W9. ..
f“,.-,,‘u.
I ‘. .‘

.
4 L
"l “1'
.

 

' a."
1"...“

31'? & “ '

: f j, .’ _ ‘ a‘ L 1:, .

-§’§ , “5. . , Egg 1233;; iv
- ”o ( “in?

.1 ‘51}
m“: .

my
‘{...3

..

11:74”
.. . 2'“.
I w‘-

‘

4r

 

This is to certify that the
thesis entitled

USE OF MARKER-ASSISTED SELECTION TO BREED FOR
RESISTANCE TO COMMON BACTERIAL BLIGHT IN
COMMON BEAN

presented by

Patrick Daniel O’Boyle

has been accepted towards fulfillment
of the requirements for the

Master of Science degree in Plant Breeding and Genetics

 

 

Major Professor’s Signature /
fj/Lflq .3, 2, c9 0 é/
9‘ / ’

Date

/

 

MSU is an Affirmative Action/Equal Opportunity Institution

 

LIBRARY
Michigan State
University

 

 

 

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

 

W35?! 0&3?

 

S

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:lClRC/DateDue.p65«p. 15

USE OF MARKER-ASSISTED SELECTION TO BREED FOR
RESISTANCE TO COMMON BACTERIAL BLIGHT IN COMMON
BEAN

By

Patrick Daniel O’Boyle

A THESIS

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

MASTER OF SCIENCE

Plant Breeding and Genetics Program
Department of Crop and Soil Sciences

2004

ABSTRACT

USE OF MARKER—ASSISTED SELECTION TO BREED FOR
RESISTANCE TO COMMON BACTERIAL BLIGHT IN
COMMON BEAN

By

Patrick Daniel O’Boyle

Resistance to common bacterial blight (CBB) in common bean is a quantitative
trait. Although previous attempts to breed CBB resistant bean cultivars have had limited
success, progress in marker-assisted selection (MAS) has created new opportunities for
breeding programs. The potential of the marker SU91 in MAS for CBB resistance was
evaluated in field experiments. SU91 was correlated with lower CBB leaf scores in East
Lansing (r = -O.50), and Saginaw (r = -O.59), and correlated with pod resistance in
Saginaw (r = -0.48). SU91 exhibited a negative correlation with yield (r = -O.20) and a
positive correlation with maturity (r = 0.22) in East Lansing, but showed no association
in Saginaw. Selections carrying SU91 were crossed with additional sources of resistance
to CBB and anthracnose. F2 progeny were screened for the marker BC420, linked to a
QTL for CBB resistance on bean linkage group B6. The effects of SU91 and BC420
were examined in two greenhouse studies. The presence of SU91 was negatively
correlated with CBB disease ratings for leaves (r = -0.20) and pods (r = -0.27). Presence
of BC420 was only correlated with pod ratings (r = -0.19) in experiment one, and CBB
leaf resistance (r = -O. 18) in experiment two. Presence of both markers resulted in lower

levels of CBB resistance than provided by either marker alone.

DEDICATION

To all the wonderful family and friends who have been there with love,
support, and friendship.

iii

ACKNOWLEDGMENTS

First, I wish to thank my major advisor, Dr. James Kelly, for giving me the
wonderful opportunity to study plant breeding and genetics in his research program. I
will always look to him as the standard for a plant breeder who has successfully been able
to combine the art and the science of plant breeding. His “breeder eye” is something that
will always be admired. The art of plant breeding, which he understands and teaches,
extremely well, is something that will never be learned from any class or from reading a
book. His willingness to work with students and impart some of this knowledge and
experience to them is part of what makes him an excellent mentor.

In addition, I wish to thank the other members of my committee. Dr. Kirk was
always willing to meet and provide me with advice regarding plant pathology, in addition
to always making me laugh on those days when I really needed it. Dr. Hancock helped
me in ways that go far above and beyond my actual research, by helping to inspire me to
become a more balanced student. Both of these individuals are great not only because of
their actual research, but more importantly, because of the passion that they share for the
education of others.

I would also like to thank the lab of Dr. Hammerschmidt, who all were very
helpful. They allowed me to use their equipment, and served as a forum for my basic
plant pathology questions.

Of course, my experience at MSU would not have been complete without my

fellow bean lab members. Jerry, Norm, Shitaye, Brian, Belinda, Marcio, Evelene,

Esteban, Karolyn, Ann, and Mark have all helped make my experience here more
enjoyable. I would not be the same without the friendship and support they have given
me. In particular, I would like to thank Veronica Vallejo. She has not only been a major
influence in my education, but is a great friend as well.

Finally, I would like to thank my wonderful family (Dad, Mom, Kelly, Mike, and
of course, my wife Tiffany). I could never have made it without their love and support. I

only hope that I am able to support them as they have for me.

TABLE OF CONTENTS

LIST OF TABLES ................................................................... viii
LIST OF FIGURES .................................................................. x
INTRODUCTION .................................................................... 1
OBJECTIVES .................................................................... 28
REFERENCES ................................................................... 29
CHAPTER ONE:

THE USE OF MARKER-ASSISTED SELECTION TO ENHANCE THE
LEVEL OF RESISTANCE TO COMMON BACTERIAL BLIGHT IN
COMMON BEAN LINES.

ABSTRACT ..................................................................... 36

INTRODUCTION ............................................................... 38

MATERIALS AND METHODS .............................................. 44

RESULTS ........................................................................ 53

DISCUSSION .................................................................... 71

REFERENCES ................................................................... 77
CHAPTER TWO:

THE USE OF MARKER-ASSISTED SELECTION TO PYRAMID
RESISTANCE TO COMMON BACTERIAL BLIGHT AND
ANTHRACNOSE IN COMMON BEAN (Phaseolus vulgaris L.)

ABSTRACT ...................................................................... 80
INTRODUCTION ............................................................... 82
MATERIALS AND METHODS .............................................. 88
RESULTS ......................................................................... 91
DISCUSSION .................................................................. 107
REFERENCES .................................................................. 11 1

vi

APPENDICES :

A1: COLLECTION OF Xanthomonas axonopodis pv. phaseoli (Xap)
ISOLATES FROM MICHIGAN DRY BEAN FIELDS .................. 114

A2: CHARACTERIZATION OF RESISTANCE TO COMMON
BACTERIAL BLIGHT (CBB) BETWEEN THE KIDNEY BEAN
CULTIVARS ‘MONTCALM’ AND ‘REDHAWK’ ..................... 122

A3: DAYS TO FLOWERING DATA FOR 38 F 5 SELECTIONS
FROM SAGINAW, MI 2002 ................................................. 127

A4: DAYS TO FLOWERING DATA AND RE-GREENING DATA
FOR F3 LINES, EAST LANSING 2003 ..................................... 129

vii

LIST OF TABLES

Table 1.1 A summary of PCR profiles used in screening for the SCAR markers
SU91, SAP6, BC409, BC420 ................................................. 48

Table 1.2 Common Bacterial Blight Leaf and Pod Ratings of 30 F6 lines and check

cultivars or breeding lines, East Lansing
(2003) ............................................................................. 55

Table 1.3 List of genotypes evaluated in preliminary yield test in Saginaw, MI
(2003) ............................................................................. 56

Table 1.4 Mean leaf and pod common bacterial blight (CBB) ratings for SU91+

and SU91- F 6 lines from field plots at East Lansing and Saginaw, MI
(2003) ............................................................................. 57

Table 1.5 Common Bacterial Blight Leaf Ratings, East Lansing (2003) ........... 56

Table 1.6 Yield (T/ha) of bean genotypes with or without the SCAR marker

SU91 compared to check varieties and cultivars, East Lansing, MI
(2003) ............................................................................ 66

Table 1.7 Yield (T/ha) of bean genotypes with or without the SCAR marker
SU91 compared to check cultivars, Saginaw, MI (2003) .............. 69

Table 2.1 Marker data, phenotypic CBB ratings on leaves and pods, and
agronomic desirability of 93 F3 selections from a field study
established in East Lansing, 2003 ......................................... 93

Table 2.2 Relationship of the SCAR markers SU91 and BC420 with leaf
and pod resistance in field inoculations of 320 F23 (East Lansing,
2003) .......................................................................... 96

Table 2.3 Molecular marker data and greenhouse inoculation results
(2003-04). Data were presented by genotype. All bean genotypes
except controls are F14 lines and were selected in the field
(2003). CBB leaf and pod ratings were averaged across 4
replications ..................................................................... 99

Table 2.4 Correlation coefficients of the SCAR markers SU91 and
BC420 with leaf and pod resistance in two greenhouse
inoculation experiments ..................................................... 102

viii

Table 2.5 A comparison of the mean CBB leaf and pod ratings for various
molecular marker combinations. Results are presented as an
average of two greenhouse experiments that exhibited similar
conclusions. Mean ratings are averages of 4 replications,
determined using 1-9 scales for leaves and pods (1= no
symptoms, 9= severe symptoms). Separation of means,
designated by different letters was determined using
LSD0_05 .............................................................................. 104

Table Al.1 Summary of information regarding Xap isolates collected from

infected bean tissue at Saginaw and Montcalm research farms, in
Michigan, in 2002 ................................................................ 119

Table A3.1 Summary of marker genotype (SU91+ or SU9l-) and days to flowering for
38 F 5 selections from Saginaw, MI 2002 ..................................... 127

Table A4.1 Summary of marker genotype (SU91+, SU91-, BC420+, and
BC420-), days to flowering, and re-greening data for 38 F5
selections from Saginaw, MI 2002 ............................................. 129

LIST OF FIGURES

Figure 1.1 Order of crosses made in the development of populations with

Figure 1.2

Figure 1.3.

Figure 1.4.

Figure 1.5

Figure 1.6

increased CBB resistance. A) The resistant parent VAX 5 was

crossed with the MSU elite black bean breeding line B98311.

B) Two of the resulting lines that were verified to carry the SCAR

marker SU91 were used in crosses with ‘Phantom’ (black bean

cultivar), N99219 (navy bean breeding line), B98306 (black bean

breeding line), and 800136 (black bean breeding line). Crosses were
performed in both directions, using each parent as pollen parent and

female parent in separate crosses. The 99L91-45 and 99L91-47 parents
were used in crosses as F335 plants ................................................ 47

Frequency of leaf CBB ratings between bean lines with SU91 (SU91+)

and those without SU91 (SU91-). All 72 entries from the navy and

black bean preliminary yield trials are represented. The rating scale

used was the previously described 1-9 scale (1 = no infection,

9 = severe necrotic lesions). The minimum and maximum mean leaf
ratings for this study were 3 and 8, respectively ................................ 58

Frequency of pod CBB ratings between dry bean lines with SU91

(SU91+) and those without SU91 (SU91-). All 72 entries from the

navy and black bean preliminary yield trials are represented. The

rating scale used was the previously described 0-3 scale (0 = no pod

lesions, 3 = severe pod lesions). The minimum and maximum mean

pod ratings for this study were 0 and 1.5, respectively ........................ 59

Comparison of mean yield between lines with (SU91+) and without
(SU91-) the SCAR marker SU91. The mean yields at East Lansing of
2.05 and 2.20 T/ha for SU91+ and SU91- lines, respectively, were
significantly different (p= 0.02). Mean yields at Saginaw of 3.08 and

2.96 T/ha, for SU91+ and SU91- lines, respectively, were not significantly
different (p= 0.08) ................................................................. 60

Maturity is compared between SU91+ and SU91- lines at East Lansing,
2003. Days to maturity was recorded as the average for 3 replications

of each line. The overall mean maturity for the experiment was 95 days.
The mean maturity for both SU91+ and SU91- genotypes was also 95

days .................................................................................. 63

Maturity is compared between SU91+ and SU91- lines at Saginaw, 2003.
Days to maturity was recorded as the average for 3 replications of each
line. The overall mean maturity for the experiment was 99 days. The mean

Figure 2.1.

Figure 2.2.

maturity for both SU91+ and SU91- genotypes was also 99
days ............................................................................................................. 64

A comparison of the mean CBB leaf and pod ratings for various
molecular marker combinations. Results are presented as an average

of two greenhouse experiments that exhibited similar conclusions.

Mean ratings are averages of 4 replications, determined using 1-9

scales for leaves and pods (1= no symptoms, 9= severe

symptoms) ....................................................................... 103

A summary of the molecular marker composition of the 93 F3 lines
selected from the field (East Lansing, 2003). SU91 is a SCAR
marker linked to a CBB resistance QTL on bean linkage group B8.
BC420 is a SCAR marker linked to a CBB resistance QTL on bean
linkage group B6. SAS13 is linked to the Middle American
anthracnose resistance gene C0-42. Marker combinations (1-8)
correspond to the following: l= -/-/-, 2= SU9l+/-/-,
3= SU91+/BC420+/—, 4= SU91+/-/SASI3+, 5= -/BC420+/SAS13+,

= -/BC420+/-, 7= —/-/SASl3+, 8= SU91+/BC 420+/SASI3+ ............. 105

xi

LITERATURE REVIEW

Impacts of Common Bacterial Blight Infection

Common bacterial blight (CBB) is one of the most serious production problems of
common bean (Phaseolus vulgaris L.) worldwide (Tar’an et al. 2001). CBB is a
seedbome disease, caused by the bacterial pathogen Xanthomonas axonopodis pv.
phaseoli (Smith) Dye (Xap). CBB is often responsible for significant yield losses and
entire crop failures (Vakili et al., 1975). Studies by Wallen and Jackson (1975) and
Yoshii et a1. (1976) detemtined that yield losses ranged from 38% to 45% in Xap-
inoculated plots, and were 22% due to natural CBB infections (Yoshii et al., 1976).
Losses are typically more severe in hot years, when average temperatures exceed 30°C
(Saettler, 1989; Kelly and Copeland, 1996).

The seedbome nature of CBB has been a major constraint in commercial bean
production, and has also influenced the distribution of the dry bean seed production
industry. To attain certification, bean seed fields must be free of the disease, because
infected seeds will transmit the infection to their progeny (Kelly and Copeland, 1996). In
areas where CBB severity is high, the production of dry bean seed is not economical for
producers, resulting in the displacement of seed production from areas that once had
highly successful production programs. In-state production of certified, disease-free seed
is ideal for Michigan growers because of reduced costs and assurance of sufficient
quantity (Anderson et al., 1970). A 92% reduction in acreage (from 5,000 ha to 400 ha)
dedicated to certified navy bean production in Michigan between 1974 and 1994 was

primarily due to losses caused by CBB (Kelly and Copeland, 1996). Competition from

westem states where seed is produced in relatively disease-free environments has resulted
in displacement of the bean seed industry from Michigan.

In direct contrast to navy and black bean seed production in mid—Michigan, the
successful production of certified dark red kidney bean seed in northern Michigan
increased in acreage over the same time period (from 400 to 1,200 hectares). This
increase was primarily due to two factors. The first is the geographic isolation of kidney
bean seed production from infected commercial bean fields downstate, where other bean
market classes are grown and produced. The second factor that accounts for the success
of kidney bean seed production in Michigan is the use of moderately resistant cultivars,
such as Montcalm (Kelly and Copeland, 1996). These factors have heavily impacted the
success of the kidney seed production industry. Over the course of a 15-year period of
observation, Kelly and Copeland (1996) noted a 2—4% average of seed lot rejection per
year for certified kidney beans. During the same period, navy beans had seed lot
rejection rates of as high as 50%. In 1994, 50% of all navy bean certified seed was
rejected primarily due to CBB infection, although the seed borne fungal disease, bean
anthracnose was also responsible for a portion of the rejections. CBB has played a
significant role in the reduction of certified navy and black bean seed production in
Michigan (Kelly and Copeland, 1996).

The seed borne fungal disease bean anthracnose is another serious threat to dry
beans in Michigan. This disease is caused by the hemibiotrophic fungal pathogen
Colletotrichum lindemuthianum (Sacc. & Magnus) Lams.-Scrib. In the early 1900’s,
anthracnose was considered as one of the most serious bean diseases in the United States

(Zaumeyer and Thomas, 1957). Since that time, however, the deployment of genetic

resistance has decreased the impact of this disease in the dry bean industry. The
combination of anthracnose resistance and CBB resistance in high yielding dry bean lines
would greatly benefit growers and seed producers by allowing them access to certified

seed with resistance to the two major seed-home diseases of dry beans in Michigan.

Causal Organism Characteristics

Xanthomonas axonopodis pv. phaseoli is a gram-negative, rod-shaped bacterium
that is aerobic and has a polar flagellum. Most isolates display mucoid growth on media,
however, some have been known to have a grainy growth habit. The majority of isolates
produce a yellowish carotenoid pigment (xanthomonadin) on nutrient media. An exudate
produced by the bacteria (in cultures and in the host plant) is a high molecular weight
extracellular polysaccharide that has a demonstrated role in symptom development and
bacterial survival (Saettler, 1989). A study by Leach et al. (1957) examined the role of
this polysaccharide in bacterial fitness. The results suggest that the exudate may play a
role in tolerance to ultraviolet light. The exudates induced wilting symptoms in beans,
but it also produced symptoms in non-host species. This indicates that the polysaccharide
induces symptoms not by specific toxicity, but by blocking water transport in the plant.
Examination of water uptake confirmed water movement disruption in tomato
(Lycopersicon esculentum Mill.) and sunflower (Helianthus annuus L.) (Leach et al.,
1957). The bacteria are capable of causing symptoms in a large number of Phaseolus and

Vigna species, including common bean and COWpea (Saettler, 1989).

CBB Symptoms and Epidemiology

The first sign of Xap infection in beans is usually the appearance of tiny water-
soaked spots on the underside of leaves. Eventually, these spots coalesce, forming
necrotic spots surrounded by a border of lemon-yellow chlorotic tissue, noticeable on the
upper leaf surface. Eventually, diseased leaves become necrotic, and plant defoliation
may result. In some cases, leaves may become entirely necrotic and remain attached to
the plant following maturity. This is true especially in cases of extremely severe
infections (Saettler, 1989).

All aerial plant organs (leaves, pods, seeds, and stems) may exhibit CBB
symptoms, although leaf symptoms are the most noticeable. Pod lesions originate as
water-soaked areas that become dark and sunken as the disease progresses. The
symptoms on pods can result in seed discoloration and/or shriveling, when infections are
severe (Saettler, 1989). Although seed infection is more commonly observed when pod
lesions are visible, Cafati and Saettler (1980c) reported infected seeds in symptomless
pods. Seed symptoms can be difficult to observe on dark-colored seed coats. Symptoms
on light colored seeds (e. g. navy beans) can easily be observed as yellow-brown spots
(Zaumeyer and Thomas, 195 7). Occasionally, the hilum of infected seeds becomes
discolored. Stern lesions, although less common, can result in girdling of the entire stem,
disrupting water and nutrient transport. This is more common in plants that originated
from contaminated seeds, compared to plants infected during the course of the growing
season (Zaumeyer and Thomas, 1957). Occasionally, yellow bacterial exudates can be

observed in the vicinity of a lesion. These exudates contain living Xap cells, and serve as

a major source of secondary inoculum, as these bacterial populations may easily be
moved, from infected leaves to healthy plants (Zaumeyer and Thomas, 1957).

Xap commonly enters the leaves through natural openings, such as wounds or
stomata, and proceeds to occupy the intercellular spaces. As bacterial colonies form, they
damage cells, and dissolve the middle lamella. The bacteria can also enter the stem,
which can result in clogging of vascular tissue, causing the plant to wilt. The seed is also
susceptible to infection, most commonly through the pod’s vascular system, although the
pedicel may also be an entry point for the pathogen (Saettler, 1989). The disease cycle is
perpetuated because an infected seed holds considerable potential for transmission of the
pathogen to the resulting seedling (Aggour, 1989). Seedlings originating from Xap
infected seeds commonly exhibit lesions on cotyledons and primary leaves. These
symptoms will typically appear earlier in the growing season, beginning around the time
of flowering (Zaumeyer and Thomas, 1957). The translocation of Xap from infected to
healthy leaves in susceptible genotypes, was observed by Cafati and Saettler (1980a).
Translocation of bacteria was not observed in CBB resistant tepary bean selections,
suggesting that the limitation of bacterial movement may be a consequence of host
resistance (Cafati and Saettler, 19803; Aggour et al., 1989).

Since CBB is a seed-transmitted disease, Xap infected seed may initiate an
epidemic after planting. Infected seeds that do not exhibit visible external symptoms are
regarded as the primary source of inoculum (Jung et al., 1997). Although the survival of
Xap in contaminated bean seeds has not been fully examined, viable cells have been

recovered from bean seeds stored forlO years (Zaumeyer and Thomas, 1957).

