95.5
96.0
95.5
96.0
96.5
97.5
95.5
95 .0
95.0
95 .5
95.5
96.5
95.5
95.5
95.9
95.5
95.4
95.5
96.0
96.5
96.5
96.5
95.5
95.0
95.5
95.6
95 .5
95.0
96.5
98.0
95.6
95.0

2.0
2.0
2.0
2.0
2.0
1.5
2.0
2.0
2.0
2.0
1.0
1.5
2.0
2.0
1.5
1.5
2.0
1.5
1.5
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
1.5
1.0
1.0
2.0
4.0
2.0
1.5
2.5
2.0
1.0

48.5
49.0
48.0
46.5
46.5
49.0
47.5
49.0
49.0
48.0
49.0
50.0
46.5
48.0
49.0
48.0
47.5
47.0
47.5
47.0
48.5
47.5
47.5
48.0
47.0
50.0
48.5
48.5
43.5
49.5
48.0
47.5
48.5
48.0
48.0
47.5
35.0
48.0
47.0
47.0
47.0
49.0

4.5
4.5
4.5
5.5
5.0
5.0
5.0
5.0
4.0
4.5
6.0
4.0
6.0
5.5
5.0
4.0
5.0
6.5
5.0
4.5
4.5
4.5
4.0
4.0
5.5
5.0
5.0
5.0
4.0
4.5
4.5
5.0
5.0
5.0
6.0
5.0
3.0
4.0
4.5
4.0
4.5
4.5

Table C.2 (cont’d.)

B04399 41 1.7 3148.0 22.2 42.0 98.0 1.5 46.5 4.5
B04401 43 1.8 3136.8 22.7 42.0 96.5 1.0 46.5 6.0
B04426 68 3.0 31 14.3 20.6 45.0 95.5 1.0 48.0 4.5
B04437 79 2.0 3114.3 22.3 43.0 95.5 2.0 47.5 4.5
B04371 98 2.5 3103.0 23.4 44.0 95.4 2.0 48.5 5.0
B04395 37 3 .0 3103.0 26.3 42.0 95.6 1.0 46.0 5.5
804436 78 2.7 3046.8 23.2 44.0 95.0 1.5 48.5 4.5
BO4415 57 2.1 3046.8 25.2 43.0 95.5 1.5 47.0 4.5
B04448 90 2.0 3024.3 22.1 42.0 95.5 2.0 48.0 4.5
804393 35 2.0 3013.1 21.8 43.0 97.0 2.0 49.0 4.0
B04389 31 1.7 3013.1 22.9 42.0 96.5 1.0 45.9 5.5
B04390 32 1.9 2979.4 23.2 42.0 95.5 1.5 47.5 5.0
B04375 17 1.3 2979.4 22.1 42.0 95.5 1.0 47.0 5.5
B04408 50 3.1 291 1.9 20.7 43.0 96.5 2.0 46.0 5.0
B04425 67 3.5 2855.7 23.1 45.0 97.5 2.0 48.5 4.0
B04392 34 1.9 2810.7 24.5 42.0 95.5 1.0 47.5 5.0
B04381 23 2.2 2743.3 24.0 43 .0 95.0 1.5 48.0 4.0
B04442 84 2.0 2473.4 24.8 43.0 95.0 2.0 46.0 4.5
MEANS 31.2 3507.8 22.8 43.6 95.9 1.8 47.6 4.7
LSD (p=.05) 314.8 1.2 0.0 0.7 0.4 1.2 0.6
LSD (p=.01) 404.7 1.5 0.0 0.9 0.5 1.6 0.8

VA=Visua1 Appearance, YLD=Yield, SDWT=100-seed Weight, FLWR=Days to Flowering,
MTR=Maturity, LDG=Lodging Score, HT=Plant Height, DS=Agronomic Desirability

136

Table Q3. 2006 Agronomic and canning data for the ‘Jaguar’ by 115M RIL population.

 

 