Xap is readily disseminated by wind-driven rain, mechanical transmission by
cultivation, and human or animal passage through wet bean fields. According to Claflin
et al. (1973), wind-blown soil can play a major role in the dissemination of Xap by
wounding bean plants, creating an entry point for the pathogen, and the soil particles
themselves may be responsible for moving bacteria. This study revealed that CBB
infection increased with wind speed and exposure time. Steadman et al. (1975)
determined the potential for Xap to spread via irrigation water in regions where bean
crops are irrigated. Another factor which impacts the degree of infection is the
pathogen’s ability to persist in the field for an extended period of time (Anderson et al.,
1970). Schuster (1970) noted that Xap can persist in bean straw for at least 10 months in
Nebraska. After 22 months, however, no Xap had survived in the field. Saettler et al.
(1986) concluded that Xap does not overwinter in Michigan. A collection of 20 isolates
was investigated, to account for possible differences in the ability of distinct isolates to
overwinter.

As with many plant pathogenic bacteria, Xap can live epiphytically on the leaves
of non-host species for a short period of time. Viable Xap cells have been recovered
from soybean (Glycine max L.), maize (Zea mays L.), cowpea (Vigna unguiculata L.),
and several weed species (e. g. C henopodium album and Amaranthus retroflexus). The
methods of bacterial transmission from infected bean plants to healthy bean plants may
also move Xap from non-host species to healthy beans (Cafati and Saettler, 1980b;
Saettler, 1989).

Kaiser and Vakili (1978) examined the possibility of insect transmission of Xap

from infected to healthy bean plants. They concluded that some insect species play a role

in transmission, however it is not considered to be a major source of spread in CBB
epidemics. This research was conducted in Puerto Rico, and it is possible that the results
obtained may not apply to other regions, depending on insect species present and climatic
conditions. Unfortunately, similar studies have not been conducted in most major areas
of dry bean production, so it is unknown whether these results apply in other regions.

An additional method by which Xap can spread is from seed to seed contact,
during threshing. Bacteria on the exterior of seeds may become airborne upon threshing
and can thus, be moved to uninfected seeds. These bacteria may serve as source of
primary inoculum the following growing season in cases where seed is kept by the farmer
for planting (Weller and Saettler, 1980). This is also a concern in the seed production
industry, where it can result in apparently uninfected seeds being sold to farmers, thereby

introducing the pathogen into bean fields the following year.

Environmental Conditions

Climatic conditions favoring CBB infection in bean fields have been studied. The
disease is generally considered a hi gh-temperature disease, although it may occur over a
wide range of temperatures. Goss (1939) found that bean plants exhibited CBB
symptoms over the entire range of temperatures tested (16 to 32°C); however, the higher
temperatures favored more rapid symptom development. Symptoms occurred most
rapidly at temperatures between 28 and 32°C. Inoculated plants grown at 32°C displayed
symptoms at 6 days after inoculation, while symptoms were not noticeable in plants
grown at 24°C until 14 days after inoculation. Plants grown at 16°C did not display CBB

symptoms until placed at 28°C for 7 days (Goss, 1939).

The effect of moisture was also examined in the experiment by Goss (1939). A
comparison of low and high humidity after inoculation revealed that symptoms
developed significantly earlier and affected a greater area in conditions of low humidity.
At least some moisture is required for symptom development, however, it seems the
disease is favored by low to moderate humidity (Goss, 1939). Timely precipitation
accompanied by wind during the period between flowering and maturity can also play an
important role in disease development, as this is one of the major methods by which Xap

spreads from plant to plant (Zaumeyer and Thomas, 195 7).

Disease Management and Breeding for CBB Resistance

Various disease management strategies have been implemented to control CBB,
but have achieved limited success. Some of the methods used have been: planting of
disease free seed, bactericide seed treatment, crop rotation management, deep-plowing,
avoiding the reuse of irrigation water from infected fields, and isolation of diseased fields
(Zaumeyer and Thomas, 1957; Anderson et al., 1970). The most effective of these
methods is to plant disease-free seed, however, this has not been practical in some
locations due to insufficient supply of seed certified free of Xap (Saettler, 1973). The use
of copper sprays and other chemical methods has also been widely examined in
controlling CBB, with limited success (Saettler, 1971; Saetller, 1973). Saettler (1971)
found that even after six applications of various bactericidal chemicals, the disease was
still not controlled effectively. Many bactericide seed treatments (e. g. Streptomycin)
have been effective in decreasing the amount of primary inoculum on the seed surface,

but this does not reduce the titer of the bacterium within the seed itself (Saettler, 1989).

When possible, early planting may allow plants to escape CBB infection by reaching
maturity before the infection becomes severe, as the bacteria typically thrive in the later,
hotter portion of the growing season (Zaumeyer and Thomas, 1957).

Until recently, breeding for genetic resistance has not been a reasonable option for
protecting against CBB infections. Genetic linkage with negative agronomic traits (in
particular, grain yield) was prohibitory to the successful development of useful CBB
resistant varieties (Kelly and Copeland, 1996). Well-adapted CBB resistant breeding
lines in various bean market classes have now emerged from different bean breeding
programs. These successes have been responsible for the emergence of CBB resistance
breeding as the preferred option in preventing CBB epidemics (Singh and Munoz, 1999).

Due to the reasons discussed above, the development of resistant cultivars
provides the single, most efficient method of CBB control (Singh and Munoz, 1999).
However, the CBB resistance sources currently available only confer partial resistance.
Resistant varieties still need to be planted in an integrated management system consisting
of: crop rotation, weed control, avoiding passage through wet bean fields, seed treatment
with bactericide slurry, isolation of known Xap infested fields, and deep-plowing of

infected bean debris (Zaumeyer and Thomas, 1957; Saettler, 1989).

Genetic Nature of CBB Resistance

CBB resistance in common bean is a quantitatively inherited trait that exhibits
low heritability, controlled by an unknown number of genes. Genetic inheritance studies
conducted in various breeding programs with different genetic populations have found

differing results regarding the genetic nature of CBB resistance. Most studies have

reported low (h2= 0.14; Coyne and Schuster, 1974b) to moderate heritability for CBB
resistance, but high heritability (h2= 0.74) values have been reported (Tar’an et al., 2001).
A range of values for narrow-sense heritability from h2= 0.11 to 0.87 were reported for
various plant organs in a study by Silva et al. (1989). Heritability for CBB resistance
seems to vary greatly based on resistance source used and genetic background of the
recipient.

Past research has also reported differing results regarding the expected number of
genes involved in the expression of CBB resistance, as well as their mode of action. Data
published by Adams et al. (1988) suggested that a single recessive gene from a snap bean
line (A-8-40) conferred leaf resistance to CBB. The possibility exists, however, that
modifying genes with minor effects could also be involved. The presence of additive
gene effects for CBB resistance has been reported, while other studies have suggested
incomplete dominant gene action. Coyne and Schuster (1974b) determined that CBB
resistance was dominant in a cross between GN 1140 (susceptible parent) x PI 207262
(resistant parent), and that a small number of genes controlled the resistant reaction.
Musaana et al. (1993) determined that the number of genes associated with resistance to
CBB varied depending on the parents used in population development, and the stage at
which the plants were evaluated. Leaf resistance was controlled by one to three genes at
an early growth stage, one to four genes later in maturity, while pod resistance was
conditioned by one to two genes.

Tepary bean (Phaseolus acutifolius), a relative of common bean that exhibits
CBB resistance, has been useful for breeders attempting to enhance resistance present in

P. vulgaris gerrnplasm (Singh and Munoz, 1999). Resistance in tepary bean leaves and

10

pods appears to be explained by a single dominant gene (Drij fhout and Blok, 1987).
When CBB resistance in tepary bean was introgressed into common bean, the resistance
was found to be quantitatively inherited (Coyne and Schuster, 1965). CBB resistance
introgressed from tepary bean was studied in three different crosses by Scott and
Michaels (1988), and resistance was determined to be controlled by two dominant genes.
A study of CBB resistance in four tepary x tepary crosses generally agreed with the
results proposed by Drij fliout and Blok (1987), determining that CBB resistance was
conditioned by one dominant gene, although in one cross, the interaction of two
complementary dominant genes was responsible for CBB resistance (Urrea et al., 1999).
Scarlet runner bean (Phaseolus coccineus) is another donor of CBB resistance
used in interspecific crosses with P. vulgaris. Resistance levels within this species are
considered intermediate to the high levels in P. acutifolius and the low levels in P.
vulgaris (Singh and Munoz, 1999). Welsh and Grafton (2001) examined the inheritance
of CBB resistance in common bean cultivars, with resistance derived from P. coccineus
and determined that resistance was controlled by a single recessive gene. Although this
gene appeared to account for most of the genetic variation in phenotypic CBB ratings, the

presence of minor modifying genes may also be involved.

Obstacles to the Development of Genetic Resistance to CBB in Dry Bean

Breeding for genetic resistance to CBB in common bean is complicated by many
factors. The complex genetic nature of the host resistance is one obstacle to the
development of cultivars with enhanced levels of CBB resistance. As previously

mentioned, CBB resistance is a quantitatively inherited trait of low to moderate

11

heritability, controlled by an unknown number of genetic loci (Park et al., 1999).
Although bean genotypes with CBB resistance have been identified, the resistance is only
partial in nature, and as a result, does not provide complete protection from pathogen
infection (Park et al., 1999). A second complication of quantitative disease resistance is
that the genetic loci involved are subject to variation depending on environmental
influences (Paterson et al. 1991). The level of expression of quantitative resistance in
contrasting genetic backgrounds is another obstacle in breeding for CBB resistance.

A third consideration in breeding for CBB resistance is the differential response
of plant organs to CBB infection. There is little correlation between seeds, pods, and
leaves in their reaction to CBB (Amaud-Santana et al., 1994). Valladares-Sanchez et al.
(1979) demonstrated that many Phaseolus genotypes may exhibit resistance in leaves,
while the pods of the same plant exhibited a susceptible reaction to CBB. In addition,
there is an absence of correlation between seed and pod infections. There may be a
differential reaction for the interior and exterior of pods within the same genotype,
however, this difference is not as pronounced as the difference between pod and leaf
reactions. In order to attain the desired level of resistance, it may be necessary for
breeders to simultaneously breed for both pod and leaf resistance. An important
consideration regarding pod resistance is that it can minimize the possibility of disease
transmission to progeny (Valladares-Sanchez et al., 1979). Drij fhout and Blok (1987)
observed high correlations between leaf and pod resistance in tepary bean. In two
different populations (developed by an interspecific cross between P. acutifolius and P.

vulgaris), leaf and pod reactions were positively correlated (r2: 0.99).

12

A final problem in breeding for CBB resistance is the tendency of the resistance
to be unstable. CBB resistance has been known to segregate after many generations of
multiple selections and evaluations. This has complicated attempts to develop highly
resistant, true breeding lines. The reasons for this instability of resistance are currently
unknown (Singh and Munoz, 1999).

When breeding for quantitative traits (e. g. CBB resistance in bean), genes that
have a negative impact in the plant may be inherited along with the trait, making
quantitative trait breeding a challenging endeavor (Dudley, 1993). The association of
resistance with negative traits, a concept known as linkage drag, has been repeatedly
noted in attempts to breed for CBB resistance. A study conducted by Valladares-Sanchez
et al. (1979) reported a linkage between CBB resistance and late maturity in common
bean. CBB resistance was also associated with indeterminate plant growth habit in this
study.

Pathogen variability further complicates development of CBB-resistant dry bean
lines. Bacteria are prone to frequent mutations, which can result in their overcoming
resistance. Mutations are not the only consequence of pathogen variability that impact
attempts to breed for resistance to CBB. Because many different strains may be present
or absent in regions where beans are grown, the possibility of the accidental introduction
of a strain from one region to another can affect the development of CBB-resistant lines

(Schuster and Coyne, 1975).

13

Pathogen Variability and Resulting Complications for Resistance Breeding

Most of the characterization of Xap has been conducted by inoculating multiple
bean cultivars. Isolates differing in pathogenicity (the capability to cause disease) across
bean genotypes are classified by bean pathologists as different strains (or descendents of
a single isolation), but a well characterized structure of genetic variability has not yet
been determined for Xap. The lack of information regarding variability of this pathogen
is an obstacle in the application of host genetic resistance.

One trait that has been examined in Xap characterization is the production of a
yellowish polysaccharide believed to be the symptom-inducing agent. Corey and Starr
(1 95 7) examined the polysaccharide production of four Xap variants, and determined that
the amount of this toxic polysaccharide produced was correlated with lesion size.
Isolates of Xap differ greatly in the production of this polysaccharide, suggesting that this
may be a useful characteristic to use in isolate differentiation. Isolates that produced this
compound were formerly classified as X. phaseoli var. fuscans, but are no longer
categorized separately from X. campestris pv. phaseoli, the former name of X.
axonopodis pv. phaseoli (Schuster and Coyne, 1975).

Pathogenicity has been more commonly used to differentiate between isolates of
Xap. Schuster and Coyne (1971) reported new strains isolated from Colombian bean
seeds that were pathogenic on the CBB resistant breeding line GNN #1 Se]. 27. This
breeding line had previously demonstrated tolerance to many North American Xap
strains, however, it was susceptible to both Colombian strains tested. Other bean lines
tested in this experiment also displayed differing disease reactions, depending on the Xap

isolate used in inoculations. Ekpo and Saettler (1976) noted significant variation in the

14

pathogenicity of 15 isolates tested on a set of differential bean lines. Observations from
this experiment indicated that highly used resistant lines, such as the CBB resistant
breeding line GNN #1 Sel. 27, varied from highly resistant to highly susceptible,
depending on the isolate used. Although this line was resistant to the majority of isolates,
it was very susceptible to specific isolates, receiving a rating of 4.0 in response to
inoculation with four different isolates. A 1-4 scale of visual CBB rating was used:

1 = tolerant, few necrotic flecks, 2 = slightly susceptible, <20% of inoculated leaf
necrotic, 3 = moderately susceptible, 20-50% of inoculated leaf necrotic, and 4 = severely
susceptible, >50% of inoculated leaf necrotic. The isolates used in screening were
obtained from various locations in North America, South America, and Africa. Such
examples illustrate the importance of pathogen variability as it pertains to breeding for
genetic resistance. Resistance of a bean line to a specific Xap isolate does not guarantee
that it will be resistant to all isolates that may be present in the area where the line will be
grown. This issue is obviously of great concern to breeders who wish to deve10p resistant
varieties for production in a specific region (Schuster and Coyne, 1975). Ekpo and
Saettler (1976), proposed the use of multiple resistance sources to resolve issues related

to differential pathogenicity of Xap.

Sources of CBB Resistance

The naturally occurring resistance to CBB found in P. vulgaris is limited.
Sources of resistance are limited, with only intermediate levels of resistance. However,
resistance sources present in various other Plraseolus species have been identified and

utilized in bean breeding programs. In addition to quantitative resistance present in P.

15

vulgaris, resistance is also found in the genomes of related species P. acutifolius and P.
coccineus (tepary and scarlet runner beans, respectively). Scarlet runner bean is a part of
the secondary gene pool of common bean, which means that it can be crossed with
common bean without the need for embryo rescue. Tepary bean is part of the tertiary
gene pool of common bean, which means that crosses between the two species require
embryo rescue (Singh, 2001). Breeding lines have been successfully developed through
interspecific hybridization of both P. vulgaris x P. acutz’folius (Singh and Munoz, 1999)
and P. vulgar-is x P. coccineus (Freytag et al., 1982). The development of gene pyramids
consisting of resistance from various backgrounds has been suggested as a strategy for
the development of CBB-resistant varieties (Singh and Munoz, 1999). Gene pyramiding
and interspecific hybridization have not yet led to the development of lines that possess a
level of resistance equal to that found in P. acutifolius. These lines have, however,
shown marked improvement over naturally occurring resistance present in P. vulgaris
(Singh and Munoz, 1999). The first interspecific hybridization between P. vulgaris and
P. acutz’folius reported by Honma (1956), involved Montana No. 5 (the common bean
parent) and Tepary 4 (the P. acutifolius parent) and resulted in a line, Great Northern
Nebraska #1 Sel. 27 (or GNN #1 Sel. 27), that was widely used as a donor of CBB
resistance in breeding programs. Recent evidence, however, has questioned whether an
interspecific hybridization was made in this experiment (Miklas et al., 2003).
lnterspecific crosses with P. coccineus have also yielded CBB resistant breeding lines,
such as XR-235-l-l (Freytag et al., 1982).

Currently, the majority of breeding for CBB resistance has implemented three

major sources of resistance: GNN #1 Sel. 27, XAN 159, and OAC 88-1. The resistance

16

present in GNN #1 Sel. 27 originated from P. vulgaris (Miklas et al., 2003), whereas the
resistance in both XAN 159 and OAC 88-1 was introgressed from P. acutifolius (Thomas
and Waines, 1984; Bai et al., 1997; Singh and Munoz, 1999).

GNN #1 Se]. 27 was selected from a cross between Montana No. 5, a great
northern cultivar (P. vulgaris) and Tepary #4 (P. acutifolius). This cross was made by
Honma (1956), resulting in Great Northern Nebraska #1, however, it was Coyne et al.
(1963), who identified and selected GNN #1 Sel. 27 for CBB resistance breeding
purposes. It was originally assumed that the resistance in CNN #1 Sel. 27 was derived
from the tepary parent, simply because of the large amount of CBB resistant tepary
gerrnplasm (as opposed to the limited amount present in P. vulgaris). Recent evidence
has contradicted these assumptions, confirming that the CBB resistance was donated by
the common bean parent, Montana No. 5 (Miklas et al., 2003).

One example of successful tepary-derived resistance through interspecific
hybridization is the breeding line XAN 159, derived from the tepary source PI 319443.
This line originated from a number of P. vulgaris x P. acutifolius crosses, followed by
embryo rescue (Thomas and Waines, 1984; McElroy, 1985; Jung et. al. 1997). Another
instance of tepary-derived resistance is present in the line OAC 88-1, that carries
resistance obtained through the interspecific hybridization of P. vulgaris and the tepary

line PI 440795.

Anthracnose Resistance
Resistance to anthracnose in common bean is conditioned as a simply inherited,

monogenic trait, with both recessive and dominant resistance genes. Anthracnose has a

17

highly characterized race structure, and known host resistance genes confer resistance in
a race-specific fashion. Additionally, multiple alleles are characterized at three of the
known anthracnose resistance loci (Co-1, Co-3, C0-4). An allele at the C0-4 locus, Co-
42, provides the most desirable, broad-based resistance, conferring resistance to 33 of 34
C. lindemutlziammz races tested by Balardin and Kelly (1998). The ideal strategy,
however, is to combine resistance genes from two gene pools, Andean and Middle
American, in a single genotype. Such a combination is proposed to provide the most
durable resistance (Kelly et al., 1994). The Co-I gene was historically the most widely
utilized anthracnose resistance gene and is of Andean origin. The Co-Iz allele, found in
the differential cultivar Kaboon, isrthe most useful Andean anthracnose resistance gene

for North America (Melotto and Kelly, 2000).

Molecular Markers and Quantitative Trait Loci (QTL) in CBB Resistance

In recent years, the ability to identify molecular markers linked to quantitative
traits has given breeders a useful tool for examining these complex traits. Molecular
markers allow the identification and genetic linkage mapping of quantitative trait loci
(QTL). QTL are regions of the genome that account for a portion of the total phenotypic
variation of a quantitative trait. QTL analysis allows breeders to identify loci associated
with a trait, verify its novelty (based on the location to which the QTL maps in a genetic
linkage map), and determine the contribution of the locus to a phenotypic trait (reported
as an r2 value). The number of genes within a QTL region is unknown, as are their
effects. It is likely that the QTL region identified by a molecular marker contains

multiple genes, which may not all be related to the trait being studied (Castro et al.,

18

2003a). The substantial progress made in molecular marker technology holds
considerable promise in the use of breeding for genetic resistance to CBB. Molecular
markers provide breeders with an opportunity to avoid unreliable direct screening
techniques that are not effective in selection for quantitative traits that are significantly
affected by environmental factors. Several reliable QTL are now available that provide
breeders with a useful tool to use in breeding for CBB resistance, but their efficiency in
new genetic backgrounds requires testing (Singh and Munoz, 1999; Miklas et al., 2000).
Data from QTL studies of CBB resistance in bean has typically agreed with the
conclusions from previous genetic studies. A small number of major-effect QTL, each
explaining a portion of the phenotypic variation has commonly been observed. For
example, Tar’an et al. (1998) examined two molecular markers linked to QTL for CBB
resistance, discovering that in combination, they accounted for 22% of the genetic
variation. A study of resistance to CBB in a cross between BAT93 and Jalo EEP558
revealed the presence of 4 QTL that explained 75% of the variation for CBB resistance.
Because one of these putative QTL accounted for a third of the variation (32%), these
results agree with the majority of previous genetic studies, that CBB resistance in
common bean appears to be controlled by a major effect QTL with a small number of
minor loci also affecting the trait (Nodari et al., 1993). Jung et al. (1997) found the
number of QTL associated with tepary-derived CBB resistance in a common bean
population varied from one to four (explaining 18% to 53% of the phenotypic variation)
depending on the pathogen isolate used in screening and the plant organ used for
phenotypic evaluation. A QTL linked to the RAPD marker BC420900 explained 62% of

the variation for CBB resistance in an experiment by Yu et al. (2000). Such data supports

19

the hypothesis of a single major locus with several contributing minor QTL controlling
the resistance reaction. The presence of four and five markers linked to QTL for leaf and
pod CBB resistance, respectively, was reported by Ariyarathne et al. (1999). The
observation that one locus controlled 44% of the variation for leaf resistance and 41% of
the variation for pod resistance support the theory that CBB resistance is conditioned by
one major locus and a small number of minor loci contribute to that resistance.

The identification of molecular markers linked to QTL provides an opportunity
for breeders to screen for the presence of specific molecular markers. This type of
approach can reduce the number of lines that require direct screening for a quantitative
trait. Widespread screening of breeding materials for the presence or absence of specific
markers, a process known as marker-assisted selection (MAS), has received considerable
attention recently. MAS is the logical result of QTL and marker discovery studies,
however it is not without its share of limitations and challenges. Unfortunately, the
majority of research has focused on the identification of genomic regions associated with
quantitative traits, and less on the application of QTL in breeding programs (Ribaut and
Hoisington, 1998).

Some serious challenges are presented when working with QTL in disease
resistance breeding. The expression of QTL between distinct genetic populations can
vary, limiting their usefirlness in breeding programs (Castro et al., 2003b). Since QTL
are not associated with the complete phenotypic variation of a trait, the resistance
provided by a QTL is only partial. In order to confer high levels of resistance to a
cultivar, more than one QTL may be required. Since quantitative resistance displays

continuous variation and is subject to environmental influences, large-scale field

20

evaluations are needed to confirm the presence and effects of QTL (Castro et al., 2003b).
Additionally, Beavis (1998) suggested that the small population sizes used in QTL
identification studies have caused problems by overestimating the effects of QTL and
underestimating the number of loci involved in a quantitative trait. Although breeding
for quantitative traits has experienced modest success in comparison to breeding for
major gene resistance, it provides breeders with the only alternative for developing

improved cultivars, where monogenic resistance does not exist.

Quantitative Trait Loci (QTL) in CBB Resistance Sources

The characterization of QTL in various CBB resistance sources has been
investigated. One report has given breeders insight regarding the origin of the resistance
QTL carried by the CBB resistant breeding line GNN #1 Sel. 27. It has been proposed
that the resistance found in GNN #1 Sel. 27 originated from P. vulgaris, not P.
acutifolius, as previously considered (Miklas et al., 2003). In field studies, Montana No.
5 (the common bean parent of GNN #1 Sel. 27) appeared to have partial CBB resistance,
similar to GNN #1 Se]. 27 (Miklas et al., 2002). Perhaps the most compelling evidence
of this theory regarding the origin of the CBB resistance in GNN #1 Sel. 27, is the
molecular marker data. The availability of molecular markers linked to CBB resistance
in common bean has resulted in new opportunities for addressing questions regarding the
origins of specific genomic regions. The SCAR marker SAP6820, which is linked to the
major resistance QTL present in GNN #1 Sel. 27, was used for screening, and the marker
present in GNN #1 Sel. 27 can be amplified in the Montana No. 5 cultivar, but not in

Tepary #4. Additionally, cluster analysis using 169 molecular markers did not reveal any

21

relationship between Tepary #4 and GNN #1 Sel. 27. The evidence presented is
insufficient to determine whether an interspecific hybridization ever occurred between
Montana No. 5 and Tepary #4 (Miklas et al., 2003). A re-examination of the F3 data
presented by Honma (1956) seems to suggest that a cross pollination did not occur.

Based on the results of other attempts at similar cross hybridizations, Miklas et al. (2003)
concluded that the probability of obtaining a cultivar with the seed quality of GN #1
without at least one backcross to P. vulgaris is highly unlikely.

It seems apparent, however, that the major-effect QTL linked to SAP6 (on bean
linkage group BIO), which is present in GNN #1 Sel. 27, did originate from the Montana
No. 5 parent. The higher levels of resistance consistently observed in GNN #1 Sel. 27
indicate the presence of additional minor loci that are involved in genetic resistance to
CBB in this population (Miklas et al., 2003). The presence of transgressive segregants
identified from a GNN #1 Sel. 27/Montana No. 5 cross, also suggests the presence of
minor-effect QTL that are not expressed in Montana No. 5 (Miklas et al., 2003). The
nature and origin of these additional resistance loci are yet to be determined.

The bean breeding line XAN 159 is now widely used in breeding programs as a
source of CBB resistance, because it carries two major QTL for CBB resistance, mapped
to linkage groups B6 and B8 of the integrated bean linkage map (Freyre et al., 1998).
Both QTL are linked to molecular markers that can be used for indirect screening. The
sequence-characterized amplified region (SCAR) markers linked to these two QTL are
referred to as SU91 and BC420, respectively (Pedraza et al., 1997; Miklas et al., 2000;
Yu et al., 2000). SCAR markers, which are derived from random amplified polymorphic

DNA (RAPD) markers, are more specific, typically amplifying only one fragment.

22

RAPD markers, which utilize single 10 nucleotide primers, are much less specific. This
results in a larger number of bands amplified, which in turn, makes them more difficult to
score and use for selection. The development of SCAR markers from RAPDs involves
sequencing the fragment amplified by the original RAPD primer, and developing longer
(and thus, more specific) primers. These primers (approximately 24 base pairs in length)
result in the amplification of a single, specific product, scored as the presence or absence
of a single band on an agarose gel (Paran and Michelmore, 1993).

The CBB resistant breeding line OAC 88-1 has also been examined in QTL
studies. The resistance present in this line appears to be conferred by at least two major
QTL, linked to the SCAR markers R7313 and R4865. The R7313 QTL has been mapped
to linkage group B8, while the other QTL remains unmapped (Bai et. al., 1997; Miklas et.
al., 2000). The QTL in OAC 88-1 may be the same QTL in XAN 159. The SCAR
markers SU91 and R7313 are mapped to the same location on bean linkage group B8,
which provides support for the idea that these markers identify the same QTL for CBB
resistance (Tar’an, 1998; Miklas et al., 2000). This example outlines the need for QTL
mapping in determining the novelty of a resistance source. Because QTL identified in
diverse germplasm may actually be the same QTL, determining their novelty based on
their location on a genetic linkage map can help ensure that research efforts are not
duplicated, prior to the initiation of crossing with the resistance source (Kelly and

Vallejo, in press).

23

Marker-Assisted Selection and Gene Pyramiding for CBB Resistance

The opportunity to pyramid QTL (as compared to using a single QTL) for
resistance to CBB offers several distinct advantages to breeders attempting to develop
bean cultivars with genetic resistance. Due to pathogen isolate variability and the
resulting potential isolate specificity of CBB resistance sources, the pyramiding of
multiple resistance QTL would provide a broader resistance base than the use of a single
resistance locus. The potential durability of genetic resistance is also higher when
multiple QTL or genes are pyramided, as it decreases the likelihood of a pathogen
overcoming the resistance (Jung et al., 1999; Castro et al., 2003a; Castro et al., 2003b).
Since the partial nature of the resistance conferred by individual QTL for CBB resistance,
the use of multiple QTL potentially provide higher levels of resistance (Singh et al. 2001;
Castro et al., 2003a). The demonstrated organ specificity of known CBB resistance QTL
illustrates another potential advantage of pyramiding QTL for resistance. Because certain
QTL are known to confer organ specific resistance, the use of multiple QTL would help
ensure that resistance levels are increased in all affected plant organs. The opportunity
currently exists to begin pyramiding various QTL associated with CBB resistance, and
this approach clearly demonstrates advantages over the use of single loci (Jung et al.,
1999; Park et al., 1999; Singh et al., 1999; Miklas et al., 2000).

Molecular markers linked to a trait can be easily used to screen for genetic
determinants of the trait. One promising approach to developing CBB resistant bean
cultivars is to utilize MAS to facilitate the introgression of gene pyramids into
agronomically acceptable lines, followed by the confirmation of resistance by extensive

greenhouse and field disease screening (Singh and Munoz, 1999). Progress has been

24

made in pyramiding resistance from various Phaseolus species. Pinto bean accession G
17341, pinto bean breeding line NY 79-3776-1, and white bean breeding line Wilkinson
2 are the result of combining resistance derived from P. vulgaris, P. coccineus, and P.
acutifolius. These genotypes exhibited improved CBB resistance when compared to that
of the three parental lines GNN #1 sel. 27, XAN 159, and OAC 88-1 (Singh and Munoz,
1999). Another group of resistance sources with pyramided CBB resistance is the VAX
lines, derived from an interspecific hybridization between common bean (the cultivar
ICA Pijao) and tepary bean (accession G 40001). The original cross was conducted by
Mejia-Jimenez et al. (1994) and resulted in the breeding lines VAX 1 and VAX 2. Four
additional VAX lines (VAX 3-VAX 6) were derived from these first two lines using
multiple CBB resistant parents, and have been utilized in a number of breeding programs.
XAN 159 was used in the development of only VAX 4 and VAX 5. The common bean
cultivar ‘Tara’ was used in developing VAX 3 and VAX 6. All six VAX lines have
resistance introgressed from the CBB resistance source P1207262. Due to the use of
additional resistance sources in the VAX 3-VAX 6 lines, their CBB resistance is greater
than the resistance in VAX 1 and VAX 2. The mean ratings for CBB reaction of the
VAX 3 — VAX 6 lines ranged from 1.5 to 2.7, based on the CBB Visual disease rating
described by Schoonhoven and Pastor-Corrales (1987), where: 1 = no visual symptoms,
3 = approximately 2% of the leaf area has lesions, 5 = approximately 5% of the leaf area
has lesions, 7 = approximately 10% of the leaf area has lesions, 9 = more than 25% of the
leaf area has lesions. These mean CBB ratings for VAX 3 — VAX 6 rank among the best

of the available CBB resistant breeding lines (Singh and Munoz, 1999).

25

Use of Marker-Assisted Selection to Improve Phenotypic Selection

As with most plant pathogens, CBB infection most commonly occurs within
specific ranges of temperature and humidity. CBB epidemics generally require
temperatures above 28°C. In addition to this, humidity is often required for CBB
infection to occur (Saettler, 1989). Environmental conditions are not the only factors that
impact the susceptibility of dry beans to Xap. The growth stage of the plant also plays a
large role in plant-pathogen interactions. As a general rule, plants that are late maturing
due to either genetic or environmental conditions are more likely to escape infection.
This is complicated by the differential response of various bean plant organs to CBB,
because of independent genetic control of resistance in the different plant organs. Plant
organs are not all susceptible at the same developmental stage in the growing season.
This differential reaction can have an enormous impact on CBB incidence in different
years, as well as attempts at breeding resistant cultivars (Coyne and Schuster, 1974).
Although an optimal breeding strategy would include the development of plants with
both leaf and pod/seed resistance, Valladares-Sanchez et al. (1979) suggested that
pathogen spread could be significantly reduced with sufficient pod resistance, since it
would reduce the primary source of inoculum.

The variability of the bean plant’s reaction to CBB makes resistant cultivar
development using visual selection alone impractical, because many factors can result in
susceptible genotypes appearing resistant in the field. The use of MAS, alone, or in
conjunction with conventional screening helps to overcome the difficulties associated
with environmental conditions (Yu et al., 2000; Tar’an et al., 2001). MAS may hold

greater promise for disease resistance than for other traits, because it helps alleviate the

26

difficulties associated with experimental systems involving multiple living organisms. It
allows breeders to screen a large number of genotypes in absence of the pathogen,
determining their resistance potential by using molecular markers linked to disease
resistance loci (Kelly and Miklas, 1999). The preliminary screening of genotypes in the
absence of the pathogen is useful if normal selection for resistance is complicated by
excessive variability in the pathogen, in experimental conditions, or if there is a need for
conducting the research in a location where the pathogen (or a particular strain) is not
naturally present. In the latter case, pathogen escape would be a concern in the use of
inoculations (Kelly and Miklas, 1999).

There are some limitations to the usefulness of MAS, despite its obvious
advantages. One of the major complications is the possible population specificity of
some molecular markers. When attempting to use markers in different populations from
those in which the marker was developed, the original linkage between the marker and
the trait of interest may be lost (Bai et al., 1997). A RAPD marker (OA141100) linked to a
rust resistance gene in beans, was only useful in the Middle American gene pool and
provided no value in breeding Andean beans (Miklas et al., 1993).

In conclusion, the strategy with the most potential for identifying CBB resistant
genotypes seems to be through the combination of molecular marker-assisted selection
and traditional phenotypic selection in the field and greenhouse. The use of both direct
and indirect selection helps to avoid some of the problems associated with using either

method of selection alone.

27

GENERAL OBJECTIVES

Common bacterial blight can be devastating for both growers and seed producers
of dry bean. The adequate control of this pathogen requires the deployment of bean
cultivars with high levels of resistance. Currently, none of the cultivars available to
Michigan dry bean growers have sufficiently high levels of CBB resistance. To identify
and improve potentially resistant genotypes, an efficient method of selection must be
developed. This is complicated by pathogen variability and environmental effects.
Marker-assisted selection provides an opportunity to overcome some of the limitations of
traditional phenotypic selection for resistance to CBB. The objectives of this research
were to: 1) examine the potential for using previously developed molecular markers
linked to CBB resistance in a marker-assisted selection strategy, 2) determine the effects
on CBB resistance of previously discovered QTL in genetic backgrounds suited for
Michigan, 3) determine the potential for combining multiple sources of CBB resistance,
and 4) combine CBB resistance with specific genes conferring anthracnose resistance in
new bean germplasm.

Chapter 1 addresses the potential of using a previously developed molecular
marker originating from another bean breeding program to select for a QTL linked to
CBB resistance in a different genetic background than the background in which the QTL
was discovered. Chapter 2 examines the use of MAS to pyramid diverse CBB resistance
sources with anthracnose resistance sources to develop cultivars with resistance to both

seed-bome common bean pathogens.

28

REFERENCES

Adams, M.W., J .D. Kelly, and AW. Saettler. 1988. A gene for resistance to common

blight (Xanthomonas campestris pv. phaseoli). Ann. Rept. Bean Improv. Coop.
31:73-74.

Aggour, A.R., D.P. Coyne, AK. Vidaver, and KM. Eskridge. 1989. Transmission of the
common blight pathogen in bean seed. J. Amer. Soc. Hort. Sci. 114:1002-1008.

Anderson, A.L., L.O. Copeland, and AW. Saettler. 1970. Grow blight-free field beans.
Michigan State University Cooperative Extension Service Extension Bulletin 680.
Michigan State University, East Lansing, Michigan.

Anderson, F.N., R.L. Forster, A.F. Hagen, E.D. Kerr, D.S. Nuland, H.F. Schwartz, R.L.
Stoltz, and RC. Wilson, Jr. 1989. Recognition and management of dry bean
production problems. North Central Regional Extension Publication 198.

Andrus, CF. 1948. A method of testing beans for resistance to bacterial blight.
Phytopathology 38:757-759.

Ariyarathne, H.M., D.P. Coyne, G. Jung, P.W. Skroch, A.K. Vidaver, J .R. Steadman,
P.N. Miklas, and M.J. Bassett. 1999. Molecular mapping of disease resistance
genes for halo blight, common bacterial blight, and bean common mosaic virus in
a segregating population of common bean. J. Amer. Soc. Hort. Sci. 124:654-662.

Amaud-Santana, E., D.P. Coyne, K.M. Eskridge, and AK. Vidaver. 1994. Inheritance;
low correlations of leaf, pod, and seed reactions to common blight disease in

common beans; and implications for selection. J. Amer. Soc. Hort. Sci. 119:116-
121.

Bai, Y.H., T.E. Michaels, and KP. Pauls. 1997. Identification of RAPD markers linked to
common bacterial blight resistance genes in Phaseolus vulgaris L. Genome
40:544-551.

Balardin, RS, and Kelly, JD. 1998. Interaction between Colletotrichum lindemuthianum
races and gene pool diversity in Phaseolus vulgaris. J. Amer. Soc. Hort. Sci.

123:1038-1047.

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

Cafati, CR. and AW. Saettler. 1980a. Effect of host on multiplication and distribution of
bean common blight bacteria. Phytopathology 70:675-679.

Cafati, CR, and AW. Saettler. 1980b. Role of nonhost species as alternate inoculum
sources of Xanthomonas phaseoli. Plant Disease 64:194-196.

29

Cafati, CR. and AW. Saettler. 19800. Transmission of Xanthomonas phaseoli in seed of
resistant and susceptible Phaseolus genotypes. Phytopathology 70:638-640.

Castro, A.J., X. Chen, A. Corey, T. Filichkina, P.M. Hayes, C. Mundt, K. Richardson, S.
Sandoval-Islas, and H. Vivar. 2003a. Pyramiding and validation of quantitative
trait locus (QTL) alleles determining resistance to barley stripe rust: effects on
adult plant resistance. Crop Sci. 43:2234-2239.

Castro, A.J., X. Chen, P.M. Hayes, and M. Johnston. 2003b. Pyramiding quantitative trait
locus (QTL) alleles determining resistance to barley stripe rust: effects on
resistance at the seedling stage. Crop Sci. 43:651-659.

Claflin, L.E., D.L. Stuteville, and D.V. Arrnbrust. 1973. Wind-blown soil in the
epidemiology of bacterial leaf spot of alfalfa and common blight of bean.
Phytopathology 63:1417-1419.

Claflin, L.E., A.K. Vidaver, and M. Sasser. 1987. Mxp, a semi-selective medium for
Xanthomonas campestris pv. phaseoli. Phytopathology 77:730-734.

Corey, RR, and MP. Starr. 1957. Colony types of Xanthomonas phaseolz'. J. Bact.
74:137-140.

Coyne, D.P., M.L. Schuster, and S. Al-Yasiri. 1963. Reaction studies of bean species and
varieties to common blight and bacterial wilt. Plant Dis. Reptr. 47:534-53 7.

Coyne, D.P., M.L. Schuster, and L. Harris. 1965. Inheritance, heritability, and response to
selection for common blight (Xanthomonas phaseoli) tolerance in Phaseolus
vulgaris field bean crosses. Proc. Amer. Soc. Hort. Sci. 86:373-379.

Coyne, DP, and ML. Schuster. 1974a. Differential reaction of pods and foliage of beans
(Phaseolus vulgaris) to Xanthomonas phaseoli. Plant Dis. Reptr. 58:278-282.

Coyne, DP, and ML. Schuster. 1974b. Inheritance and linkage relations of reaction to
Xanthomonas phaseoli (E.F. Smith) Dowson (common blight), stage of plant
development and plant habit in Phaseolus vulgaris L. Euphytica 23: 195-204.

Drij fhout, E., and W.J. Blok. 1987. Inheritance of resistance to Xanthomonas campestris
pv. phaseoli in tepary bean (Phaseolus acutifolius). Euphytica 36:803-808.

Dudley, J .W. 1993. Molecular markers in plant improvement: Manipulation of genes
affecting quantitative traits. Crop Sci. 33:660-668.

Ekpo, E.J.A., and AW. Saettler. 1976. Pathogenic variation in Xanthomonas phaseoli
and X. phaseoli var. fuscans. Plant Dis. Reptr. 60:80-83

30

F ehr, W.R. 1987. Single-Seed Descent Method. Pages. 319-327. In: Principles of Cultivar
Development, Vol. 1. W. R. F ehr (ed.). Macmillan Publishing Company, New
York.

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.

Freytag, G.F., M.J. Bassett, and M. Zapata. 1982. Registration of XR-235-1-1 bean
germplasm. Crop Sci. 22:1268-1269.

Goss, R.W. 1939. The relation of temperature to common and halo blight of beans.
Phytopathology 30:258-264.

Honma, S. 1956. A bean interspecific hybrid. J. Hered. 47:217-220.

Jung, G., P.W. Skroch, D.P. Coyne, J. Nienhuis, E. Amaud-Santana, H.M. Ariyarathne,
S.M. Kaeppler, and M.J. 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.