Line Entry VA YLD SDWT FLWR MTR LDG HT DS
kg/ha g days days cm
804391 33 3.5 4261.1 18.9 44.9 98.5 2.6 47.9 2.0
804431 73 2.9 4013.7 21.2 46.0 95.1 2.0 49.9 4.0
804411 53 2.9 3957.5 19.8 46.1 95.8 3.0 48.6 3.0
804398 40 2.5 3912.5 18.8 46.4 94.8 2.1 48.4 4.0
804370 12 3.1 3912.5 18.6 45.0 97.2 3.5 48.4 3.0
804444 86 2.9 3867.6 16.7 45.9 93.2 2.0 49.4 4.5
804445 87 3.6 3856.3 20.7 45.0 96.5 2.6 49.4 3.5
804404 46 3.7 3856.3 19.0 46.4 92.0 2.5 49.1 4.0
804446 88 3.4 3856.3 16.3 46.0 96.1 3.0 47.5 2.5
804429 71 3.4 3822.6 18.8 44.0 92.2 2.0 49.0 4.0
804443 85 2.6 3822.6 17.1 44.0 94.2 2.0 50.1 4.0
804423 65 3.4 3822.6 18.0 48.0 94.3 3.5 48.9 3.0
804414 56 3.4 3800.1 18.1 44.0 95.9 2.5 49.1 3.5
804400 42 3.3 3766.4 18.1 44.4 94.7 2.1 48.9 4.0
804386 28 3.3 3755.1 20.1 43.0 91.8 1.4 50.1 5.0
804397 39 2.7 3732.6 19.8 44.4 95.9 2.0 48.1 4.0
804412 54 3.6 3721.4 19.2 43.5 92.6 1.5 50.6 4.5
804450 92 3.7 3721.4 19.4 44.0 93.3 1.6 48.0 4.5
804451 93 3 .4 3721.4 20.3 44.4 93.5 2.0 48.0 4.0
804376 97 3.3 3710.2 18.7 44.5 94.3 2.0 48.5 4.0
804410 52 3.2 3710.2 19.4 45.0 97.0 3.0 48.5 3.0
804435 77 2.8 3698.9 17.3 45.0 94.7 2.5 48.0 3.5
804440 82 3.2 3698.9 17.7 44.0 92.2 1.5 48.5 4.5
804364 6 2.4 3687.7 20.3 45.5 93.5 1.5 52.5 4.0
1 15M 100 2.9 3665.2 19.8 45.0 94.2 2.0 49.0 3.5
804396 38 2.7 3665.2 19.0 43.0 96.0 3.0 48.5 3.0
804387 29 2.9 3653.9 19.3 44.5 91.9 2.4 48.1 3.5
804405 47 3.7 3642.7 19.2 43.5 93.3 1.5 50.0 5.5
804437 79 3.5 3642.7 19.6 44.1 91.1 2.5 48.5 4.0
804409 51 3.0 3631.5 19.0 47.4 95.2 3.0 49.0 3.0
804402 44 2.7 3620.2 16.0 45.5 96.0 2.0 49.5 4.0
804428 70 2.4 3575.2 20.1 44.0 95.9 2.0 50.5 3.0
804413 55 3.0 3575.2 18.6 45.5 93.8 2.5 47.9 3.0
804433 75 3.6 3564.0 17.5 45.0 94.9 2.0 49.9 4.0
804382 24 3.6 3564.0 19.2 44.6 95.1 2.0 50.0 4.0
804371 98 3.4 3564.0 19.0 44.0 93.3 2.0 48.9 4.5
804420 62 2.7 3541.5 18.8 45.0 93.7 2.0 48.4 4.0
804368 10 3.6 3530.3 18.1 46.1 93.4 2.6 48.4 3.0
804360 2 3.1 3530.3 18.8 48.1 95.9 3.0 46.9 3.0
804365 7 2.2 3519.0 19.2 44.7 93.2 2.0 49.0 4.0

137

Table C.3 (cont’d.)

804366
804447
804372
804361
804406
804384
804454
804416
804439
804421
804388
804385
804369
80437 8
804367
804362
804453
804418
804374
804417
804394
804401
804422
804419
8043 83
804436
804426
804395
804389
804441
Jaguar
804425
804379
804434
Tacana
804432
80437 7
804438
804393
T-39
804392
804449

8
89
14

3
48
26
96
58
81
63
30
27
ll
20
9
99
95
60
16
59
36
43
64
61
25
78
68
37
31
83
4
67
21
76
18
74
19
80
35
13
34
91

2.7
3.4
2.7
3.2
3.1
2.6
2.7
2.9
3.8
3.9
3.2
2.8
3.4
3.7
3.1
3.2
3.7
3.3
3.0
2.9
3.4
2.8
2.6
3.5
2.8
3.1
2.8
3.5
2.9
3.7
3.9
3.7
2.6
3.5
3.1
3.4
3.2
2.6
3.0
3.7
2.4
3.4

3507.8
3507.8
3507.8
3507.8
3507.8
3496.5
3485.3
3485.3
3485.3
3474.1
3462.8
3440.3
3429.1
3429.1
3417.8
3417.8
3417.8
3417.8
3417.8
3417.8
3406.6
3395.4
3372.9
3361.6
3361.6
3361.6
3350.4
3339.1
3339.1
3327.9
3316.7
3305.4
3271.7
3271.7
3260.4
3260.4
3260.4
3238.0
3238.0
3238.0
3238.0
3226.7

18.6
19.0
20.5
18.6
18.0
18.0
19.2
19.3
19.2
18.7
17.8
18.8
20.1
19.0
19.1
17.2
20.4
20.9
19.4
18.6
20.0
19.2
20.1
18.6
18.8
19.2
18.6
21.1
18.9
19.1
18.6
19.1
19.4
19.4
19.6
20.4
18.9
19.6
17.5
18.8
20.5
17.9

138

44.1
45.1
45.9
45.5
45.1
44.5
44.0
44.4
44.6
44.4
44.6
48.5
44.1
44.0
44.8
44.4
45.9
45.9
44.6
44.1
46.0
44.1
46.4
44.1
44.5
44.4
44.5
44.0
44.1
45.8
44.0
46.6
44.1
47.0
44.5
44.5
44.4
45.0
45.1
44.4
44.4
45.0

94.3
96.2
95.4
97.1
93.4
94.7
91.5
94.0
92.8
92.0
91.4
93.8
92.4
95.1
95.5
93.2
94.7
95.4
96.6
91.0
95 .6
94.4
95.8
92.0
94.2
95.4
93.8
91.6
94.5
95.4
92.4
95.5
93.0
96.3
94.8
95.5
92.4
93.9
96.4
92.5
94.9
91.7

2.0
2.0
2.0
2.5
3.0
1.5
2.5
1.0
3.0
2.0
2.0
3.4
1.9
2.1
2.0
2.6
2.1
3.0
3.5
2.0
3.0
2.0
2.0
1.9
1.4
1.5
1.0
1.1
2.0
2.0
1.5
1.9
2.0
3.0
1.9
2.4
1.5
1.0
2.6
4.1
2.0
2.0