Jung, G., P.W. Skroch, J. Nienhuis, D.P. Coyne, E. Arnaud-Santana, H.M. Ariyarathne,
and J .M. Marita. 1999. Confirmation of QTL associated with common bacterial
blight resistance in four different genetic backgrounds in common bean. Crop Sci.
39:1448-1455.

Kaiser, W.J., and NC. Vakili. 1978. Insect transmission of pathogenic xanthomonads to
bean and cowpea in Puerto Rico. Phytopathology 68:1057-1063.

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

Kelly, JD. and LO. Copeland. 1996. Status of common bacterial blight problems in
Michigan dry beans. p. 173-177. ler. Taller Interacional sobre bacteriosis somun
del fiijol. Oficina de Coordinacion Regional de Profiijol, Cesda, San Cristobal,
Republica Domincana.

Kelly, J .D., and RN. Miklas. 1999. Marker-assisted selection. Pages 93-123. In:

Common Bean Improvement in the Twenty-First Century, 1 ed. S. P. Singh (ed.).
Kluwer Academic Publishers, Dordrecht, The Netherlands.

31

Kelly, J .D., and VA. Vallejo. 2004. QTL analysis of multigenic disease resistance in
plant breeding. In: Multigenic and Induced Systemic Resistance in Plants. S.
Tuzun and E. Bent (eds.) In press.

Leach, J .G., V.G. Lilly, H.A. Wilson, and MR. Purvis, Jr. 1957. Bacterial
polysaccharides: the nature and function of the exudate produced by
Xanthomonas phaseoli. Phytopathology 47:59-1 12.

McElroy, J .B. 1985. Breeding dry beans, Phaseolus vulgaris L., for common bacterial
blight resistance derived from Phaseolus acutifolius A. Gray. Ph.D. Cornell
Univ., Ithaca, N.Y.

Mejia-Jimenez, A., C. Munoz, H.J. Jacobsen, W.M. Roca, and SP. Singh. 1994.
Interspecific hybridization between common and tepary beans: increased hybrid
embryo growth, fertility, and efficiency of hybridization through recurrent and
congruity backcrossing. Theor. Appl. Genet. 88:324-331.

Melotto, M., L. Afanador, and J .D. Kelly. 1996. Development of a SCAR marker linked
to the I gene in common bean. Genome. 39: 1216-1219.

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

Miklas, P.N., J .R. Stavely, and J .D. Kelly. 1993. Identification and potential use of a
molecular marker for rust resistance in common bean. Theor. Appl. Genet.
85:745-749.

Miklas, P.N., J .R. Smith, R. Riley, K.F. Grafton, S.P. Singh, G. Jung, and DP. Coyne.
2000. Marker-assisted breeding for pyramided resistance to common bacterial
blight in common bean. Ann. Rept. Bean Improv. Coop. 43:39-40.

Miklas, P.N., D.P. Coyne, K.F. Graflon, N. Mutulu, J. Reiser, and SP. Singh. 2002. A
major QTL for common bacterial blight resistance in GN#1 Sel. 27 derives from
the great northern landrace Montana No. 5, not tepary bean. Ann. Rept. Bean
Improv. Coop. 45:52-53.

Miklas, P.N., D.P. Coyne, K.F. Grafton, N. Mutulu, J. Reiser, D.T. Lindgren, S.P. Singh.
2003. A major QTL for common bacterial blight resistance derives from the
common bean great northern landrace cultivar Montana No. 5. Euphytica
131:137-146.

Musaana, S.M., O.D. Mwandemele, and J .A. Kapuya. 1993. The number of genes

controlling resistance in beans to common blight. Ann. Rep. Bean Improv. Coop.
36: 156-157.

32

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

Paran, 1., and R.W. Michelmore. 1993. Development of reliable PCR-based markers
linked to downy mildew resistance genes in lettuce. Theor. Appl. Genet. 85:985-
993.

Park, S.O., D.P. Coyne, N. Mutlu, G. Jung, J .R. Steadman. 1999. Confirmation of
molecular markers and flower color associated with QTL for resistance to
common bacterial blight in common beans. J. Amer. Soc. Hort. Sci. 124:519-526.

Paterson A.H., S. Damon, J .D. Hewitt, D. Zamir, H.D. Rabinowitch, S.E. Lincoln, E.S.
Lander, and SD. Tanksley. 1991. Mendelian factors underlying quantitative traits
in tomato: comparison across species, generations, and environments. Genetics
127:181-197.

Pedraza, F ., G. Gallego, S. Beebe, J. Tohme. 1997. Marcadores SCAR y RAPD para la
resistencia a la bacteriosis comun (CBB). Pages 130-134. In: Taller de
Mejoramiento de F rijol para el Si glo XXI: Bases para Una Estrategia para
America Latina. S.P. Singh and O. Voysest (ed.). CIAT, Cali, Colombia.

Ribaut, J .M., and D. Hoisington. 1998. Marker-assisted selection: new tools and
strategies. Trends in Plant Sciences 3:236-239.

Saettler, AW. 1971. Chemical sprays for the control of common-fuscous bacterial blight
in Navy (pea) beans. Ann. Rept. Bean Improv. Coop. 14:54-55.

Saettler, AW. 1973. Chemical control of bean bacterial diseases. Ann. Rept. Bean
Improv. Coop. 16:27-28.

Saettler, AW. 1989. Common bacterial blight. Pages 261-283 In: Bean Production
Problems in the Tropics, 2nd ed. H. F. Schwartz, and M. A. Pastor-Corrales
(eds). CIAT (Centro Intemacional de Agricultura Tropical), Cali, Colombia.

Saettler, A.W, C.R. Cafati, and D.M. Weller. 1986. Nonoverwintering of Xanthomonas
bean blight bacteria in Michigan. Plant Disease 70:285-287.

Schoonhoven, A.v., and MA. Pastor-Corrales. 1987. Measurement of damage caused by
diseases. Pages 16-44. In: Standard System for the Evaluation of Bean
Germplasm. A. v. Schoonhoven and M. A. Pastor-Corrales (eds.). CIAT (Centro

Intemacional de Agricultura Tropical), Cali, Colombia.

Schuster, ML. 1955. A method for testing resistance of beans to bacterial blights.
Phytopathology 45:519-520.

33

Schuster, ML. 1970. Survival of bacterial pathogens of beans. Ann. Rept. Bean Improv.
Coop. 13:68-70.

Schuster, ML. and DP. Coyne. 1971. New virulent strains of Xanthomonas phaseoli.
Plant Dis. Reptr. 55:505-506.

Schuster, ML, and DP. Coyne. 1975. Genetic variation in bean bacterial pathogens.
Euphytica 24: 143-147.

Scott, ME. and TE. Michaels. 1988. Inheritance of resistance to common bacterial blight
in common bean. Ann. Rept. Bean. Improv. Coop. 31 :72.

Singh, SP. 2001. Broadening the genetic base of common bean cultivars: a review. Crop
Sci. 41 :1659-1675.

Singh, SP, and CG. Munoz. 1999. Resistance to common bacterial blight among
Phaseolus species and common bean improvement. Crop Sci. 39:80-89.

Singh, S., J .S. Sidhu, N. Huang, Y. Vikal, Z. Li, D.S. Brar, H.S. Dhaliwal, and GS.
Khush. 2001. Pyramiding three bacterial blight resistance genes (xa5, MB, and
Xa21) using marker-assisted selection into indica rice cultivar PR106. Theor.
Appl. Genet. 102:1011-1015.

Steadman, J .R., C.R. Maier, H.F. Schwartz, and ED. Kerr. 1975. Pollution of surface
irrigation waters by plant pathogenic organisms. Water Resour. Bull. 11:796-804.

Tar'an, B., T.E. Michaels, and KP. Pauls. 1998. Stability of the association of molecular
markers with common bacterial blight resistance in common bean (Phaseolus
vulgaris L.). Plant Breed. 117:553-558.

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

Thomas, C.V., and J .G. Waines. 1984. Fertile backcross and allotetraploid plants from
crosses between tepary beans and common beans. J. Hered. 75:93-98.

Urrea, C.A., P.N. Miklas, and J .8. Beaver. 1999. Inheritance of resistance to common
bacterial blight in four tepary bean lines. J. Amer. Soc. Hort. Sci. 124:24-27.

Vakili, N.G., W.J. Kaiser, J .E. Perez, and A. Cortes-Monllor. 1975. Bacterial blight of
beans caused by two Xanthomonas pathogenic types from Puerto Rico.

Phytopathology 65 :40 1 -403.

Valladares-Sanchez, N.E., D.P. Coyne, and ML. Schuster. 1979. Differential reaction of
leaves and pods of Phaseolus germplasm to strains of Xanthomonas phaseoli and

34

transgressive segregation for tolerance from crosses of susceptible germplasm. J.
Amer. Soc. Hort. Sci. 104:648-654.

Wallen, V.R., and HR. Jackson. 1975. Model for yield loss determination of bacterial
blight of field beans utilizing aerial infrared photography combined with field plot
studies. Phytopathology 65:942-948.

Weller, D.M., and AW. Saettler. 1980. Evaluation of seedbome Xanthomonas phaseoli
and X. phaseolz’ var. fuscans as primary inocula in bean blights. Phytopathology
70: 148-152.

Welsh, M.M. and KP. Grafton. 2001. Resistance to common bacterial blight of bean
introgressed from Phaseolus coccineus. HortScience 36:750-751.

Wilson, H.A., V.G. Lilly, and JG. Leach. 1965. Bacterial polysaccharides, IV: longevity
of Xanthomonas phaseoli and Serratia marcescens in bacterial exudates.
Phytopathology 55:1135-1138.

Yoshii, K., G.E. Galvez, and A. Alvarez. 1976. Estimation of yield losses in beans caused
by common blight. Proc. Amer. Phytopath. Soc. 32298-299.

Young, R.A., M. Melotto, R.O. Nodari, and J .D. Kelly. 1998. Marker-assisted dissection
of the oligogenic anthracnose resistance in the common bean cultivar, 'G2333'.
Theor. Appl. Genet. 96:87-94.

Yu, K., S.J. Park, and V. Poysa. 2000. Marker-assisted selection of common beans for
resistance to common bacterial blight: efficacy and economics. Plant Breed.
119:411-415.

Zaumeyer, W.J., and HR. Thomas. 1957. Bacterial diseases of major importance. Pages
65-88. In: A monographic study of bean diseases and methods for their control.
W. J. Zaumeyer and H. R. Thomas (eds.).United States Department of Agriculture
(USDA), Washington, DC.

35

CHAPTER ONE

THE USE OF MARKER-ASSISTED SELECTION TO ENHANCE THE
LEVEL OF RESISTANCE TO COMMON BACTERIAL BLIGHT IN
COMMON BEAN LINES

ABSTRACT

Common bacterial blight (CBB) is a destructive disease of common bean
(Phaseolus vulgaris L.), caused by the bacterial pathogen Xanthomonas axonopodis pv.
phaseoli. Previous attempts to develop highly resistant cultivars have not succeeded due
to complications including: pathogen variability, complexity of host genetic resistance,
and lack of reproducibility in direct screening with the pathogen. In this study, the
potential for utilizing marker-assisted selection (MAS) was examined, to determine if this
approach could be used to overcome the limitations of phenotypic CBB evaluations.

Field studies were established in East Lansing and Saginaw to compare the CBB
resistance of bean genotypes with or without the sequence characterized amplified
polymorphism (SCAR) marker SU91 (linked to a QTL for CBB resistance on bean
linkage group B8). Presence of SU91 was correlated with enhanced CBB leaf resistance
at both locations (p<0.0001 and p<0.0001, respectively). Presence of SU91 was
correlated with enhanced CBB pod resistance only at the Saginaw location (p<0.0001).

The presence of negative correlations between SU91 and important agronomic
traits was also examined. Presence of SU91 was not correlated with late maturity and
was correlated with reduced yields in the East Lansing experiment (r = 0.22 and r = -

0.20*, respectively). Despite this negative association between the SU91 marker and

36

seed yield, the selection of early-maturing high-yielding, CBB resistant genotypes was
achieved. Presence of SU91 was not correlated with late maturity or reduced yields in
the Saginaw experiment (p= 0.24 and p= 0.08, respectively).

These results indicate that SU91 is an effective molecular marker for the selection
of bean lines with enhanced CBB resistance from Michigan’s dry bean breeding and
genetics program. These findings are encouraging, because multiple reports suggest
population specificity and linkage drag problems are associated with tepary-derived CBB

I'CSISIEIIICC SOUTCCS 111 common bean.

37

INTRODUCTION

Common bacterial blight (CBB), caused by the bacterial pathogen Xanthomonas
axonopodis pv. plzaseoli (Smith) Dye (Xap) is a destructive disease of common bean
(Phaseolus vulgaris). CBB infections generally produce negative impacts on various
agronomic traits, such as: lodging, maturity, pods per plant and grain yield (Tar’an et al.,
2001). In severe cases, CBB may reduce bean yields, with losses potentially exceeding
60% (Zaumeyer and Thomas, 1957). Perhaps the most significant consequence of CBB
epidemics has been the impact on the bean seed production industry. CBB infection is
especially problematic for seed producers because of the seed borne nature of the disease
(Kelly and Copeland, 1996). Seed producers contend with potential yield losses due to
CBB and also the rejection of certified seed beans. Certified seed fields are inspected for
several criteria, one of which is the absence of CBB symptoms. Since Xap is able to
survive in the seed, transmission to seedlings is common (Weller and Saettler, 1980).
CBB infection in certified bean seed is not permissible. In years with serious CBB
epidemics, the rejection rate in Michigan has approached 50% (Kelly and Copeland,
1996).

Due to favorable climatic conditions for CBB development in Michigan, there has
been a dramatic decrease of in-state production of certified seed. From 1974 to 1994, the
area devoted to certified navy bean seed production in Michigan decreased from 5,000
hectares to 400 hectares. This reduction, a 92% decrease, was primarily due to CBB.
Interestingly, kidney bean seed production in Michigan has increased during this time

period. Isolation of kidney bean seed production from commercial fields, as well as the

38

use of moderately resistant cultivars (such as ‘Montcalm’) has played a significant role in
this growth (Kelly and Copeland, 1996).

Many cultural and chemical controls have been used to protect bean crops from
CBB, none of which have proven successful (Zaumeyer and Thomas, 1957; Saettler,
1971). The use of genetic resistance would be the most desirable method of disease
prevention. Previous attempts to develop bean cultivars with CBB resistance have not
been successful (Singh and Munoz, 1999).

Many factors account for the inability to breed resistant bean cultivars. The most
notable factor is due to the nature of the host resistance. Resistance to CBB in common
bean is a quantitatively controlled trait, which exhibits low to moderate heritability. The
inability to identify discrete genetic loci conditioning CBB resistance in beans has been a
limiting factor in attempts to develop resistant cultivars. The expression of quantitative
traits, such as CBB resistance, is typically complicated by environmental interactions that
make it difficult to accurately identify genotypes with CBB resistance (Yu et al., 2000).
Another major complication of traditional approaches to breeding for CBB resistance has
been the inaccuracy of using Xap in direct screening. Problems with reproducibility in
direct screening with the pathogen are partially due to fluctuations in environmental
conditions (Tar’an et al., 2001). There are two major reasons for this: 1) sensitivity of
Xap to temperature and moisture, and 2) variability in virulence among isolates between
experiments; which combine to adversely impact the successful selection of resistant
genotypes (Goss, 1939). The variability of Xap is relatively uncharacterized, and because
the usefulness of resistance sources may be strain specific, pathogen variability presents a

problem for resistance breeding (Schuster and C oyne, 1971; Schuster and Coyne, 1975;

39

Ekpo and Saettler, 1976). A genotype identified as resistant in a particular experiment
may not exhibit the same resistance level in a different experiment using a different Xap
isolate. An additional constraint to breeding for CBB resistance in common bean is the
association of resistance with undesirable agronomic traits. Linkage of a quantitative
trait and undesirable traits, or linkage drag, is a common complication in breeding
quantitative traits (Dudley, 1993). Beebe (1989) reported linkage of CBB resistance to
both negative seed traits (brilliance and color instability) and low bean yield. Linkage
between CBB resistance and late maturity was also reported by Valladares-Sanchez et al.
(1979)

The recent introduction of molecular marker technology has created additional
opportunities for disease resistance breeding strategies, and the potential to overcome
these challenges is now available (Yu et al., 2000). The use of MAS has the potential to
overcome the major limitations of traditional phenotypic selection for CBB resistance.
Molecular markers can be used to identify specific genomic regions linked to a
quantitative trait, and provide the opportunity to screen for the presence of quantitative
trait loci (QTL), which are genetic loci associated with the quantitative trait-of-interest
(Lande and Thompson, 1990). MAS also resolves problems associated with
reproducibility of direct screening, due to environmental effects. Plants may be selected
in absence of the pathogen, which alleviates the concerns of working with both biological
organisms, although direct inoculations may be required to confirm results of MAS
(Kelly and Miklas, 1999).

Despite the obvious advantages over direct screening, MAS has certain

limitations. The issues of pathogen genetic variability and potential specificity of

40

bacterial isolates by resistance loci are not resolved by using MAS. Ekpo and Saettler
(1976) reported that the widely used CBB resistance breeding line GNN #1 Sel. 27
exhibited a highly susceptible reaction when screened with specific isolates. Verification
of resistance provided by specific resistance sources to local Xap isolates would help
ensure that a desired resistance source is useful for that particular region. Another
limitation to the use of MAS in breeding programs is the expression of QTL in different
genetic backgrounds. The usefulness of a molecular marker in a genetic background that
is different from the background in which the QTL was identified is not guaranteed. The
performance of QTL linked to the marker must be examined to ensure that it confers the
desired phenotype (Dudley, 1993; Yu et al. 2000). Prior to the widespread adoption of
MAS for disease resistance, usefulness of markers developed by other research programs
in successfully predicting resistance must be verified. Yu et al. (2000) reported a
successful attempt to use a molecular marker linked to CBB resistance in common bean
population differing genetically from that in which it was developed. Jung et al. (1999)
examined a SCAR marker linked to a different CBB resistance QTL, and determined that
it was associated with resistance in all four of the genetic populations tested. In contrast,
Paterson et al. (1991 ), determined that only four of 29 QTL for fruit size and quality traits
of tomato were associated with the same trait in three environments, while 10 QTL were
stable across two environments.

The practical application of MAS in selection for disease resistance, including
CBB, has been examined, with notable successes and failures (Yu et al., 2000). Research
conducted by Yu et al. (2000) determined that selection based on SCAR marker BC420

was 92.5% as effective as direct selection in identifying resistant bean genotypes, and

41

was also more cost-effective. Various CBB resistance sources are available for use in
bean breeding programs, and robust molecular markers have been developed to screen for
CBB resistance (Singh and Munoz, 1999).

The SCAR marker SU91 is linked to a QTL for CBB resistance on common bean
linkage group B8 (Pedraza et al., 1997). This QTL originated from tepary bean
(Phaseolus acutifolius) and was introgressed into common bean through an interspecific
hybridization. The original resistance source from which SU91 was characterized is
XAN 159, a breeding line that has been used extensively in breeding for CBB resistance.
SU91, which amplifies a single 700 bp fragment, was derived from the RAPD marker U9
(Pedraza et al., 1997).

The SCAR markers SAP6 and BC409 are both linked to the same QTL for CBB
resistance on common bean linkage group B10 (Miklas et al., 2000; Jung et al., 1999).
The original source from which this QTL was identified is the CBB resistant breeding
line GNN #1 Sel. 27. SAP6 exhibits a tighter linkage with the QTL than BC409 (Miklas
et al., 2000). SAP6 and BC409 were derived from the RAPD fragments AP6 and BC409,
and amplify single fragments of 820 and 1250 bp, respectively (Miklas et a1, 2000; Jung
et al., 1999).

The primary objective of this research was to develop black and navy bean lines
with enhanced levels of CBB resistance for production in Michigan, using the SCAR
marker SU91 in a marker-assisted selection program as a complement to traditional direct
screening using local Xap isolates. A second objective was to evaluate the potential for

using SU91 in a genetic background that is different from the original one in which it was

42

identified and evaluate the association between CBB resistance linked to the SU91

marker and important agronomic traits in common bean.

43

MATERIALS AND METHODS

Parental Material and Population Development. The resistant parents (99L91-
45 and 99L91-47) initially used for the development of CBB resistant populations in this
study were developed from VAX 5, a breeding line developed at CIAT (Intemational
Center for Tropical Agriculture), from an interspecific hybridization between P. vulgaris
and P. acutifolius (Singh and Munoz, 1999). VAX 5 has CBB resistance derived from
both XAN 159 and GNN #1 Sel 27, two of the most widely used resistance sources.
VAX 5 is a small-seeded, type II (indeterminate upright), black bean with high levels of
CBB resistance (Singh and Munoz, 1999). The resistance found in VAX 5 (as well as the
other VAX lines) is lower than the level of resistance found in P. acutifolius, however, it
is superior to the resistance found in most resistant P. vulgaris accessions.

The MSU black bean breeding line B98311 was crossed with VAX 5 to develop
two breeding lines (99L91-45 and 99L9l-47). These breeding lines were crossed with a
commercial cultivar (‘Phantom’) and elite breeding lines (N99219, B98306, and B00136)
to combine disease resistance with favorable agronomic characteristics (Frahm, 2002). A
summary of the crosses made in this study is included in Figure 1.1.

In total, eight F 2 populations were selected for advancement via the single seed
descent (SSD) breeding method (Johnson and Bernard, 1962). One pod was harvested
from each F2 individual within each population from the field in the fall of 2001. The F3
plants were grown in the greenhouse during the fall/winter 2001-02, in 6-inch clay pots
and sterile potting soil. Two seeds were planted for each genotype, and thinned to one
seedling/ genotype afier emergence. Upon maturity, l pod from each plant was again

harvested for SSD. Advancement of the F4 generation was conducted in the same

44

manner as the F3 generation during the spring of 2002. Tissue was collected from young
trifoliate leaves of F3 plants, for use in DNA extraction (as described below). All
individuals were screened for presence of the SCAR marker SU91. The F5 populations
were planted on June 10, 2002 at the Saginaw Valley Bean and Beet Farm in one-row
plots. The plots were space-planted (ranging from 2-25 seeds planted per plot, with
variable spacing) and fertilized with 200 lbs of 33-1 1-0 at planting. These five F5
populations consisted of a total of 454 F5 lines that were evaluated based on evaluated
agronomic traits (e. g. time to flowering, pod set, plant architecture). These lines were
compared to parental lines (99L91-45, 99L91-47, 800136, B98306, N99219, and
Phantom) and check lines (HR45, Jaguar, N97774, VAX 3, VAX 5, VAX 6, Vista, and
XAN 159). A select group of 38 F5 lines were chosen and harvested for further study

(for flowering data on these selections, see Appendix A3).

Molecular Marker-Assisted Selection. The SCAR markers used in selection for CBB
resistance were easily scored as the presence or absence of a single band on an agarose
gel. The PCR protocol used was described by Melotto et. al. (1996), except Invitrogen

T aq polymerase was used (Invitrogen, Carlsbad, CA). The SCAR marker SU91, linked
to a QTL for CBB resistance on bean linkage group B8, was used in screening. A SCAR
marker (SAP6) associated with a QTL for CBB resistance on bean linkage group 810
was used to screen parental material. This marker was present in the susceptible parents,
so it was not used in screening the F3 populations. A SCAR marker (BC409), linked to
the same CBB resistance QTL as SAP6 (on bean linkage group BIO) was used to screen

the parental materials. This marker was present in the susceptible parents, and was not

45

used to screen the F3 populations. The SCAR marker BC420, which is linked to a CBB

resistance QTL on bean linkage group B6, was not present in any parental materials, and
was not used to screen the F3 populations. All reactions were completed with a final step
of 5 minutes at 72°C, and samples remained at 4°C until their preparation for agarose gel

electrophoresis. A summary of PCR profiles for this research is presented in Table 1.1.

46

Figure 1.1 Order of crosses made in the development of populations with increased
CBB resistance. A) The resistant parent VAX 5 was crossed with the MSU elite black
bean breeding line B98311. B) Two of the resulting lines that were verified to cm the
SCAR marker SU91 were used in crosses with ‘Phantom’ (black bean cultivar), N99219
(navy bean breeding line), B98306 (black bean breeding line), and B00136 (black bean
breeding line). Crosses were performed in both directions, using each parent as pollen
parent and female parent in separate crosses. The 99L91-45 and 99L91-47 parents were

used in crosses as F35 plants.

A) B98311 x VAX 5

B) Phantom x 99L91-45 99L91-47 x Phantom
N99219 x x N99219
B98306 x x B98306
B00136 x x B00136

47

Table 1.1 A summary of PCR profiles used in screening for the SCAR markers SU91,
SAP6, BC409, BC420.

SCAR PROFILE
SU91 94°C/10", 58°C/40", 72°C/2' 34 cycles; 72°C/5', 4°C soak

 

 

SAP6 94°C/10", 55°C/40", 72°C/2' 34 cycles; 72°C/5', 4°C soak

 

BC409 94°C/10", 70°C/ 1', 72°C/2' 34 cycles; 72°C/5', 4°C soak

 

 

BC420 94°C/30", 50°C/30" , 72°C/1' 35 cycles; 72°C/5', 4°C soak

 

 

 

48

Multiplex PCR was used in some instances to screen genotypes for SU91, BC420,
and SAP6. For this multiplex reaction, the SAP6 profile was used, which has an
annealing temperature intermediate to that of SU91 and BC420. For each reaction, the
components were present in the same concentrations suggested by the aforementioned
protocol (Melotto et. al., 1996), with the following exceptions: 1.5 pl (10 ng/ pl) of each
primer, 16.35 pl H20.

PCR results were analyzed using a 1.4% agarose gel stained with ethidium
bromide (0.02 p g/mL). Bands present on the gel were compared by size to a 100 bp
ladder (Invitrogen, Carlsbad, CA). When multiplex PCR was utilized, SU91 (700 bp),
BC420 (900 bp), and SAP6 (820 bp) were distinguishable by size using ultra-violet light

fluorescence.

Collection and Culture of Bacterial Isolates. Diseased plant tissue was collected from
the field for the purpose of isolating Xap for use in field and greenhouse inoculations.
Twelve samples of infected bean leaves were collected from diverse genotypes, locations,
and market classes at the Saginaw Valley Bean and Beet Farm. An additional 12 samples
of infected bean leaves were collected at the Montcalm Research Farm.

The protocol used for bacterial isolation was as described by Claflin et. a1. (1987),
with some minor modifications. Approximately 0.5 g sample of tissue removed from the
margin of chlorotic regions was ground using an autoclaved mortar and pestle. The
extraction buffer was autoclaved prior to use in grinding. A 50 pl aliquot of each sample
was plated on the semi—selective media MXP (Claflin et. al., 1987). Petri plates were

immediately covered with aluminum-foil to prevent photodegredation of antibiotics in the

49

MXP media. The plates were incubated at 28°C for approximately 48 hours, and were
stored in a refrigerator at 4°C. For long-term storage, bacterial cultures were stored in

50% glycerol at -80°C.

Field Inoculations and Disease Evaluation. Field plots were established at the
Agronomy Farm of Michigan State University in East Lansing, Michigan on May 28,
2003. All plots were fertilized at planting with 46 kg/ha of 19-19-19, which was handed.
Weed control was achieved by applying 0.38 L/ha ethafluralin and 0.25 L/ha metolachlor
prior to planting. Plots were treated with 0.19 L/ha dimetholate for insect control as
needed during the growing season. The experiment consisted of 42 entries, with each
plot consisting of two 15-foot rows and 3 replications arranged in a 6 x 7 lattice design.
Each replication included an additional block consisting of 12 rows of ‘Midland’ as a
spreader, to increase disease pressure. The 42 entries consisted of 30 experimental F6
lines (chosen from the 38 F5 lines selected for evaluation of CBB resistance), parental
lines (N99219, 99L91-45, 99L91-47, B98306, ‘Phantom’, and B00136), resistant checks
(VAX 5, HR45, XAN 159), and susceptible checks (‘Othello’ and ‘Midland’). The 30
experimental F 6 lines were selected on the basis of SU91, to ensure that 15 lines carried
the marker, and 15 did not. ‘Midland’ seed infected with Xap was used to facilitate the
spread of bacterial inoculum. The ‘Midland’ seed was harvested from diseased field
plots grown at the Montcalm Research Farm the previous year.

Field plots were inoculated with a Xap solution (106 cfu/ml) of the isolate 9712-3
(obtained from the University of Nebraska) on July 16 and July 24, 2003. A power

sprayer was used to apply the bacterial solution at approximately 150 psi

50

and the inoculum was prepared in 76 liter batches. Plants were sprayed until water-
soaking occurred and upper leaf surfaces were completely covered with inoculum.

The field plots were evaluated for various agronomic traits. Days until flowering
(at least 50% of the plants in a plot had a minimum of one open flower) was recorded for
each plot. Leaf disease reaction was evaluated on August 6 and 13, 2003 (13 and 20 days
after the 2nd inoculation, respectively), using the 1-9 scale described by Schoonhoven and
Pastor-Corrales (1987). This rating scale was described by the following guidelines: 1 =
no visible disease symptoms, 3 = 2% of the leaf surface has visible lesions, 5 = 5% of the
leaf surface has visible lesions, 7 = 10% of the leaf surface has visible lesions, 9 = More
than 25% of the leaf surface has visible lesions (Schoonhoven and Pastor-Corrales,
1987). Pod disease reaction was recorded separately from leaf reaction on a scale of 0 to
3 (using intervals of 0.5). The scale was described as follows: 0 = no pod lesions, 1 =
approximately one to three pod lesions, 2 = approximately four to eight pod lesions, 3 =
more than eight pod lesions. Agronomic traits evaluated prior to harvest were: height,
lodging, maturity, and desirability (DS). Height was recorded in centimeters (cm).
Lodging was evaluated on a 1 to 5 scale: 1 = no lodging, 3 = intermediate lodging, and 5
= excessive lodging. Maturity was recorded as days after planting (DAP) with respect to
the day that the plot was entirely mature and ready to be harvested. DS was determined
as a visual rating of overall agronomic desirability. This rating used a 1 to 9 scale with
the following guidelines: 1-4 = above average agronomic traits, 5 = average (compared
to commercial checks), 6—9 = unacceptable agronomic traits. Those entries with a DS of
below 5 are typically selected, with some selections made within those genotypes

receiving a rating of 5. Plants were mechanically harvested using the bean knife method

51

 

on September 5, 2003. Yield data was calculated by harvesting 4.57 meters of the center

2-rows for each 4-row plot.

Experimental Design and Statistical Analysis. Analysis of variance (ANOVA) was
determined for each experiment, using the PROC GLM function of SAS (SAS Institute
Inc., 2000). Traits showing association (p>0.05) with SU91 based on F-values of single-
factor ANOVA were analyzed using Fisher’s Protected LSD (or=0.05). The PROC REG
function of SAS was used for regression analysis of all traits, and r-values were

determined. Correlations were examined using PROC CORR of SAS.

52

RESULTS

Genotypes Evaluated in East Lansing and Saginaw Field Studies. The list of
genotypes included in the East Lansing experiment was presented in Table 1.2. This
experiment included 15 SU91+ F6 lines, 15 SU91- F6 lines, and 12 check lines. The
Saginaw field experiment included additional SU91- lines that were grown as part of the
preliminary yield trial. These additional lines were included in molecular marker

analysis. A list of genotypes grown in Saginaw was presented in Table 1.3.

Use of the SCAR Marker SU91 in Marker-Assisted Selection. In the East
Lansing field experiment, the 1St rating of leaves for CBB resistance was significantly
different between SU91+ and SU91- genotypic classes (p<0.0001), as was the 21nd rating
of leaves for CBB resistance (p<0.0001). The mean leaf ratings for SU91+ lines were
3.37 and 3.48 for the 15‘ and 2nd leaf ratings, respectively. Comparatively, the mean leaf
ratings for the SU91- lines were 4.26 and 4.71. No significant differences were detected
at the 0.05 significance level between genotypic classes in either the 1St or 2"‘1 rating of
pods for CBB resistance (p=0.09 and p=0.052, respectively). A comparison between
phenotypic disease ratings between lines with or without SU91 is summarized in Table
1.4.

The results obtained from the PYT at the Saginaw Valley Bean and Beet Research
Farm differed in that the presence of SU91 was significantly associated with both leaf
(p<0.0001) and pod (p<0.0001) phenotypic disease ratings. Only one rating each of

leaves and pods was recorded at this location. The SU91+ genotypes had a mean leaf

53

disease rating of 4.67, compared to a mean of 6.56 for those lines without SU91. The
mean pod disease ratings were 0.53 and 1.05 for SU91+ and SU91- lines, respectively
(Table 1.4).

Mean CBB leaf ratings for each of the 42 entries included in the East Lansing
experiment were presented (Table 1.2). Means were listed in rank, beginning with the
lowest mean CBB leaf rating (2.30) and ending with the highest (7.63). SU91 genotypic
class was included for each entry (SU91+ or SU91-). Mean CBB pod ratings were not
included, because pod ratings were not significantly associated with SU91.

Mean CBB leaf ratings for each of the 72 entries included at the Saginaw study
were presented in Figure 1.2. Pod ratings were significantly associated with SU91 in the
Saginaw experiment. The mean CBB pod ratings are presented by genotypic class in

Figure 1.3.

SU91 and Yield. A significant negative correlation (r = -0.20*) was observed between
SU91 and yield at the East Lansing location. The mean yield for SU91+ and SU91- lines
were 2.05 i 0.06 and 2.20 i 0.09 T/ha, respectively, which were significantly different
(p= 0.02). The presence of SU91 was not significantly associated with reduced yields at
the Saginaw location (p= 0.08), where the mean yields were 3.08 d: 0.11 and 2.96 i 0.07

T/ha for SU91+ and SU91- lines, respectively (Figure 1.4).

54

Table 1.2 Common Bacterial Blight Leaf and Pod Ratings of 30 F6 lines and check

cultivars or breeding lines, East Lansing (2003).

 

Genotype‘l' SU91: Leaf§ Podfi] Genotype'l‘ SU91: Leaf§ Podfil

 

 

 

HR45 + 2.30 0.89 N03616 - 3.78 1.00
803644 - 2.90 1.00 803647 - 3.81 1.05
803637 + 3.02 1.00 803622 + 3.85 0.92

99L91-45 + 3.03 1.22 803625 + 3.96 0.92
803633 + 3.06 0.94 898306 - 3.99 1.00
803645 - 3.06 1.00 803623 - 4.05 1.00

VAX 5 + 3.06 1.08 N99219 - 4.10 1.00

NO3611 + 3.12 1.00 803646 - 4.11 1.00

XAN 159 + 3.20 0.94 803628 + 4.35 0.75
803632 + 3.34 1.00 803631 + 4.41 0.83
99L91-47 + 3.34 0.92 N03618 - 4.49 1.17
803639 + 3.37 1.00 Phantom - 4.50 1.00
803635 - 3.38 0.92 803613 - 4.52 1.00
803643 + 3.38 0.92 800136 - 4.60 1.08
803630 + 3.43 0.94 803638 - 4.63 1.00
803641 - 3.43 1.00 803620 - 4.73 1.00
N03614 + 3.47 1.17 803629 - 4.84 1.00
803642 + 3.55 1.17 803636 - 4.87 1.08
803627 - 3.57 1.00 Othello - 5.99 0.67
N03617 + 3.60 1.00 Midland - 7.30 2.00
803634 + 3.74 1.00 Midland - 7.63 2.50
T = Genotypes beginning with “B" and “N” denote black and navy bean breeding lines, respectively.

Also included are check genotypes (HR45, 99L91-45, VAX 5, XAN 159, 99L9l-47, N99219,
Phantom, Othello, and Midland).
I = “+" denotes bean genotypes which carry the SCAR marker SU91.
“-“ denotes bean genotypes which do not carry the SCAR marker SU91.
§ = Indicates mean CBB leaf rating for specified bean genotype. The rating scale used was a 1-9 rating of
visual CBB symptoms (1= no visible disease symptoms, 3: Approximately 2% of leaf surface has
lesions, 5: Approximately 5% of leaf surface has lesions, 7= Approximately 10% of leaf surface has
lesions, 9: More than 25% of leaf surface has lesions).
1] = Indicates mean CBB pod rating for specified bean genotype. The ratings scale used was a 0-3 rating of
visual CBB symptoms ( 0= no pod lesions, 1= small pod lesions, 2: large pod lesions, 3= multiple large
pod lesions).

55

Table 1.3 List of genotypes evaluated in preliminary yield test in Saginaw, MI (2003).

 

Genotype‘t SU91: Leaf§ Pod§ Genotypei' SU91: Leaf§ Pod§

 

N03614" + 3.0 0.0 N00729 - 6.0 1.0
N03617* + 3.5 0.0 803613“ - 6.5 1.0
803632" + 3.5 0.0 803624 - 6.5 1.5
803633“ + 3.5 0.5 803640 - 6.5 0.5
803637“ + 4.0 0.5 800136 - 6.5 1.0
803639* + 4.0 0.0 I03361 - 6.5 1.5
803643“ + 4.0 0.5 l03366 - 6.5 1.0
N02234 - 4.0 1.0 103370 - 6.5 1.0
l03373 - 4.0 0.0 N00792 - 6.5 1.0
803622’ + 4.5 0.5 N00838 - 6.5 1.0
803647' - 4.5 0.5 l03390 - 6.5 1.0
N02250 - 4.5 1.0 803619 - 7.0 1.5
103367 - 4.5 1 .0 803620* - 7.0 1 .0
N03611* + 5.0 0.5 803623" - 7.0 1.0
N03618* - 5.0 1.5 803626 - 7.0 1.0
803630" + 5.0 1 .0 803636* - 7.0 0.5
803634“ + 5.0 0.5 803638‘ - 7.0 1.0
80364? + 5.0 0.5 803641 * - 7.0 1.5
N02247 - 5.0 1.0 Jaguar - 7.0 1.0
898306 - 5.0 1.0 Condor - 7.0 1.0
l03362 - 5.0 1.0 Vista - 7.0 1.0
803612 - 5.5 1.0 l02521 - 7.0 1.0
803621 - 5.5 1.0 N00756 - 7.0 1.5
803625* + 5.5 1.0 803615 - 7.5 1.5
803635' - 5.5 1.0 N03616' - 7.5 1.5
l03371 - 5.5 1.0 803627" - 7.5 1.5
803631 * + 6.0 1.0 803629* - 7.5 1.0
803644" - 6.0 1 .0 803645* - 7.5 1 .0
803646' - 6.0 1.5 T-39 - 7.5 1.0
N02249 - 6.0 1.0 l03364 - 7.5 1.0
N99219 - 6.0 1.0 N00760 - 7.5 1.0
115-M - 6.0 1.0 N00794 - 7.5 1.0
l03363 - 6.0 1.5 l01794 - 7.5 1.0
l03365 - 6.0 1.5 803628‘ + 8.0 1.5
l03369 - 6.0 1.0 103368 - 8.0 1.0
l03372 - 6.0 1.0 Seahawk - 8.0 1.0

 

 

* = Denotes genotypes evaluated in both East Lansing and Saginaw.

1' = Genotype names beginning with “8” or “N” are black or navy bean lines, respectively.
Genotype names beginning with “I” are bean breeding lines from other breeding programs.

I = “+” denotes bean genotypes which carry the SCAR marker SU91.
“-“ denotes bean genotypes which do not carry the SCAR marker SU91.

§ = “Leaf" denotes the mean C88 leaf rating for the genotype. Leaf ratings were recorded on a 1-9 scale,
with 1 = uninfected and 9 = severe necrotic regions.
“Pod” denotes the mean C88 pod rating for the genotype. Pod ratings were recorded on a 0-3 scale,
with O = uninfected and 3 = severe pod infection.

11 =Denotes susceptible bean cultivars or breeding lines included as checks for yield comparisons.

56

Table 1.4 Mean leaf and pod common bacterial blight (CBB) ratings for SU91+ and

SU91- F6 lines from field plots at East Lansing and Saginaw, MI (2003).

 

Location SU911' Leafli Leat21 Mean-Leaf Pod1§ Pod2§ Mean-Pod

 

East Lansing + 3.37 3.48 3.43 0.95 1.01 0.98

- 4.26 4.71 4.49 1.05 1.17 1.11

LSDoos'll 0.40* 0.38* 0.39* 0.11 0.16 0.14
Saginaw# + 4.67 - - 0.53 - -
- 6.56 - - 1.05 - -
LSDoosfil 0.64* - - 0.23* - -

 

* = Denotes significance at the 0.05 probability level.
1' = “+” represents the class of genotypes which carry the SCAR marker SU91.
= “-“ represents the class of genotypes which do not carry the SCAR marker SU91.

I = Denotes mean CBB leaf ratings 1 and 2 (13 and 20 days after the final inoculation, respectively) in
East Lansing (2003). Leaf ratings were recorded on a 1-9 scale, with 1 = uninfected and 9 = severe
necrotic regions.