49.0
47.9
49.1
48.0
48.5
49.5
48.1
51.0
46.0
48.5
50.5
47.6
49.1
48.5
49.0
49.0
48.9
48.1
47.5
49.0
48.5
49.1
49.4
48.5
49.2
49.0
50.0
48.9
48.5
48.9
48.5
49.6
48.9
49.0
48.6
49.0
49.5
49.6
47.0
46.0
49.0
48.0

3.5
3.5
3.5
2.5
3.5
4.5
3.5
5.0
2.5
4.0
5.0
3.0
4.5
4.0
3.5
3.5
4.0
4.0
2.5
3.5
3.0
4.0
3.5
4.0
5.5
4.0
4.0
5.5
4.0
3.5
5.0
3.0
3.0
3.0
4.5
3.5
4.5
4.0
3.0
1.5
4.0
3.5

Table C.3 (cont’d.)

5.5
3.5
4.5
3.5
3.5
2.5
3.0
3.0
5.5
4.0
3.5
3.0
3.5
4.0
4.0
2.0
4.0
5.0
3.7

304375 17 2.2 3215.5 18.7 44.1 94.4 0.9 50.6
304390 32 3 .0 3193.0 19.4 45.5 95.2 2.0 48.9
304415 57 2.3 3193.0 20.1 44.1 91.8 0.9 48.1
B04452 94 3 .5 3170.5 18.6 47.4 95.3 3.0 48.4
304399 41 2.6 3159.3 18.8 43.9 95.9 1.5 49.0
B04373 15 2.9 3159.3 20.0 46.6 97.5 2.0 50.1
304448 90 3 .1 3148.0 20.5 45.5 95.9 2.9 47.6
304363 5 3.2 3136.8 17.6 43.4 95.4 3.5 49.0
B04380 22 2.7 3125.5 20.4 45.6 91.6 1.0 50.0
B04407 49 3.3 3125.5 19.1 44.4 92.7 2.0 49.5
B04430 72 3.3 3114.3 20.7 45.0 93.6 1.5 50.0
B04403 45 3.4 3103.0 17.8 47.6 94.5 2.5 47.0
304424 66 2.4 3091.8 18.8 44.0 94.8 2.0 49.5
304408 50 3.1 3058.1 16.2 44.0 94.9 2.0 47.5
304381 23 3.2 3001.9 18.9 45.0 91.5 1.1 47.5
304359 1 3.0 2990.6 18.0 43.9 92.2 4.0 47.5
304427 69 3.1 2990.6 19.8 45.1 94.6 1.5 48.9
304442 84 2.7 2900.7 20.4 45.2 94.1 1.0 48.5
MEANS 3474.1 19.0 45.0 94.2 2.2 48.8
LSD (p=.05) 303.6 1.4 1.1 1.4 0.5 1.1 0.7

LSD (p=.01) 393.5 1.8 1.4 1.9 0.7 1.4 0.9

VA=Visua1 Appearance, YLD=Yield, SDWT=100-seed Weight, FLWR=Days to Flowering,
MTR=Maturity, LDG=Lodging Score, HT=Plant Height, DS=Agronomic Desirability

139

Table CA. 2007 Agronomic and canning data for the ‘Jaguar’ by 115M RIL population.

 

 

Line Entry VA YLD SDWT FLWR MTR LDG HT DS
kg/ha g days days cm
804391 33 1.2 3710.2 22.2 55.5 103.1 2.1 54.0 3.0
804431 73 1.7 3631.5 22.3 54.9 101.6 1.5 54.0 3.5
804384 26 1.8 3575.2 18.8 54.5 101.5 1.1 53.5 5.0
804423 65 2.2 3474.1 18.9 57.4 102.6 2.0 54.5 3.1
804411 53 1.6 3440.3 21.1 56.1 101.9 1.9 53.5 4.0
804404 46 2.0 3395.4 19.2 54.5 101.5 2.6 53.5 3.0
804429 71 1.9 3350.4 19.3 54.5 101.5 2.0 52.5 3.5
804445 87 2.1 3350.4 20.6 54.1 101.4 1.5 53.5 3.5
804366 8 1.9 3339.1 18.7 54.1 101.5 1.5 54.0 3.5
804394 36 2.1 3339.1 21.3 55.5 100.6 2.6 53.0 3.0
804443 85 2.2 3327.9 18.7 55.1 101.4 1.5 53.5 4.0
804444 86 2.2 3316.7 19.2 55.0 101.5 2.0 54.0 3.0
804360 2 2.2 3305.4 20.4 55.4 102.1 2.5 53.5 3.5
804400 42 2.1 3305.4 19.5 55.0 100.6 2.1 53.5 3.5
804385 27 1.8 3294.2 20.7 55.0 101.9 2.1 53.5 3.5
804413 55 2.1 3294.2 19.7 54.5 101.1 1.1 53.1 4.4
804414 56 1.9 3282.9 20.3 54.6 101.9 0.9 52.5 4.5
804430 72 2.8 3282.9 19.9 54.1 101.0 1.7 52.5 3.5
804440 82 2.5 3282.9 19.0 53.5 100.1 1.1 52.5 4.5
804452 94 3.1 3271.7 19.2 55.9 101.5 1.9 54.0 3.1
804405 47 1.9 3260.4 18.8 54.5 99.9 0.9 51.0 5.0
804421 63 2.8 3249.2 17.8 53.9 101.4 1.9 53.4 4.0
804439 81 3.1 3238.0 19.8 54.6 101.4 1.5 52.5 3.5
804446 88 2.4 3238.0 19.3 54.9 100.8 1.9 53.9 3.5
804387 29 2.0 3215.5 19.7 52.5 100.8 2.6 52.5 3.0
804451 93 2.4 3215.5 20.4 53 .4 99.9 1.4 51.5 4.0
804412 54 2.3 3204.2 18.7 54.9 100.1 1.0 52.0 5.0
115M 100 2.3 3204.2 20.8 54.0 102.0 1.6 53.5 4.0
804386 28 2.6 3193.0 19.2 52.9 100.5 1.0 51.5 5.0
804372 14 2.2 3181.7 18.8 56.1 101.9 1.6 53.0 4.0
804364 6 3.0 3170.5 19.3 53.0 100.6 1.1 51.6 3.9
804453 95 1.6 3170.5 22.2 55.5 101.5 2.0 53.5 3.5
804377 19 1.8 3159.3 18.8 53.6 100.5 0.9 53.0 5.0
804383 25 2.6 3159.3 18.6 52.6 100.9 1.1 52.5 4.0
804382 24 2.7 3148.0 19.5 55.4 101.1 1.6 53.6 3.9
804426 68 2.4 3148.0 19.5 54.4 100.6 1.1 51.1 5.0
804454 96 2.3 3136.8 20.4 54.5 101.0 1.0 52.0 4.0
804361 3 1.9 3125.5 18.7 54.5 101.1 1.5 53.0 4.4
804373 15 1.8 3103.0 20.4 55.1 103.0 2.0 52.6 3.9
804370 12 2.0 3091.8 19.4 55.0 100.9 2.0 53.0 3.6