§ = Denotes mean CBB pod ratings 1 and 2 (l3 and 20 days after the final inoculation, respectively) in
East Lansing (2003). Pod ratings were recorded on a 0-3 scale, with 0 = uninfected and 3 = severe pod
infection.

1] = Fisher’s Least Significant Difference (LSD)

# = Only one CBB leaf rating and one CBB pod rating were taken at the Saginaw location. This
experiment was not inoculated, natural infection was sufficient for screening.

57

 

 

 

 

 

 

 

 

 

 

12
a 10

d)

C

.3

”‘6 8

3 mSU91+
8 ISU91-
C

0

3

0'

2

LI.

 

 

 

CBB Leaf Rating

 

 

 

Figure 1.2. Frequency of leaf CBB ratings between bean lines with SU91 (SU91+) and
those without SU91 (SU91-). All 72 entries from the navy and black bean preliminary
yield trials are represented. The rating scale used was the previously described 1-9 scale
(1 = no infection, 9 = severe necrotic lesions). The minimum and maximum mean leaf

ratings for this study were 3 and 8, respectively.

58

 

01
O

 

 

 

 

|-—l
O

 

O

’u?

a:

.E 40

.J

3 3O EllSU91+
35

> 20 I SU91-
u

5

3

3'

E

0 0511.5 2 2.5 3
C88 Pod Rating

 

 

 

Figure 1.3. Frequency of pod CBB ratings between dry bean lines with SU91 (SU91+)
and those without SU91 (SU91—). All 72 entries from the navy and black bean
preliminary yield trials are represented. The rating scale used was the previously
described 0-3 scale (0 = no pod lesions, 3 = severe pod lesions). The minimum and

maximum mean pod ratings for this study were 0 and 1.5, respectively.

59

 

.09
01

 

EISU91+
ISU91-

 

 

 

Yield (Mn)
N

 

EastLansmg Sagmaw

Locafion

 

 

 

Figure 1.4. Comparison of mean yield between lines with (SU91+) and without (SU91-)
the SCAR marker SU91. The mean yields at East Lansing of 2.05 and 2.20 T/ha for
SU91+ and SU91- lines, respectively, were significantly different (p= 0.02). Mean yields
at Saginaw of 3.08 and 2.96 T/ha, for SU91+ and SU91- lines, respectively, were not

significantly different (p= 0.08).

60

SU91 and Maturity. The presence of SU91 was not associated with later maturity
(compared to lines without the marker), in either location. The mean days to maturity for
both SU91+ and SU91- lines was 95 days in East Lansing (Figure 1.5), ranging from 91
to 97 days. The mean days to maturity was 99 days in Saginaw (Figure 1.6), ranging
from 97 to 103 days. The mean maturity of SU91+ and SU91- lines was not significantly

different (p= 0.22 and p= 0.24) for East Lansing and Saginaw, respectively.

SU91 and Seed Weight. The presence of SU91 was not significantly associated with
bean seed weight at either location. The mean seed weight for SU91+ and SU91- lines in
East Lansing was 21.7 g/ 100 seeds and 20.9 g/ 100 seeds, respectively. These means
were not significantly different. The mean seed weight for SU91+ and SU91- lines in
Saginaw was 18.7 g/ 100 seeds and 18.0 g/ 100 seeds, respectively. These means were not

significantly different (p= 0.065).

Yield and Phenotypic Disease Ratings. To evaluate the impact of C88 resistance on
yield, the visual disease ratings were correlated to yield at both locations. For the East
Lansing experiment, only the 2nd pod rating was associated with yield. A negative
correlation was observed (r = -0.26**) between yield and the 2nd pod rating. In contrast,
an examination of the Saginaw study illustrated an association between yield and leaf
ratings (r = -0.35**), while there was no significant association between yield and pod

ratings.

61

Leaf and Pod Disease Ratings. Leaf and pod ratings were compared to examine
potential differences in tissue susceptibility (Table 1.2 and Table 1.3). For the East
Lansing study, the 1" leaf rating showed a positive correlation with the lSt pod rating (r =
0.39***). The 2nd leaf rating was also positively correlated with the 2nd pod rating (r =
0.59***). The two leaf ratings were strongly correlated (r = 0.77***), as were the two
pod ratings (r = 0.52***). The leaf and pod ratings were also correlated in the Saginaw

study as well (r = 0.65***).

62

 

EISU91+
ISU91-

Frequency (# of Lines)
#

 

91 92 93 94 95 96 97
Days to Maturity

 

 

 

Figure 1.5 Maturity is compared between SU91+ and SU91- lines at East Lansing, 2003.
Days to maturity was recorded as the average for 3 replications of each line. The overall
mean maturity for the experiment was 95 days. The mean maturity for both SU91+ and

SU91- genotypes was also 95 days.

63

 

 

, lsus1+
ISU91-

Frequency (#of Lines

 

 

97 98 99 100 101 102 103
Days to Maturity

 

 

Figure 1.6. Maturity is compared between SU91+ and SU91- lines at Saginaw, 2003.
Days to maturity was recorded as the average for 3 replications of each line. The overall
mean maturity for the experiment was 99 days. The mean maturity for both SU91+ and

SU91- genotypes was also 99 days.

64

Yield Results from CBB Resistance Screening Study -— East Lansing 2003

The 30 genotypes included in the C88 screening experiment in East Lansing are
listed in Table 1.2. Of the 30 lines, 15 had the SU91 marker, while the remaining 15 did
not. Also included were 11 checks (N99219, 99L91-45, 99L91-47, B98306, ‘Phantom’,
800136, ‘Othello’, VAX 5, HR45, XAN 159, and ‘Midland’).

Yield summaries of the SU91+ and SU91- genotypes are presented in Table 1.6.
Check genotypes are included for a yield comparison. Yields of this experiment (across
genotypes) ranged from 1.44 to 2.73 T/ha, respectively and the overall mean yield for this

study was 2.13 T/ha.

65

Table 1.6 Yield (T/ha) of bean genotypes with or without the SCAR marker SU91

compared to check varieties and cultivars, East Lansing, MI (2003).

 

 

Ranki' Genotypei SU91 Yield (T/ha)
1 803625 + 2.73
2 803645 - 2.59
3 803613 - 2.58
4 803629 - 2.51
5 Phantomtl - 2.49
6 800136 - 2.45
7 803635 - 2.41
8 803623 - 2.39
9 803638 - 2.35
10 803630 + 2.33
11 803631 + 2.33
12 803644 - 2.32
13 N99219§ - 2.26
14 898306 - 2.20
15 803637 + 2.20
16 Othello§ - 2.18
17 803633 + 2.18
18 803642 + 2.18
19 VAX 51] + 2.18

20 99L91-471] + 2.17
21 803647 - 2.13
22 803643 + 2.08
23 803641 - 2.08
24 N03618 - 2.07
25 N03616 - 2.06
26 803634 + 2.05
27 803620 - 2.04
28 N03614 + 2.03
29 803627 - 2.03
30 803632 + 2.02
31 99L91-451] + 2.00
32 N03611 + 1.98
33 803622 + 1.96
34 803639 + 1.93
35 803628 + 1.92

 

66

Table 1.6, continued.

 

 

 

Rank‘l' Genotype: SU91 Yield (T/ha)
36 803646 - 1.91
37 803636 - 1.83
38 XAN 15911 + 1.78
39 N03617 + 1.73
40 HR4511 + 1.65
41 Midland§ - 1.62
42 Midland§ - 1.44

Overall Mean 2.13
LSD (0.05) 0.39
CV 11.4

 

'1' = Indicates rank by yield (T/ha) of genotype among 42 entries.

I = Genotypes beginning with “B” and “N” denote black and navy bean breeding lines, respectively.
§ = Denotes susceptible (SU9l-) check varieties or breeding lines.

1] = Denotes resistant (SU91+) check varieties or breeding lines.

67

Yield Results from Preliminary Yield Trial (PYT) - Saginaw 2003

All lines evaluated in the East Lansing study were also evaluated in the test in
Saginaw. Additional lines were included in the Saginaw experiment for the purpose of
yield comparison. A summary of relevant genotypes evaluated in the Saginaw field study
is presented in Table 1.3.

Among the 37 experimental F 6 lines tested, 15 had the SU91 marker (SU91+),
while 22 lines did not have the marker (SU91-). Of the 15 SU91+ lines, 10 yielded above
the average yield of the test, which was 2.92 T/ha. Additionally, 5 of these lines yielded
higher than all of the checks included for yield comparison. The yields of the 15 SU91+
lines are summarized in Table 1.7 in comparison to the check genotypes (B98306,
Condor, 115-M, Vista, 800136, Jaguar, N99219, Seahawk and T-39). The yield data is
presented by rank among the 72 total entries for the test. The minimum and maximum
yields of the study were 2.08 and 3.65 T/ha, respectively.

In contrast, only 11 of the 22 SU91- lines yielded higher than the mean yield of
2.92 T/ha. A summary of the yields for these lines is also presented in Table 1.7. The
yield data for the check genotypes is included for comparison. Four of these SU91- lines
(803623, 803621, 803627, and 80361) yielded higher than the highest-yielding check

(Condor, which yielded 3.16 T/ha).

68

Table 1.7 Yield (T/ha) of bean genotypes with or without the SCAR marker SU91

compared to check cultivars, Saginaw, MI (2003).

 

Genotypet SU911 Yield (T/ha)

 

803625 + 3.65
803623 - 3.60
803637 + 3.55
803643 + 3.51
803621 - 3.47
803639 + 3.41
803622 + 3.40
103363 - 3.30
898306 - 3.29
803627 - 3.25
803641 - 3.20
Condor - 3.16
803635 - 3.15
803634 + 3.12
103369 - 3.12
N03614 + 3.10
803642 + 3.08
NO3618 - 3.08
803647 - 3.08
803626 - 3.06
103390 - 3.05
l 15-M§ - 3.05
803636 - 3.03
803631 + 3.01
Vista§ - 3.00
803613 - 3.00
803640 - 2.99
102521 - 2.98
800136 - 2.96
103368 - 2.96
803632 + 2.95
103364 - 2.93
803624 - 2.90
103366 - 2.90
Jaguar§ - 2.90

 

69

Table 1.7, continued.

 

 

 

Genotypet SU91: Yield (T/ha)
N99219§ - 2.89
803633 + 2.88
803620 - 2.86
803630 + 2.83
Seahawk§ - 2.81
N03616 - 2.79
103365 - 2.77
N03617 + 2.75
N03611 + 2.74
803615 - 2.74
803629 - 2.73
803612 - 2.71
803619 - 2.71
803645 - 2.68
103362 - 2.66
803638 - 2.64
101794 - 2.63
803646 - 2.57
803628 + 2.54
103361 - 2.53
103371 - 2.52
803644 - 2.51
103370 - 2.47
103367 - 2.43
T-39§ - 2.43
103373 - 2.42
103372 - 2.08
Overall Mean 2.92
LSD (0.05) 0.37
CV % 11.4

 

'1' = Genotypes beginning with “8” and “N” denote black and navy bean breeding lines, respectively.
I = Denotes presence (+) or absence (-) of the SCAR marker SU91
§ = Denotes check cultivars or genotypes used in yield comparison.

70

DISCUSSION

In the current study, the association between the SU91 SCAR marker and C88
field resistance was examined to address the situation of population specificity of marker-
trait associations. Additionally, because linkage drag has been reported for a number of
C88 resistance sources, SU91 was examined to determine if it was significantly
associated with negative agronomic traits (Valladares-Sanchez et al., 1979; Beebe et al.,
1989).

The use of MAS as a tool in the selection of quantitative traits is ideal in cases
where traditional phenotypic selection to improve the trait has proven inadequate. MAS
can provide an alternative for selection of traits that exhibit low heritability, and are not
responsive to visual selection (Dudley, 1993). Resistance to C88 in common bean is a
quantitative trait that exhibits low to moderate heritability, and should be an excellent
candidate for MAS (Yu et al., 2000). In order for MAS to be useful in a breeding
program, however, several factors must be considered. Since QTL-marker associations
in breeding programs may be population specific, the usefulness of a marker in selecting
for the quantitative trait must be verified in other populations (Dudley, 1993; Yu et al.,
2000). Based on field experiments at two locations in 2003, presence of the SCAR
marker SU91 was clearly associated with significant differences in C88 foliar resistance
in the populations under study, but the association was less clear in the pods. The marker
was associated with high levels of leaf resistance in both experiments, and was also
associated with pod resistance in one of the two experiments. The results from the

Saginaw study were especially encouraging, because the difference in mean C88 leaf

71

ratings was approximately 2 units on the 1-9 scale used in C88 leaf evaluations, which
was a significant difference.

It is interesting to note that the CBB-resistant black bean check VAX 5, which
was the source of the resistance in these lines, received an average C88 leaf rating of 3.1.
As no other resistant parents were used to derive these lines, this C88 leaf rating gives an
approximation as to the lowest ratings that should be expected. Although values less than
this were observed, they were not significantly different.

The lack of significant association between SU91 and pod disease resistance in
the East Lansing study may be a result of the rating scale used. The rating scale of 0-3
may not successfiilly discriminate between severities of pod infections. To better
differentiate between C88 resistant and susceptible bean pods, a 1-9 scale
complementary to the scale used for leaf disease ratings was adopted for future field
evaluations. Although the rating scale used in leaf disease ratings, as proposed by
Schoonhoven and Pastor-Corrales (1987), allowed for simultaneous leaf and pod ratings,
the scale was not used in this manner. Previous studies have determined that differences
in tissue susceptibility exist between genotypes (Aggour et al., 1989; Valladares-Sanchez
et al., 1979), and that a scale with only one value for leaves and pods is not satisfactory.
This scale will continue to be used, however, for the evaluation of leaf disease reaction in
both field and greenhouse experiments. Given past reports of differences in tissue
susceptibility within bean cultivars (Aggour et al., 1989; Valladares-Sanchez et al., 1979)
it is possible that the SU91 QTL does not confer adequate levels of resistance to C88 in
the pods of the population used in this study. This suggests an alternative explanation for

the lack of association between the marker and C88 pod resistance.

72

Relationships between important agronomic traits in common bean and SU91
were examined at both locations. Previous research has indicated a potential for linkage
drag associated with C88 resistance. Late maturity and low yield are two important
traits that have been problematic in past bean breeding efforts (Valladares-Sanchez et al.,
1979). The possibility of linkage between C88 resistance and unfavorable agronomic
traits must be examined prior to the development of a successful marker-assisted
selection program.

The SU91 marker was weakly correlated with lower bean yields in the East
Lansing study. No such correlation was evident for the Saginaw experiment. Despite the
negative implications of the association in East Lansing, high yielding SU91+ lines with
high levels of field C88 resistance were observed. The highest yielding line in East
Lansing, which carried the SU91 marker, was also the highest yielding line in the test at
Saginaw. The test in Saginaw was direct-harvested, while the East Lansing C88
screening study was knife-pulled prior to threshing. Based on this preliminary evidence,
this black bean line (803625) seems to be well-suited for either type of harvest method.
Although a slight negative correlation between SU91 and yield was apparent in East
Lansing, the identification of genotypes such as 803625 is promising for the
development of high-yielding cultivars with enhanced levels of C88 resistance. The
observation of yield drag associated with SU91+ in East Lansing may indicate a yield
penalty caused by the presence of the QTL that can only be observed in cases of low
disease pressure, as the disease pressure was lower at that location. The C88 resistance
conferred by SU91 may have obscured this effect in environments where the disease

pressure is high. Yield comparison of these lines in the complete absence of the disease

73

will be required to determine whether the presence of SU91 is associated with linkage
drag in these genotypes. The correlation between SU91 and reduced yield in the location
with lower disease pressure indicates that such an association between SU91 and linkage
drag may exist.

Late maturity is another negative agronomic trait that has commonly been
reported as being associated with C88 resistance in beans (Coyne and Schuster, 1974).
This linkage has been broken in the past by other research projects (Coyne and Schuster,
1973), however it is a concern that must be addressed. SU91 was not associated with late
maturity in either field location. This information is encouraging for the development of
bean cultivars with suitable levels of C88 resistance, as it allows for the selection of
early-maturing, high-yielding, C88 resistant bean lines.

The presence of SU91 was not correlated with the other agronomic traits
evaluated in either location (days to flowering, lodging, height, and seed weight). The
lack of linkage between C88 resistance and these important agronomic traits indicates
that selection of C88 resistant genotypes with favorable agronomic characteristics is
feasible in the genetic backgrounds described in the present study.

In conclusion, although the use of MAS will not completely eliminate the need for
direct selection for C88 resistance, it can be applied in earlier generations to reduce the
number of lines that require direct screening. Potential C88 resistant breeding materials
could be screened for molecular markers, such as SU91, in the F2 generation. Although
MAS is more expensive than direct screening for many traits, a study by Yu et al. (2000)
determined that MAS is more cost effective than conventional screening in selecting

C88 resistant lines. The primary reason for this is the inefficiency of direct screening

74

using Xap, because of the effects of environment on the screening results. By eliminating
genotypes that do not have SU91 from the screening process, direct screening programs
can be optimized. Inclusion of exclusively SU91+ lines in field and greenhouse
screening will increase the possibility of selecting highly resistant genotypes. The direct
screening in the greenhouse and field could begin in later generations, for those
genotypes which carry the marker, as determined by the indirect selection in the F2
generation. Conducting the direct screening in later generations is ideal, as the seed
supply at that point would allow for replicated testing, whereas seed supply in early
generations may be insufficient for replicated studies. Screening of potentially resistant
materials in several locations and multiple years in the field would be desirable.

Although it may be difficult to screen breeding materials with multiple Xap isolates in the
field, this portion of the C88 resistance verification could be performed in the
greenhouse.

The results reported here are based on data from two field locations in 2003. It
would be important to confirm these results across multiple years and locations, as
suggested above. Because of the unreliability of visual phenotypic ratings for C88
resistance, these additional evaluations would validate the usefulness of screening for
SU91 in a MAS program.

Advanced materials selected from this research are currently being utilized to
introgress C88 resistance into other market classes of dry bean. Suitable levels of
resistance are not currently present in commercial cultivars of the major bean market

classes cultivated in the United States. Highly resistant black and navy bean lines

75

identified in this research may be useful parents in the introduction of resistance into

other market classes, where resistance is lacing.

76

REFERENCES

Aggour, A.R., D.P. Coyne, and AK. Vidaver. 1989. Comparison of leaf and pod disease
reactions of beans (Phaseolus vulgaris L.) inoculated by different methods with
strains of Xanthomonas campestris pv. Phaseoli (Smith) dye. Euphytica 43: 143—
152.

Beebe, 3.8. 1989. Quantitative genetics in Phaseolus vulgaris: the example of resistance
to Xanthomonas campestris pv. phaseoli. Pages 213-238. In: Current Topics in
Breeding of Common Bean. Proc. Int. Bean Breeding Workshop Working Doc.
No. 47. S. Beebe (ed.) CIAT, Cali, Colombia.

Claflin, L.E., A.K. Vidaver, and M. Sasser. 1987. MXP, a semi-selective medium for
Xanthomonas campestris pv. phaseoli. Phytopathology 77:730-734.

Coyne, DP. and ML. Schuster. 1973. Phaseolus germplasm tolerant to common blight
bacterium (Xanthomonas phaseoli). Plant Dis. Reptr. 57:11 1-1 14.

Coyne, DP. and ML. Schuster. 1974. Inheritance and linkage relations of reaction to
Xanthomonas phaseoli (E. F. Smith) Dowson (common blight), stage of plant
development and plant habit in Phaseolus vulgaris L. Euphytica 23:195-204.

Dudley, J .W. 1993. Molecular markers in plant improvement: manipulation of genes
affecting quantitative traits. Crop Sci. 33:660-668.

Ekpo, E.J.A., and AW. Saettler. 1976. Pathogenic variation in Xanthomonas phaseoli
and X. phaseoli var. fuscans. Plant Dis. Rptr. 60:80-83.

F rahm, MA. 2002. Effects of terminal drought stress on black beans. M.S. Michigan
State University. East Lansing.

Goss, R.W. 1939. The relation of temperature to common and halo blight of beans.
Phytopathology 30:258—264.

Johnson, H.W., and R. L. Bernard. 1962. Soybean genetics and breeding. Adv. Agron.
14:149-221.

Jung, G., P.W. Skroch, J. Nienhuis, D.P. Coyne, E. Amaud-Santana, H.M. Ariyarathne,
and J .M. Marita. 1999. Confirmation of QTL associated with common bacterial
blight resistance in four different genetic backgrounds in common bean. Crop Sci.
39:1448-1455.

Kelly, J .D., and LO. Copeland. 1996. Status of common bacterial blight problems in
Michigan dry beans. Pages 173-177. In: ler. Taller Interacional sobre bacteriosis
Comun del frijol. Oficina de Coordinacion Regional de Profrijol, Cesda, San
Cristobal, Republica Domincana.

77

Kelly, J .D., and RN. Miklas. 1999. Marker-assisted selection. Pages 93-123. In:
Common Bean Improvement in the Twenty-First Century, lSt ed. Kluwer
Academic Publishers, Dordrecht, The Netherlands.

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

Miklas, P.N., J .R. Smith, R. Riley, K.F. Grafton, S.P. Singh, G. Jung, and DP. Coyne.
2000. Marker-assisted breeding for pyramided resistance to common bacterial
blight in common bean. Ann. Rept. Bean Improv. Coop. 43:39-40.

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

Paterson, A.H., 8. Damon, J .D. Hewitt, D. Zamir, H.D. Rabinowitch, S.E. Lincoln, 8.
S. Lander, and SD. Tanksley. 1991. Mendelian factors underlying quantitative
traits in tomato: comparison across species, generations, and environments.

Genetics 127:181-197.

Pedraza, F ., G. Gallego, S. Beebe, and J. Tohme. 1997. Marcadores SCAR y RAPD para
la resistencia a la bacteriosis comun (CBB). Pages 130-134. In: Taller de

Mejoramiento de Frijol para el Siglo XXI: Bases para Una Estrategia para
America Latina. S.P. Singh and O. Voysest (eds.). CIAT, Cali, Colombia.

Saettler, AW. 1971. Chemical sprays for the control of common-fuscous bacterial blight
in Navy (pea) beans. Ann. Rept. Bean Improv. Coop. 14:54-55.

Schoonhoven, A. van, and MA. Pastor-Corrales. 1987. Standard system for the
evaluation of bean germplasm. CIAT, Cali, Colombia.

Schuster, ML. and DP. Coyne. 1971. New virulent strains of Xanthomonas phaseoli.
Plant Dis. Rptr. 55:505-506.

Schuster, ML. and DP. Coyne. 1975. Genetic variation in bean bacterial pathogens.
Euphytica 24:143-147.

Schuster, M.L., D.P. Coyne, T. Behre, and H. Leyna. 1983. Sources of Phaseolus species
resistance and leaf and pod differential reaction to common blight. HortScience.

18:901-903.

Singh, PS, and CG. Munoz. 1999. Resistance to common bacterial blight among
Phaseolus species and common bean improvement. Crop Sci. 39:80-89.

78

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

Valladares-Sanchez, N.E., D.P. Coyne, and ML. Schuster. 1979. Differential reaction of
leaves and pods of Phaseolus germplasm to strains of Xanthomonas phaseoli and
transgressive segregation for tolerance from crosses of susceptible germplasm. J.

Amer. Soc. Hort. Sci. 104:648-654.

Weller, D.M. and AW. Saettler. 1980. Evaluation of seedbome Xanthomonas phaseoli
and X. phaseoli var. fuscans as primary inocula in bean blights. Phytopathology
70: 148-1 52.

Yousef, G.G. and J .A. J uvik. 2001. Comparison of phenotypic and marker-assisted
selection for quantitative traits in sweet corn. Crop Sci. 41:645-655.

Yu, K., S.J. Park, and V. Poysa. 2000. Marker-assisted selection of common beans for
resistance to common bacterial blight: efficacy and economics. Plant Breeding
1192411-415.

Zaumeyer, W.J. and HR. Thomas. 1957. Bacterial diseases of major importance. Pages
65-88. In: A monographic study of bean diseases and methods for their control.
W.J. Zaumeyer and HR. Thomas (eds.). United States Department of Agriculture
(USDA), Washington, DC.

79

CHAPTER TWO

THE USE OF MARKER-ASSISTED SELECTION TO PYRAMID RESISTANCE
TO COMMON BACTERIAL BLIGHT AND ANTHRACNOSE IN COMMON
BEAN (Phaseolus vulgaris L.)

ABSTRACT

The used of marker-assisted selection to develop dry bean lines with resistance to
common bacterial blight (C88) and anthracnose was examined. Sources of C88 and
anthracnose resistance were combined in 4-way crosses with previously developed CBB
resistant bean breeding lines. Selections that carried the SCAR marker SU91 (linked to a
C88 resistance QTL on bean linkage group 88), which were identified in a previous
study, were crossed to breeding lines with the anthracnose resistance gene C0-42. The
resulting F1 progeny were crossed with Fl plants from the cross HR45/Kaboon to
generate 4-way crosses. HR45 is a C88 resistant small white navy bean used as a donor
parent of a QTL for CBB resistance (BC420) mapped to bean linkage group 86. Kaboon
is an Andean anthracnose differential cultivar that carries the anthracnose resistance gene
Co-I‘ that mapped to bean linkage group 81.

The F2 populations originating from the 4-way crosses were screened for SCAR
markers SU91, BC420, and SAS13, linked to C88 resistance QTL (on bean linkage
groups 88 and B6) and anthracnose resistance loci (on bean linkage group 88). Potential
resistance conferred by the C88 resistance QTL linked to SU91 (on linkage group 88)
and BC420 (on linkage group 86), was examined in both the field and greenhouse. The
SAS13 marker was used to screen for the C 0-42 Middle American anthracnose resistance

gene. Results from anthracnose screening are based only on presence of molecular

80

markers, as direct screening for anthracnose resistance was not included in this study
because a differentiating race of the pathogen was unavailable.

Neither SU91 nor BC420 was associated with C88 resistance in the field
experiment. This was unexpected, as the results obtained in Chapter One illustrated the
potential for SU91 to identify CBB resistant bean genotypes. The level of disease
incidence in this study may have been inadequate for a comparison of marker genotypes.

In two greenhouse experiments, however, both SU91 and BC420 were associated
with increased CBB resistance. SU91 was associated with both leaf (r= -0.20**) and pod
(r= -0.17*) resistance in experiment 1. The results were similar in experiment 2, as SU91
was associated with both leaf (r= -0.15**) and pod (1'2 —0.27***) resistance to C88. The
marker BC420 was only associated with pod resistance (r= -0.19*) in experiment 1, and
leaf resistance (r= -0.18**) in experiment 2. Unexpectedly, the presence of both SCAR
markers in a single genotype conferred levels of C88 resistance inferior to that of

genotypes with either SU91 or BC420 alone.

81

INTRODUCTION

Common bacterial blight (CBB), caused by the bacterial pathogen Xanthomonas
axonopodis pv. phaseoli (Smith) Dye (Xap), is a destructive disease of common bean
(Phaseolus vulgaris L.). Resistance to C88 is controlled as a quantitative trait exhibiting
low to moderate heritability (Amaud-Santana et al., 1994; Coyne and Schuster, 1974),
although partial resistance exhibiting high (h2 = 0.87) heritability has been observed in
some genetic studies of C88 resistance in common bean (Silva, 1989).

Research conducted in various bean breeding programs, using diverse genetic
backgrounds have differed greatly in estimating the number of genes or QTL controlling
resistance to C88. Resistance to C88 was controlled by only a single dominant gene in
one genetic study, however, that resistance was obtained from tepary bean (Phaseolus
acutifolius), not common bean (Drijfhout and Blok, 1987). McElroy (1985) determined
that one major gene and several minor genes were responsible for the C88 resistance in
XAN 159, a C88 resistant breeding line derived from a cross between P. vulgaris and P.
acutifolius. Eskridge and Coyne (1996) found that as many as five genes may control
CBB resistance in common bean. Additionally, molecular information has suggested the
association of up to six distinct QTL with C88 resistance (Nodari et al., 1993; Jung et al.,
1996). Between two and six QTL were associated with C88 resistance, depending on
the isolate used for evaluating disease reaction and the bean tissue (leaves, pods, or seeds)
inoculated (Jung et al., 1996). The polymorphic markers associated with resistance in
analysis by Jung et al. (1996) each explained 14% to 34% of the total phenotypic
variation, depending on the combination of isolate and bean tissue. Nodari et al. (1993)

determined that at least four independent QTL were involved in C88 resistance.

82

Together, these four putative QTL explained 75% of the total phenotypic variation for
CBB resistance in common bean.

One possible explanation for the different numbers of estimated genes or QTL
controlling quantitative traits was proposed by Paterson (1995). Based on quantitative
trait data in rice (Oijyza sativa) and sorghum (Sorghum bicolor), such traits are controlled
by a small number of major-effect genes and a large number of minor-effect genes that
make extremely small contributions to the overall phenotypic variance. This scenario
provided an explanation of the results obtained in studies of C88 resistance in common
bean. A small number of QTL responsible for a large proportion of the total phenotypic
variance for C88 resistance, have been discovered. At least four major-effect QTL
conditioning C88 resistance have been identified. These four QTL have been mapped to
bean linkage groups 86, B7, 88, and 810, respectively (Nodari et al., 1993; Pedraza et
al., 1997; Ariyarathne et al., 1999; Miklas et al., 2000). These QTL do not entirely
account for the total phenotypic variance, and it is therefore possible that a large number
of minor QTL are also involved. There is limited information, however about the
individual genes involved in C88 resistance, or the mechanism by which they confer
resistance.

Colletrichium Iindemuthianum (the causal agent of bean anthracnose), is a
destructive seed-bome pathogen of common bean localized to humid production areas in
North America. Two highly virulent races (7 and 73) are present in Michigan (Kelly et
al., 1994). The availability of resistance sources with dominant R-genes conditioning
resistance to specific races of the pathogen have reduced the impact of anthracnose in

North America. The C 0-42 anthracnose resistance gene, identified in the Middle

83

American cultivar G 2333, provides a broad spectrum of resistance to the majority of
known C. lz'ndemutlzianum races (Young et al., 1998). The Co-I‘ gene, an anthracnose
resistance gene of Andean origin is another valuable source of genetic resistance
(Balardin and Kelly, 1998). Introgression of both an Andean and a Middle American
anthracnose resistance gene into single cultivars is the most effective means of
developing broad-based protection to anthracnose (Young and Kelly, 1996; Melotto and
Kelly, 2000).

The combination of anthracnose and C88 resistance in a single cultivar would be
a valuable contribution to the development of improved bean cultivars. A small white
navy bean HR45, with C88 resistance derived from XAN 159 was released in 1994
(Park and Dhanvantari, 1994). High levels of C88 pod resistance have been observed in
this line. Although HR45 also exhibited leaf resistance, the level of leaf resistance was
lower than that exhibited by XAN 159. HR45 was used to introgress an additional QTL
for CBB resistance into lines selected from the research presented in Chapter One of this
thesis. The Andean differential cultivar Kaboon was utilized as a donor of the Co-I‘
anthracnose gene, an allele of the Co-I gene identified in Michigan Dark Red Kidney
(Melotto and Kelly, 2000). Two small-seeded black bean lines developed at Michigan
State University, Phantom*Z/SEL 1308 or J aguar*2/SEL 1308, were used to introduce
the C0-42 anthracnose resistance gene (Middle American). The donor of C0-42 was SEL
1308, which is a bean breeding line with anthracnose resistance derived from the highly
resistant cultivar G 2333 (Young et al., 1998).

Breeding strategies involving quantitative traits, such as CBB resistance, create

challenges for breeders that may not be encountered when breeding qualitative traits.

84

With disease resistance, for example, the partial resistance conferred by a single QTL
may not provide adequate resistance. The pyramiding of multiple resistance QTL in a
single genotype offers a possible solution. Gene pyramiding for quantitative disease
resistance may not provide total protection against a pathogen, however the level of
resistance may be higher than that of a single QTL (Castro et al., 2003).

Gene (or QT L) pyramiding became much more feasible with the introduction of
molecular marker technology. The phenomenon of epistasis, or interaction between
multiple loci that control expression of a phenotype, made gene pyramiding for disease
resistance difficult prior to the advent of linked markers. Traditional phenotypic
evaluation may not adequately separate the effects of QTL or individual genes. QTL
pyramiding of multiple loci associated with disease resistance allows not only the
development of lines with increased resistance, but also the opportunity to study the
effects of QTL in a novel genetic background. Interactions with the environment and
other QTL can also be examined in populations with pyramided resistance (Castro et al.,
2003)

Multiplex PCR provides an advantage in pyramiding multiple loci for resistance
in a single cultivar, as it allows for the simultaneous screening of multiple molecular
markers in a single reaction (Miklas et al., 2000). For this approach to be useful,
however, the annealing temperature of the primers must be similar. The SU91 and
BC420 primers used in selection for this research, have annealing temperatures of 58 and
50°C, respectively. Another SCAR linked to C88 resistance in common bean, SAP6,
has an intermediate annealing temperature of 55°C (Miklas et al., 2000). Although this

marker was monomorphic in our genetic population, and is therefore not useful for

85

selection, its PCR profile is compatible for amplification of SU91 and BC420 in
multiplex PCR. Using this program in PCR for both SU91 and BC420 can reduce the
cost and time involved in conducting marker-assisted selection (MAS). SAP6 is useful
for selection of C88 resistance in some genetic backgrounds, such as the kidney bean
cultivar ‘Montcalm’ (see Appendix A2).

The marker SU91 is linked to a QTL on bean linkage group 88 that confers
partial resistance to C88 in both leaves and pods (Pedraza et al., 1997). This marker was
derived from the random amplified polymorphic DNA (RAPD) marker U9700. XAN 159
was the original source of the resistance conferred by the QTL linked to SU91. Another
SCAR marker R7313, derived from the RAPD marker BC73 (also known as R7313), is
believed to be linked to the same QTL (Bai et al., 1997; Beattie et al., 1998; Miklas et al.,
2000). In this case, the resistance source was the tepary bean selection PI 440795,
however, the resistance among tepary beans may be similar (Bai et al., 1997).

The SCAR marker BC420, which is linked to a QTL for C88 resistanCe on bean
linkage group 86 (Miklas et al., 2000), was derived from the fragment amplified by the
RAPD marker BC420900 (Yu et al., 2000). In a recombinant inbred (RI) population
originating from the cross HR67/W1744d, the BC420 SCAR marker was responsible for
62% of the phenotypic variation (R2 = 0.62). Additionally, greenhouse results
determined that BC420 is 94.2% accurate in the identification of resistant genotypes (Yu
et al., 2000). Although the usefulness of this marker across different genetic populations
has not been previously examined, the original RAPD marker, BC420900, has been
studied. In three different genetic populations, this marker explained 8% to 30% of the

total phenotypic variation for CBB leaf resistance (Park et al., 1999). This range in

86

variation accounts for differences between the genetic populations used, the Xap isolates
used in screening, and both greenhouse and field studies. BC420900 was not associated
with pod resistance to C88 in any of the populations studied by Park et al. (1999).

The dominant SCAR marker SASl3, developed by Young et al. (1998) is linked
to the C 0-42 allele. This marker was derived from the OAS13950 RAPD fragment, both of
which are very tightly (0.39 cM) linked to C0—42 (Young et al., 1998). This resistance
allele was identified in the breeding line SEL 1308, and confers resistance to 33 of 34
races of C. lindemuthianum tested (Balardin and Kelly, 1998).

The primary objective of this research was to develop pyramided resistance to
common bacterial blight and anthracnose, based on the use of indirect selection of
various combinations of molecular markers. Selections were made for SCAR markers
linked to two QTL for CBB resistance a single anthracnose resistance gene. A second
objective was to evaluate the effects of the SCAR markers SU91 and BC420 on
resistance in novel genetic backgrounds, using field and greenhouse inoculations with
local isolates of X. axonopodis pv. phaseoli. The combination of the SU91 and BC420
markers in the same genotype was examined to determine its effectiveness in enhancing

CBB resistance compared to single QTL resistance sources.

87

MATERIALS AND METHODS

Parental Material and Population Development. A series of crosses were conducted,
with the purpose of combining in single genotypes, higher levels of resistance to
anthracnose and C88. One set of crosses resulted in 11 F1 progeny derived fi'om a cross
between HR45 and Kaboon. A second set of crosses involved 15 F5 black and navy bean
lines with SU91 crossed with either Phantom*Z/SEL 1308 or Jaguar*Z/SEL 1308 (donors
of the C0-42 anthracnose resistance gene). The resulting single cross F1 progeny from
these two sets of crosses were inter-mated to develop 40 four-way F1 populations, which
were self-pollinated to produce the F2 populations. Approximately one-half the seed
from each F2 population was planted (due to greenhouse space limitations), resulting in a
total of 432 F2 plants. Some F2 individuals were lost in the greenhouse due to late
maturity, genetic incompatibility, and failure to produce seed. A total of 320 F23 lines
were grown as single-row plots in the field (East Lansing) during the summer of 2003.
The one—row plots were space-planted and fertilized with 113 kg of 19-19-19, which was
banded at planting. The F3 lines were evaluated for agronomic characteristics (days to
flowering, lodging, re-greening and agronomic desirability). Days to flowering was
recorded as the number of days after planting that at least 50% of the plants in a plot had
a minimum of one open flower. Lodging was evaluated on a one to five scale: 1) no
lodging, 3) intermediate lodging, and 5) excessive lodging. Re-greening was recorded as
the presence (1) or absence (0) of re-greening at maturity. Desirability (DS) was
determined as a visual rating of overall agronomic desirability. This rating used a l to 9
scale with the following guidelines: 1-4 = above average agronomic traits, 5 = average

(compared to commercial checks), 6-9 = unacceptable agronomic traits. The F3 lines

88

were also inoculated and evaluated for CBB resistance. Inoculations were conducted as
described in Chapter One, on July 17, July 24, and August 6, 2003. The Xap isolate that
was used for inoculations was ‘9712-3’, an isolate from Nebraska that had originally been
isolated from infected bean seed grown in Michigan. Two CBB ratings (15 and 22 days
after the final inoculation) were recorded for the field experiment. Single plant selections
were made within the most desirable 93 of these 320 F3 lines based on agronomic traits,

marker data, and disease reaction.

Greenhouse Inoculations and Disease Evaluation. Greenhouse experiments were
conducted at Michigan State University. Plants were grown in 15.2 cm clay pots, with
sterile potting soil used as growth medium. Slow-release fertilizer was added to each pot
at the primary leaf stage. Plants were watered as needed. Inoculation experiments
consisted of four replications, each composed of 93 F3 genotypes selected from the 320
F2; 3 lines in the field in addition to four control genotypes of known phenotypic reaction
to C88 inoculation. The four control genotypes were present twice in each replication
(as inoculated and non-inoculated controls). This resulted in a total of 101 entries per
replication.

Two greenhouse inoculation experiments were conducted. Trifoliate leaves were
inoculated twice. The first inoculation was when all plants had at least one flower, and
the second was one week later. Flowering was delayed in these experiments due to slow
emergence. The first inoculation was at 65 and 66 days after planting (DAP) in the first
and second experiment, respectively, which was when the majority of plants had begun

flowering. Leaves were evaluated for visual CBB symptoms when they became

89

apparent. The leaf ratings for the first experiment were recorded at 7 days after the final
inoculation and at 14 days after the final inoculation for the second experiment. Two
pods were inoculated per plant with Xap, during the flat pod stage of pod development.
Pods were only inoculated once, and were evaluated 14 days after inoculation in the first
study. Ratings were delayed in the first study, due to slow disease development. Pods

were evaluated 9 days after the pod inoculation in the second experiment.

Polymerase Chain Reaction (PCR) and Molecular Marker Screening.

DNA extraction, PCR, and agarose gel electrophoresis were conducted as
previously described in Chapter One of this thesis. When multiplex PCR was used for
simultaneous screening of the SCAR markers SU91 and BC420, the thermal cycler
profile for a marker with an intermediate (55°C) annealing temperature (SAP6) was used.
This profile is described in Chapter One (Table 1.1).

Screening for the SCAR marker SASl3, linked to the Co-4‘ anthracnose
resistance gene, was conducted as previously described for SCAR markers. The PCR
profile used was as described by Young et al. (1998), and consisted of 34 cycles of: 10
seconds at 94°C, and 2 minutes and 40 seconds at 72°C. The reactions were completed
by a final step of 5 minutes at 72°C. The single resulting band, an amplified fragment of
approximately 950 base pairs (bp), was visualized as previously described using

fluorescence.

90

RESULTS

A total of 93 lines were selected from the 320 lines examined as F3 lines in a field
study at East Lansing (2003). These lines were selected on the presence of molecular
markers (SU91 and BC420), overall agronomic suitability, and low phenotypic ratings
for C88 in the field. A summary of the 93 lines selected is presented in Table 2.1.
These lines were screened for SAS13 marker, which is linked to the C0-4‘ anthracnose
resistance gene. All 93 resulting F4 lines were also evaluated for CBB resistance in the

greenhouse by direct inoculations.