140

Table C.4 (cont’d.)

804447
804390
804396
804401
804441
804376
804395
804407
804415
804422
Jaguar
804365
804449
804450
Tacana
804435
804392
804437
804369
8043 80
804418
804363
804427
804398
804410
80437 8
804448
804367
804432
804403
804420
804406
804374
804402
804379
8043 89
804393
804424
804375
804409
804417
804419

89
32
38
43
83
97
37
49
57
64
4
7
91
92
18
77
34
79
1 1
22
60
5
69
40
52
20
90
9
74
45
62
48
16
44
21
31
35
66
17
51
59
61

2.4
2.2
2.4
2.4
2.5
2.8
2.4
1.9
2.5
2.5
3.4
2.8
2.0
2.8
2.3
1.8
2.3
1.4
1.9
1.3
1.9
1.9
2.0
1.3
1.1
1.4
1.7
1.4
1.5
1.4
1.7
1.5
1.6
1.7
1.6
2.0
2.0
1.6
1.3
2.7
2.8
1.9

3080.6
3069.3
3069.3
3069.3
3069.3
3058.1
3058.1
3058.1
3058.1
3058.1
3046.8
3035.6
3035.6
3035.6
3035.6
3024.3
3013.1
3013.1
3001.9
2990.6
2990.6
2979.4
2979.4
2968.1
2968.1
2956.9
2956.9
2945.6
2945.6
2934.4
2934.4
2923.2
291 1.9
2900.7
2889.4
2878.2
2878.2
2878.2
2855.7
2855.7
2855.7
2855 .7

20.4
21.1
18.2
19.1
20.3
17.8
20.2
18.4
20.6
19.7
18.5
18.6
18.6
19.3
19.3
19.3
19.3
18.4
19.2
20.7
20.3
18.9
21.4
20.5
20.0
19.1
19.6
19.2
20.2
17.7
19.1
18.0
20.2
18.0
18.8
18.8
20.4
19.6
19.3
20.8
17.7
20.0

141

54.0
53.4
52.0
52.5
55.0
54.5
54.1
53.6
53.0
54.9
54.0
54.5
53.4
52.0
51.6
55.6
54.0
55.1
52.5
54.5
55.1
55.0
54.6
55.0
55.5
54.5
53.6
54.0
54.9
55.0
55.0
55.0
54.4
54.9
54.5
53.5
55.0
55.1
54.0
55.6
50.9
54.0

102.0
102.0
101.9
100.4
101.0
100.9
100.0
101.0
101.0
101.5
99.8

102.6
101.4
99.6

98.9

102.6
102.0
100.5
101.1
99.5

101.9
100.0
104.1
101.6
101.6
100.5
102.0
102.0
106.0
101.1
100.4
100.5
101.5
103.0
101.6
101.0
103.6
101.5
100.0
102.4
100.5
99.4

1.9
0.9
0.9
1.0
1.0
0.9
1.5
1.6
2.0
1.4
0.9
1.0
2.0
1.1
1.7
1.4
1.0
2.0
2.1
0.9
1.5
1.4
1.0
1.5
2.0
1.9
1.5
0.9
1.5
2.1
0.9
1.5
2.5
1.0
0.9
1.5
2.0
0.9
1.2
1.0
1.0
1.0

54.0
53.5
52.9
51.9
52.5
53.5
51.5
52.5
52.5
53.5
51.5
53.5
52.9
50.5
52.0
54.0
53.5
52.0
52.5
51.5
53.0
52.5
51.5
54.0
54.0
52.5
53.4
52.5
54.0
53.0
53.0
52.5
52.5
52.1
52.5
52.0
53.0
52.0
51.1
54.0
52.0
50.9

3.0
4.6
4.1
4.6
4.5
4.5
4.0
4.0
3.5
4.1
5.0
4.0
3.5
4.4
4.0
3.0
3.0
3.5
4.0
5.0
3.5
4.5
3.0
4.0
4.0
4.0
4.0
4.0
5.0
3.4
4.6
4.5
2.9
3.5
4.5
4.5
3.0
4.0
5.0
3.5
5.0
5.0

Table C.4 (cont’d.)

804436 78 2.1 2833.2 20.3 54.0 101.4 0.9 53.0 4.0
804416 58 1.5 2788.2 19.2 53.1 100.0 0.9 52.0 4.5
804428 70 1.2 2777.0 21.4 54.0 99.9 1.0 53.9 4.1
804371 98 1.4 2765.8 18.9 54.9 101.0 1.5 53.0 4.0
T-39 13 2.5 2754.5 20.1 54.5 100.1 4.1 42.0 2.0
804397 39 1.5 2732.0 18.6 53.5 101.9 0.9 54.0 4.0
804388 30 1.9 2709.5 18.3 55.0 101.0 1.0 53.0 5.0
804359 1 1.7 2698.3 20.1 55.0 102.5 1.4 52.5 4.0
804433 75 1.7 2698.3 17.3 55.4 100.4 0.9 51.9 5.1
804362 99 2.1 2675.8 17.7 53.0 101.0 2.0 53.0 3.5
804408 50 1.5 2675.8 17.3 54.4 102.4 0.9 51.5 4.0
804425 67 2.4 2630.8 23.1 55.9 103.1 2.1 53.6 3.0
804368 10 1.2 2619.6 18.2 55.1 101.6 1.5 53.0 4.0
804438 80 1.1 2585.9 19.3 54.4 102.4 0.9 52.4 3.6
804399 41 1.7 2518.4 18.3 53.1 99.9 0.9 52.0 5.0
804434 76 2.6 2495.9 22.3 55.5 107.1 2.0 54.5 3.0

804381 23 1.9 2462.2 19.5 53.5 100.4 1.0 51.5 4.0
1304442 84 1.7 2406.0 21.4 53.1 102.1 1.0 52.0 4.0
MEANS 3046.8 19.6 54.3 101.3 1.5 52.7 4.0
LSD (p=.05) 343.5 1.1 0.9 1.1 0.4 1.0 0.5
LSD (p=.01) 449.7 1.4 1.2 1.4 0.6 1.2 0.7

 

VA=Visual Appearance, YLD=Yield, SDWT=100-seed Weight, FLWR=Days to Flowering,
MTR=Maturity, LDG=Lodging Score, HT=Plant Height, DS=Agronomic Desirability

142

Table C.5. Three year averages for processed bean color (Hunter L-value), texture (Kg-
force), and washed-drained weight (g) measured in the ‘Jaguar’ by 115M RIL population.

 

 

Line CLR Line TXT Line WDWT
804398 20.0 Tacana 84.1 804447 260.5
804391 18.5 804363 79.0 804441 256.3
804428 18.4 804359 72.9 804374 256.0
804420 18.2 804385 71.6 8043 77 255.7
804396 17.7 115M 71.5 804440 255.5
804415 17.6 804387 70.7 T-39 254.7
804431 17.5 804451 70.5 804369 254.4
804429 17.5 804419 68.9 804395 254.0
804413 17.5 804453 68.7 804417 253.5
804435 17.3 804428 68.5 804361 253.1
804375 17.2 804421 68.3 Jaguar 253.0
804372 17.2 804424 67 .6 804405 252.9
115M 17 .2 804395 67 .6 804414 252.7
804385 17 .2 804408 67.3 804365 252.5
804409 17.1 804362 67.1 804399 252.4
804387 17.1 804404 67 . 1 804438 252.0
804384 17.0 804454 66.8 804379 251.8
804374 17.0 804439 66.6 804400 251.8
804402 16.8 804413 66.5 804450 251.3

304380 16.8 304449 66.1 304419 251.3
304370 16.7 304366 66.0 304434 251.3

304403 16.7 304446 66.0 304415 251.2
304397 16.7 304360 65.7 304444 251.2
304417 16.7 304415 65.4 304402 251.0
304405 16.6 304432 65.3 304449 251.0
304365 16.6 304445 64.9 304384 251.0
304443 16.5 304425 64.5 304409 250.8

304360 16.5 304389 64.3 304382 250.8
304449 16.4 304386 64.2 304412 250.7
304438 16.4 304382 64.0 304378 250.7
304444 16.3 304429 64.0 304439 250.6

Tacana 16.3 304420 64.0 304425 250.6
304364 16.3 304430 63.9 304418 250.6
304416 16.3 304407 63.6 304436 250.5
304400 16.3 304368 62.9 304388 250.5
304410 16.3 304390 62.5 304392 250.3
304382 16.2 3043 80 62.1 304394 250.1
304388 16.2 304396 61.9 304423 250.0
304411 16.2 304376 61.8 304364 249.9

143

Table C.5 (cont’d.)

304453 16.2 30441 1 61.4 304422 249.8
304379 16.1 3043 79 60.9 304443 249.8
304366 16.1 304452 60.5 304442 249.7
304442 16.1 304367 60.3 304372 249.7
304371 16.1 304450 60.2 304371 249.5
3043 76 16. 1 304423 60.1 304404 249.4
304383 16.0 304364 60.0 304396 249.4
304392 16.0 304369 59.5 3043 86 249.4
304437 16.0 304437 59.4 304381 249.3
304445 16.0 304431 58.9 304367 249.3
304395 16.0 304435 58.8 304445 249.2

304406 16.0 304427 58.8 304383 249.2
304368 15 .9 304394 58.6 3043 70 249.1

304404 15.9 304412 58.4 304410 249.1
304423 15.9 304397 57.9 304362 249.0
304418 15.9 304403 57.9 304368 248.9
304419 15.8 304410 57.7 304451 248.7

304414 15.8 304388 57.7 304431 248.6
304390 15.8 304381 57.1 304366 248.5

3044-46 15.8 304448 56.7 304373 248.3
304427 15.8 304391 56.7 304406 248.3
304401 15.8 304426 56.6 304454 248.3
304432 15.7 304405 56.6 304448 248.2
304369 15.6 304392 56.3 304432 248.1
304424 15 .6 304434 56.1 304416 248.1
304407 15.6 304418 56.0 304413 248.1
304447 15 .6 304375 55 .9 304390 247.9
3043 89 15 .5 3043 84 55 .2 304426 247.9
304434 15.5 304401 55 .0 304408 247.7
304451 15 .5 304436 55 .0 304389 247.7
304433 15 .4 3043 70 54.9 304401 247.6
304454 15 .4 3043 78 54.9 304391 247.6
304394 15 .4 304400 54.7 304398 247.2
304367 15 .4 3043 73 54.5 304427 247.1
304362 15 .4 304422 54.3 304393 246.9
304399 15.4 304433 53.6 304452 246.9
304430 15 .3 304409 53.2 304433 246.5
304422 15 .3 304416 52.8 304397 246.3
304377 15.3 304393 52.7 304437 246.2
304393 15.3 304443 52 .5 304420 246.1
304450 15.2 304444 52.5 115M 246.1
3043 86 15 .2 304365 52.3 304421 245.9

144

Table C.5 (cont’d.)

804436 15.2 304441 52.1 Tacana 245.8
304441 15.2 804406 52.0 304453 245.2
304452 14.9 304417 51.9 304375 245.1
Jaguar 14.9 304442 51.8 804360 244.8
304408 14.9 304374 51.6 304407 244.7
304440 14.8 304372 51.5 304424 244.6
304412 14.7 304399 51.5 304430 244.5
304421 14.6 304377 50.7 804380 244.3
T-39 14.6 304371 50.7 804446 244.1
804378 14.6 304440 50.7 304359 244.0
304361 14.5 304383 50.6 304435 243.4
804381 14.5 304402 49.3 304363 243.4
804363 14.5 304414 49.2 804376 243.0
304373 14.4 T-39 48.5 304429 241.9
804448 14.4 Jaguar 48.4 304385 241.8
304425 14.2 804398 48.4 304411 238.4
304359 14.2 804438 46.2 304428 238.2
304439 14.1 804361 45.7 304403 235.2
804426 13.7 304447 45.5 304387 234.9
Mean 16.0 59.4 248.7
LSD (.05) 1.3 9.6 6.8

LSD (.01) 1.7 12.7 8.9

 

CLR=Canned bean color, TXT=Texture, WDWT=Washed-drained weight

Table C.6. Mean values by year for processed bean color, texture, and washed-drained weight measured in
the ‘Jaguar’ by 115M RIL population during 2005-2007.

 

 

YEAR CLR TXT WDWT
2005 17.6 63.9 243.3
2006 14.2 60.8 248.9
2007 16.2 53.5 253.9

LSD(.05) 0.4 2.5 1.5
LSD(.01) 0.5 3.3 2.0

 

CLR=Canned bean color, TXT=Texture, WDWT=Washed-drained weight

145

Table C.7. Reaction of 96 ‘Jaguar’ by 115M RILs following inoculation with race 73 of C.

lindemuthianum in the greenhouse. (S=susceptible, Rqesistant).

 

 

Plants Plants Plants
Accession S R Accession S R Accession S R
804359 7 804404 3 2 804449 5
804360 6 804405 7 804450 7
804361 6 804406 7 804451 2 4
Jaguar 13 804407 8 804452 1 4
804363 9 804408 3 2 804453 6
804364 5 804409 7 804454 9
804365 6 804410 9 804376 5
804366 10 804411 5 2 804371 7
804367 6 804412 10 804362 7
804368 5 804413 11 115-11M 13
804369 11 804414 1 5 Blackhawk 6
804370 7 804415 8
T-39 11 804416 7
804372 5 804417 3 4
804373 6 804418 5
804374 6 804419 7
804375 2 6 804420 6
Tacana 5 804421 3 2
804377 7 804422 7
804378 5 804423 6 1
804379 5 1 804424 4 2
804380 6 804425 7
804381 11 804426 6
804382 6 804427 7
804383 5 804428 6
804384 5 804429 6 1
804385 9 804430 7
804386 7 804431 7
804387 7 804432 6
804388 5 804433 6
804389 2 7 804434 7
804390 6 804435 1 6
804391 1 6 804436 3 3
804392 9 804437 6
804393 6 804438 7
804394 7 804439 7
804395 6 804440 7
804396 6 804441 7
804397 5 2 804442 6
804398 6 804443 7
804399 3 4 804444 5
804400 4 3 804445 5
804401 6 804446 6
804402 4 1 804447 7
804403 10 804448 7

 

146

Table C.8. SRAP primer sequences used in pairwise combinations to screen for genomic
polymorphisms in a ‘J aguar’/ 1 15M RIL population. Sequence information based on Li
and Quiros (2001) TAG 1032455-461.

 

 

Code Forward Primer Code Reverse Primer

M1 TGA GTC CAA ACC GGA TA El GAC TGC GTA CGA ATT AAT
M2 TGA GTC CAA ACC GGA GC E2 GAC TGC GTA CGA ATT TGC
M3 TGA GTC CAA ACC GGA AT E3 GAC TGC GTA CGA ATT GAC
M4 TGA GTC CAA ACC GGA CC E4 GAC TGC GTA CGA ATT TGA
M5 TGA GTC CAA ACC GGA AG ES GAC TGC GTA CGA ATT AAC
M6 TGA GTC CTT TCC GGT AA 86 GAC TGC GTA CGA ATT GCA
M7 TGA GTC CTT TCC GGT CC E7 GAC TGC GTA CGA ATT CAA
T1 TGT GTG GTT AAT ATG AGC E8 GAC TGC GTA CGA ATT CAC

 

Table C.9. TRAP primer sequences used in pairwise combinations to screen for genomic
polymorphisms in a ‘J aguar’/ 1 15M RIL population. Sequence information based on Hu
and Vick (2003) Plant Mol. Bio. Rept. 21 :289-294.

 

 

Code Forward Primer Code Reverse Primer

Fl CAA CCG AAA ACC AGC AAT R1 GCG AGG ATG CTA CTG GTT

F2 CGA TCT AGA ATC CAA GCC R2 CTA TCT CTC GGG ACC AAA C
F3 CGA ATC TCC ACT AAA CCC R3 TTC TAG GTA ATC CAA CAA CA
F4 CCG AGT TGG TAT GCT TGT R4 TTA CCT TGG TCA TAC AAC ATT
F5 ATC AGT TCA TTA GGG CAC R5 TTC TTC TTC CCT GGA CAC AAA
F6 GGA ACA TTT GTC TCT CGC R6 TCA TCT CAA ACC ATC TAC AC
F7 CTT CAG CAG TGT CTC TCC R7 GGA ACC AAA CAC ATG AAG A
F8 CTC GAT AAC ATC CTC CCA R8 CCA AAA CCT AAA ACC AGG A
F9 TGG ATT TTC ACC AGC GTC R9 CAC AAG TCG CTG AGA AGG

F 10 GAA ATT AAC GGG GTT GGA R10 ATA AGA ATC AGC AGA CGC AT
F11 GCT TCA ATT GGC CCT TAC

F12 CAG AAC TTG TTG GTG GTG

F13 CAT CGC ATA CTG ATG GAG

F14 GCA GAC ATC GGT AGA AAG

F15 AGT TGT TCC CAG ATG GAG

F16 GTG GGA ACC TAG AAA TGG

F 17 CCT AAA TGG GAG GAA GTG

F18 AAG ATC ACA CCT TGT CCC

F19 AAT CTC AAG GAC AAA AGG

F20 GCT TCA GAG CAT TGA AGT

F21 GAA AGA CGA AGG AAC AGG

F22 CGT TTA TIT CCT CGC CTC

 

147

 

148

m < M 53425 < < m 3 gm m < M _wmvom
m < M @3425 < I m m _ 34cm M < M owmvcm
M < M mvvvom M < M N— 36m A m I abmvom
m < M 34%ch M m I _ Svom m < I whmvcm
m < I mvvvcm M < M o I 35m < m m Rmvcm
n— < I Nix—5m m < M aogom m < M chmvom
m < M _ 334cm m < M wovvom < m m mbmvom
m < M 93425 I < M hogm . m M 3.33—
m < I ovacm n— < M. ocvvom m n— I mumvom
M < M wmvvcm m < M 334cm M < M thvcm
m < M :34ch m < M 3340m— m < M Ibmvom
n— < M 0325 A < M 834cm m < I onmvom
< M m m mvvom m a— I mogom < M I acmvom
I < M .494ch m < M 335m m < M womvom
< n— m mmvgm < m m ocvvom m < M hemgm
n— < M 3?ch m < M gmvom I < m ocmvom
n— < M 5343— m m M womvom m < m 894cm
< L m cmvvom m < M hamvcm a— < M 3215
< n— I amvvom m < M wavom n— < M mcmvcm
m < M wmvgm < < I nonvom < m M mom—4am
m < M hmvvcm < m I 395m A < M .325
n— < M cmgm m < M mamvcm m < M oemvom
< n— m mmvvcm m < M Namvom M < I ommvcm
_MPM o _ Mm _ m HmDM =o_mmooo< _MFM c _Mm _ m HmDM commmooo< MIL o _Mm _ M HmDM =o_mmooo<

 

.38.qu SE
95 Mo 8:033 ..o 8:805 28 mmnm 88 ”53:33:33? .3 8 mos: BEE Esp—5:88.. 2m: 43 .3293? cm Mo 2283M .20 2an

Eomn< u< 4:085 um A3829 3.8—a 0333325 93 «53283 508 msoocowESoI "I 633885 um .EfiflmoM HM

 

 

a < x .85an < a m 83.8 a < m 39.8
< a m 3:8. a < a 83.8 < a m 39.8
a < m 22 _ a < a 343m 4. < a 32.8
< a 2 :38 a < a $48 a < a 255m
3 < m 33.8 a < a 83.8 < a 2 52.8
< a m $38 a. < 2 23.8 a < 2 £38
a < a :38 a < x 23.3 a < a 39.8
< a : 83.8 a < m 23.8 a < x $38
a < a 238 < a m 23.8 a < x 33m
.— a a ”$48 a < a 243m a < a $9.8
SE 2%: 3:2 8583.. SE 2%: .53. 8:88.... :5 2%: Sam 858%

 

338v 2 .u 2.3

149

40-
35-
30—

20-
15‘
10‘

Freque

 

2000 2300 2600 2900 3200 3500 3800 4100
Yield (kg ha")

40-
35-
30-1
25‘
20'
15‘
104

Frequency

 

18 19 20 21 22 23 24
100 seed weight (g)

40-
35+
30~
25-
20-
15~
10«

Frequency

 

 

0 a
45.25 46 46.75 47.5 48.25 49 49.75 50.5 51.25

Days to Flower

Figure C. 1. Frequency distributions for agronomic and canning quality traits in the ‘Jaguar’
by 115M RIL population.

150

407
35‘
30‘
25‘
20-

Frequency

15—
10‘

 

   

—
7 Y

93.5 94.5 95.5 96.5 97.5 98.5 99.5 100.5

35-
30-
25-
20-
15-
10-

Frequency

 

40-
35-
30-
25-
207
15-
104

Frequency

 

Days to Maturity

1.3 1.6 1.9 2.2 2.5 2.8 3.1
Lodging Score

 

 

47 47.8 48.6 49.4 50.2 51 51.8
Plant Height

Figure C.1 (cont’d).

151

50
4O
30
20

Frequency

2.8 3.2 3.6 4 4.4 4.8 5.2 5.6 6

Desirability Score

35
30
25
20
15
10

Frequency

1.5 1.8 2.1 2.4 2.7 3 3.3 3.6

Visual Appearance

Frequency

13.25 14 14.7515.516.25 17 17.7518.519.25 20 20.75

Canned Bean Color

Figure C.1 (cont’d).

152

Frequency

42 45 48 51 54 57 60 63 66 69 72 75 78 81 84

Texture (kg—force)

Frequency

U1
__.i

234 238 242 246 250 254 258 262 266
Washed-drained weight (3)

Figure C.1 (cont’d).

153

0.0
2.6

7.2

16.9

26.6

0.0

1 2.7

19.9

33.1

81

 

 

 

 

 

 

 

 

 

 

 

 

 

F1 3R2.250
F4125.“ 0

PVBR1 07

0.0

PVBR233
19.9

F3R1 .235

I7E1 .625

81 b 82

 

 

IAC26 0.0 M006 0.0

6.3

12.3
16.7

 

14.7 BM156

 

 

 

Ant73

P113394 21-2

PVBR76

 

22.1
26.4

 

 

 

M16
”451 .375

 

IAC66 F22R1.400

M1 53.1 75
851620

 

7.1

F7R1.150

Ruat3:22 F10R6.150

 

10.1
PVItOM

F6R10.200
F15R10.560

 

1 3.4

FJ16

 

16.3
BM171

 

”356.320 F12R7.250
F20R4.190

F20R4.250

 

 

22.1

F16R6.900

F6R5.650
IAC96

F1 R9375
PVBR61

F1 3R5.230
F10R5.275
F10R5.265
F7R2.600
F1 3R5.220
F1 3R5.1 75
85121 3

 

B3

 

“1 56.550

 

IAC32

 

”121.150
"23.325
Pvuoos

 

 

 

 

 

m

0.0
03>

. Qé mango
1.5 /

\ PVBR14

I

 

F6R1 .540
F6R2.350

10.6
14.2

 

 

F1 3125.420
MR163

2 / 3111187
111436.600
14355.326
F2131.3so
/ 111736.690
/ "453.1200
—- 111cc:

» smears
11111574100
\ F10R9.380
\ M3E5.375

20.0
22.6

 

8

io

/
111111

 

.5".

~6
\

[I

56.0 ~—w 137M2
56.6 ’5“ "756.140

 

 

Figure C.2. Linkage map of ‘Jaguar’ll 15M RIL population consisting of 119 SSR, SRAP,
TRAP, and phenotypic markers placed on 15 linkage groups covering a combined distance of
4600M.

154

87 86 B10 B11

 

 

 

 

 

 

 

 

 

 

 

  
   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.0 F22R4.550 0.0 31439.1000 0.0 PvBR181 0.0 3736.620
2.9 _.... 3111166
6.6 N.» 3111210 3°“ ”“3““
7-1 ““9113”? 10.4\ /317310.320 7,7 3141194200
110 "”550 16.6 ... 3232.676
' 12.6 32232.326 25-8\ PvBRm 12.6 31736.420
14.7 31133269 37.3 . 31637.266
40-4 L11 WW” 17.2 3234.200
42.0 3636.410
22.0 3333.200 43-2 / F17R3-580
24.3 3132.336 43.6 316361100
:55: 1511;616:3620 26.6 31333130
26.6 3.126 “.7 M1113.
47.2 3636.660
47.6 F13R3.135 36.2 36310.476
46.3 - 317310.360
49.2 1 31734.226
46.6 Ir 3131.660
45,0 3139.400 60.0 1 'F5R5.690
61.3 1 31936790
493 1155340 52.2 F17R4.275
62.7 F17R4.265
63.0 F18R2.775
63.9 F19R6.1200
66.6 312310.260
L61 L62 L63

 

 

0.0 F1 6117.290 0.0 M3E3.545 0.0 IAC63

 

 

 

9.4 31634.490 7-1 "466.235

9.7 ”256.400

 

 

 

15.2 M3E4.470

Figure C.2 (cont’d.)

155