Field Results — East Lansing (2003)

The 320 F3 lines in the field study consisted of 4 genotypic classes, based on the
presence or absence of molecular markers: SU91+/BC420+, SU9l+/-, -/BC420+, and -/-.
The effects of both markers (SU91 and BC420) were examined alone and in combination
for relationships with various phenotypic traits. The traits evaluated in this study
included: days to flowering, CBB leaf rating 1 (14 days after inoculation), CBB leaf
rating 2 (21 days after inoculation), C88 pod rating 1 (14 days after inoculation), CBB
pod rating 2 (21 days after inoculation), re-greening, lodging and overall agronomic
desirability score (DS).

Neither SU91 nor BC420 were significantly associated with C88 leaf or pod
resistance in the field study. The results of this study are summarized in Table 2.2.
These findings disagree with previous data from Chapter One, where presence of the

SU91 marker was associated with increased CBB resistance.

91

A relationship was noted, however between both markers and days to flowering.
Presence of the SU91 and BC420 markers were both negatively correlated with days to
flowering (r = —0.13* and —0.18** for SU91 and BC420, respectively). In combination,
SU91 and BC420 were also negatively correlated with days to flowering (r = -0.22**).
Data for days to flowering was presented in Appendix A4.

BC420 was associated (r = 015*) with re-greening of plants late in the growing
season. This trait complicates harvest and reduces the overall quality of potential bean
cultivars. SU91 was not, however, associated with re-greening in this study. Re-
greening data from this study was reported in Appendix A4. Because of extensive re-
greening in this experiment, actual maturity dates were not recorded, and therefore could

not be included in the analysis.

92

Table 2.1 Marker data, phenotypic CBB ratings on leaves and pods, and agronomic

desirability of 93 F3 selections from a field study established in East Lansing, 2003.

 

GenotypeiL SU9II BC420: Leaf1§ Leaf2§ Pod1§ Pod2§ D811

 

03T-9037 - -
03T-9056 - -
03T-9098 - -
03T-9 165 - -
03T-9291 - -
03T-9306 - -
03T-9 195 - +
03T-9 199 - +
03T-9204 +
03T-9019
03T-9026
03T—9048
03T-9049
03T-9058
03T-9061
03T-9062
03T-9067
03T-9076
03T-9079
03T-9082
03T-9092
03T-9093
03T-9099
03T-9104
03T-9l l l
03T-9l l3
03T-9l 15
03 T-91 l9
03T-9124
03T-9126
03T—9132
03T-9l 33
03 T-9 l 37
03T—9 l 39
03T-9142
03T-9 l 46
03T-9 l 48
03T-9153
03T-9154
03T—9156
03T-9 l 63
03T—9 1 66
03T-9 l 69
03T-9172

+++++++++++++++++++++++++++++++++++-

I I
wwwbwwwbut42.82854;wwwwwwrowwwmwwhmruwwwwwhwww—wbwmw
moat»AwwwbmwuwbAwwbwwwwwwwwwwwwawwwwwwwwwmwwww
_.._._.._._...__._.........._._.._.._.._.._....._._.._...._..._.._.._._..._._...._._._.._._._.._._........_.._._._.._.
—————————--—-——-—-—-—-—-——-—-—-—-—-—-—-—-—-—-——-—-N-——-——-—-—-—-——-—-—-—
MMMMilli/rmAMMMMMMMMQRMMMMMMMMMMMthMMQMMMAb-fih##-

93

Table 2.1 Continued.

 

Genotype'l SU9II BC4201 Leafl§ Leaf2§ Pod1§ Pod2§ D811

 

03T-9178
03T-9l 80
03T-9l 84
03T-9 l 85
03T-9l 86
03T-92 l 7
03T-924 l
03T-9245
03T-9247
03T-9250
03T-9259
03T-9263
03T-9267
03T-9268
03T-9277
03T-928 l
03T-9288
03T-9299
03T-93 l9
03T-9007
03T-901 3
03T-9029
03T-9036
03T-9046
03T-9050
03T-905 l
03T-905 5
03T-9059
03T-9060
03T-9090
03T-9 1 12
03T-91 16
03T-9128
03T-9 l 36
03T-9 l 4 l
03T-91 87
03T-9l 88
03T-9 l 89
03T-91 90
03T-919 l
03T-920 l
03T—9202
03T-9207
03T-9222

++++++++++++++++++++++++++++++++++++++++++++

wwwwwAwbwwwNNw-bwhbwhw—N—wwwwwwwwwawwwwwm
wwwwwwwwwLamb.AwwwwwwwwAAbANNNmmwm-bwwwwmmwwwww
._.._.._.._.._.._.._.._I._..—I._..—._.._.._.._.._a._.._.._.N__._._._._.o__._.___.____._.______
—-——————~————~—-—-—-—--—-—-—-—-—-—-—-—-—-—-—-—-—-—-I\J—-—-—-—-——-—-——-——-
L)!MMMO‘MUIMQMMML’I'JI'JIQO‘IMMMQ©MQMMM$MMMMMMAMMMMMMMM§

+++++++++++++++++++++++++u

WWWDJ

94

Table 2.1 Continued.

 

Genotype'l' SU9II BC42OI Leafl§ Leaf2§ Pod1§ Pod2§ DS11

 

03T-9269 + + 3 3 1 l 6
03T-9270 + + 4 4 1 l 6
03T-927l + + 3 4 l l 5
03T-9305 + + 2 2 1 1 5
03T-9309 + + 2 2 1 1 4

 

'1' = Indicates the F 3 genotype screened for C88 resistance using direct inoculations and molecular markers.

3; = SU91 and BC420 are SCAR markers linked to C88 resistance. “+” and “-“ indicate presence or
absence of the marker.

§ = Bean leaves and pods were rated twice for visual C88 symptoms. “Leafl” and “Leat2” refer to C88
leaf ratings, recorded 15 and 22 days after inoculation, respectively. Leaf ratings were recorded on a
scale of 1-9 (1= no symptoms, 9= severe symptoms. “Podl” and “Pod2” refer to C88 pod ratings,
recorded 15 and 22 days after inoculation, respectively. Pod ratings were recorded on a 0-3 scale
(0= no symptoms, 3= severe symptoms).

11 = DS represents overall agronomic desirability (1= superior, 5= average, 9= poor).

95

Table 2.2. Relationship of the SCAR markers SU91 and BC420 with leaf and pod

resistance in field inoculations of 320 F23 lines (East Lansing, 2003).

 

Field — Rating 1§

Field — Rating 2§

 

 

Marker‘l' CBB-LeafI CBB-PodI CBB-Leafi CBB-PodI
SU91+ 3.13 NS 1.02 NS 3.19 NS 1.01 NS
SU91 - 3.21 NS 1.03 NS 3.23 NS 1.04 NS
BC420 + 3.13 NS 1.02 NS 3.26 NS 1.04 NS
BC420 - 3.18 NS 1.02 NS 3.20 NS 1.02 NS
SU91+/BC420+ 3.17 NS 1.03 NS 3.23 NS 1.00 NS
SU91-[BC420- 3.22 NS 1.03 NS 3.24 NS 1.03 NS

 

NS = Not significant

1 = “SU91+” and “SU91-“ designate means for genotypes carrying or lacking the SU91 marker, linked to a
C88 resistance QTL mapped to bean linkage group 88.
“BC420+” and “BC420-“ designate means for genotypes carrying or lacking the BC420 marker, linked
to a C88 resistance QTL mapped to bean linkage group 86.

I = CBB-Leaf ratings were recorded on a 1-9 scale (1= no symptoms, 9 = severe symptoms).
CBB-Pod ratings were recorded on a 0-3 scale (0= no symptoms, 3= severe symptoms).

§ = Rating 1 and Rating 2 refer to visual CBB ratings taken 15 and 22 days after inoculation, respectively.

96

Greenhouse Results —- 2003

Greenhouse inoculations, which consisted of four replications of 101 (93 F3
selections and 8 controls) entries, were conducted as two separate experiments one week
apart. A summary of each entry included in the experiment is listed in Table 2.3. The
information included for each entry consists of: presence or absence of BC420 and
SU91, mean leaf rating (lSt experiment), mean pod rating (lSt experiment), mean leaf
rating (2nd experiment), and mean pod rating (2nd experiment). VAX 5, HR45, ‘Othello’
and ‘Midland’ are included as controls, because they have well characterized responses to
Xap infection. It is not known whether the resistant checks (VAX 5 and HR45) carry the
C0-4‘ gene for anthracnose resistance. The checks are included once as non-inoculated
controls, and once as inoculated controls for disease reaction.

SU91 was significantly associated with both leaf and pod resistance in greenhouse
experiments, although the correlation was not as strong as was observed in Chapter One.
The effects of this marker were evaluated both alone and in combination with the SCAR
marker BC420. The presence of SU91 alone was correlated with leaf resistance in
experiment 1 (r = 020*) and experiment 2 (r = -0.15**). Additionally, SU91 was
correlated with C88 pod resistance in experiment 1 (r = -0.l7*) and experiment 2 (r = -
0.27***). In the greenhouse, BC420 was associated with only pod resistance in
experiment 1 (r = -0.19*) and C88 leaf resistance in experiment 2 (r = -0.18**). A

summary of the statistical analysis of SU91 and BC420 for both greenhouse experiments
is presented in Table 2.4.
Despite the associations between CBB resistance (leaf and pod) and the SCAR

markers (SU91 and BC420), genotypes with both markers (SU91+/BC420+) did not

97

perform as expected. CBB leaf ratings for both experiments, and pod ratings for
experiment 2 were actually higher than ratings from genotypes with either marker alone
(SU91+/- or -/BC420+). A graphical interpretation of these results is shown in Figure

2.1.

98

Table 2.3. Molecular marker data and greenhouse inoculation results (2003-04). Data
were presented by genotype. All bean genotypes except controls are F 34 lines (93
entries) and were selected in the field (2003). C88 leaf and pod ratings were averaged

across 4 replications.

 

Genotype SU911' BC4201' SAS13'i LeaflI PodlI Leaf21 Pod21

 

03T-9056 - - - 3.0 4.0 4.5 2.3
03T-9098 - - - 2.3 1.0 2.3 2.7
03T-9291 - - - 2.5 1.0 2.0 2.0
O3T-9306 - - - 4.3 3.0 3.0 1.7
03T-9026 . . - 6.8 2.5 4.8 2.3
Midland§ - - N/A* 1.0 1.0 1.0 1.0
Midland1l - - N/A* 4.8 7.0 7.3 7.3
Othello§ - - N/A“ 1.0 1.0 1.0 1.0
Othello11 - - N/A“ 7.3 3.0 5.8 3.8
03T-9014 - - + 3.0 2.5 3.5 2.3
03T-9037 - - + 4.5 2.5 4.8 2.7
03T-9165 - - + 6.0 6.0 5.0 3.0
03T-9l95 - + - 2.8 1.0 2.5 1.0
03T-9204 - + - 4.3 6.0 2.8 1.3
03T-9l99 - + + 1.0 1.0 2.0 2.0
03T-9058 + - - 3.3 4.0 5.3 2.7
03T-9067 + - - 1.3 1.0 1.3 1.3
03T-9079 + - - 2.0 2.0 2.5 4.5
03T-9093 + - - 5.8 4.0 3.8 2.7
03T-9099 + - - 3 .0 1.0 2.5 3.0
03T-9113 + - - 2.0 2.5 2.8 2.3
03T-9119 + - - 1.5 2.0 2.3 2.3
03T-9l32 + - - 2.0 1.5 1.3 2.3
03T-9146 + - - 4.5 6.5 5.0 3.5
03T-9148 + - - 5.8 5.0 5.8 3.0
03T-9153 + - - 2.5 3.0 3.0 3.0
03T-9166 + - - 2.5 1.0 2.0 1.3
03T-9169 + - - 1.7 1.0 2.8 2.0
03T-9l72 + - - 1.5 2.5 2.8 2.0
03T-9184 + - - 4.0 5.0 4.3 2.0
03T-9241 + - - 3.0 3.0 2.8 2.3
03T-9247 + - - 1.5 4.5 2.3 4.0
03T-9250 + - - 2.0 4.0 1.8 1.5
03T-9263 + - - 1.5 1.0 2.0 2.0
03T-9267 + - - 2.3 1.0 3.0 2.0
03T-9277 + - - 1.0 2.0 1.5 1.5
03T-9281 + - - 1.5 5.0 2.0 1.7
03T-9288 + - - 3.7 1.0 4.0 3.3

99

Table 2.3 Continued.

 

Genotype SU911’ BC4201‘ SASl31‘ LeaflI PodlI LeaflI Pod2§:

 

03T-9299 + - - 1.3 1.0 1.3 2.0
03T-9319 + - - 3.3 7.0 2.7 2.0
03T-9019 + - + 1.8 5.0 2.5 2.3
03T-9048 + - + 1.8 2.5 2.0 1.0
03T-9049 + - + 1.3 1.0 3.0 1.5
03T-906l + - + 1.3 1.5 2.5 2.0
03T-9062 + - + 1.0 2.5 2.7 2.0
03T-9076 + - + 3.3 1.5 2.8 1.5
03T-9082 + - + 2.0 2.0 3.0 1.0
03T-9092 + - + 1.3 . 3.3 3.0
03T-9111 + - + 1.7 1.0 4.0 1.0
03T-9115 + - + 4.3 1.5 5.0 1.3
03T-9124 + - + 3.3 2.0 2.0 1.0
03T-9126 + - + 3.0 2.0 3.0 1.0
03T-9133 + - + 6.8 . 5.0 1.5
03T-9137 + - + 2.3 2.0 2.3 1.0
O3T-9139 + - + 1.8 3.0 1.8 1.7
03T-9142 + - + 3.5 2.5 3.0 2.0
03T-9154 + - + 4.5 3.5 6.5 2.7
03T-9156 + - + 1.3 6.0 1.8 1.3
03T-9163 + - + 3.3 9.0 2.0 2.0
03T-9178 + - + 2.3 1.0 2.0 2.3
03T-9l80 + - + 2.0 3.5 2.3 1.5
03T-9185 + - + 1.5 2.0 2.5 2.3
03T-9217 + - + 2.3 3.0 3.0 2.5
03T-9245 + - + 1.5 2.0 2.3 1.7
03T-9259 + - + 2.0 3.0 2.8 1.3
03T-9268 + - + 2.0 2.0 2.3 1.0
VAX 51] + - N/A* 1.0 1.5 2.0 1.5
VAX 5§ + - N/A“ 1.0 1.0 1.0 1.0
O3T-9007 + + - 2.7 1.5 4.8 3.0
03T-9036 + + - 2.8 3.0 4.3 1.0
03T-9050 + + - 1.8 1.0 4.8 2.3
03T-9090 + + - 3.0 1.0 2.5 1.7
03T-9112 + + - 1.3 3.0 2.8 2.3
03T-9116 + + - 2.5 1.0 2.8 1.7
O3T-9128 + + - 2.8 2.0 3.3 1.5
03T-9136 + + - 3.0 4.5 3.7 3.3
03T-9l41 + + - 1.3 1.0 2.3 1.0
03T-9187 + + - 2.5 4.5 6.0 2.7
03T-9l88 + + - 4.3 1.0 6.7 1.7
O3T-9202 + + - 4.0 4.5 5.5 2.0
03T-9222 + + - 2.3 4.0 4.0 2.0
03T-9271 + + - 3.7 1.5 4.5 4.3
03T-9305 + + - 2.8 1.0 3.0 1.5

100

Table 2.3 Continued.

 

Genotype SU91'1' BC4201 SASl3‘l LeaflI PodlI LeafZI Pod2I

 

O3T-9309 + + - 1.5 1.0 2.0 1.0
03T-9013 + + + 2.3 1.0 4.0 2.0
03T-9029 + + + 4.0 1.0 3.8 1.5
03T-9046 + + + 3.3 3.5 4.0 1.5
03T-9051 + + + 4.3 2.0 3.5 1.0
03T-9055 + + + 3.5 1.0 2.8 1.5
03T-9059 + + + 4.0 3.0 5.0 3.0
03T-9060 + + + 1.3 1.5 4.0 1.7
031-9186 + + + 5.7 4.0 5.5 3.0
03T-9l89 + + + 1.5 2.0 2.5 1.0
03T-919O + + + 5.0 3.0 5.3 3.0
03T-919I + + + 1.7 3.0 6.0 2.7
03T-9201 + + + 6.3 1.0 3.3 1.7
03T—9207 + + + 3.5 1.0 3.0 2.0
03T-9269 + + + 2.0 2.0 2.7 1.3
03T-9270 + + + 4.0 4.0 4.8 3.7

HR45§ + + N/A* 1.0 1.0 1.0 1.0

HR451| + + N/A“ 1.3 2.0 2.5 1.5

 

" = N/A indicates “not available.”
'I' = Presence or absence of SCAR markers is reported as “+” and “-“, respectively.
SU91 = SCAR marker linked to C88 resistance QTL on bean linkage group 88.
BC420 = SCAR marker linked to C88 resistance QTL on bean linkage group 86.
SAS 1 3 = SCAR marker linked to Co-4‘ anthracnose resistance gene on bean linkage group 88.

I = Leaf 1 and Leaf2 indicate visual CBB leaf ratings recorded 7 and 14 days after inoculation,
respectively. Leaves were rated for C88 symptoms on a 1-9 scale (1= no symptoms, 9= severe
symptoms).

Podl and Pod2 indicate visual CBB pod ratings recorded 7 and 14 days after inoculation,
respectively. Pods were rated for CBB symptoms on a 0-3 scale (0= no symptoms, 3=severe
symptoms).

§ = Designates resistant (VAX S and HR45) and susceptible (Othello and Midland) controls inoculated
with water.

11 = Designates resistant (VAX 5 and HR45) and susceptible (Othello and Midland) controls inoculated
with Xap.

101

Table 2.4. Correlation coefficients of the SCAR markers SU91 and BC420 with leaf and

pod resistance in two greenhouse inoculation experiments with 101 entries.

 

 

 

 

Greenhouse — Experiment 1 Greenhouse - Experiment 2
Marker't CBB-Leaf: CBB-Pod: CBB-LeafI CBB-PodI
SU91 -0.1979** -0.1724* -0.1477** -0.2713***
BC420 0.0600 NS -0.1924* -0.1833** -0.0592 NS

 

NS, 1‘, **, *** = non-significant, significant at the 5%, 1%, and 0.1% probability levels, respectively.

1“ = SU91 is a SCAR marker linked to a C88 resistance QTL on bean linkage group 88.
BC420 is a SCAR marker linked to a C88 resistance QTL on bean linkage group 86.

I = CBB-Leaf and CBB-Pod refer to visual CBB ratings based on a 1-9 scale for leaves (1= no symptoms,
9: severe symptoms) and a 0-3 scale for pods (O= no symptoms, 3= severe symptoms). Correlation
coefficients (r) are presented between the SCAR markers and the visual disease ratings.

102

 

El Leaves
I Pods

 

 

 

 

Figure 2.1. A comparison of the mean CBB leaf and pod ratings for various molecular
marker combinations. Results are presented as an average of two greenhouse
experiments that exhibited similar conclusions. Mean ratings are averages of 4

replications, determined using 1-9 scales for leaves and pods (1= no symptoms, 9= severe

symptoms).

103

Table 2.5. A comparison of the mean CBB leaf and pod ratings for various molecular
marker combinations. Results are presented as an average of two greenhouse
experiments that exhibited similar conclusions. Mean ratings are averages of 4
replications, determined using 1-9 scales for leaves and pods (1= no symptoms, 9: severe

symptoms). Separation of means, designated by different letters was determined using

 

 

LSDocs.
Marker Combination Leaves Ms
-/- 4.4a 3.7a
SU91+/- 2.6b 2.3b
-/BC420+ 2.5b 1.8b
SU9l+/BC420+ 3.4c 2.0b

 

104

Screening for the SCAR marker SASl3. The 320 F3 lines evaluated in the field were
also screened for the SCAR marker SAS13, linked to the anthracnose resistance gene C0-
42. The results of screening the 93 lines selected based on CBB resistance for SASl3
(described above) are presented in Table 2.3.

Of the 93 lines selected for CBB resistance, 45 (48%) carried the SAS13 marker,
while the remaining 48 did not. The 45 lines that carried SASl3 were grouped into the
following categories of molecular marker combinations (illustrated in Figure 2.2):
SU91+/-/SASl3+ (26 lines), -/8C420+/SAS13+ (1 line), -/-/SASl3+ (3 lines),

SU91+/8C420+/SASI3+ (15 lines).

105

 

 

30
25
20
15
10 —

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Number of Lines

 

 

 

 

 

 

 

 

 

 

 

 

 

I l I I I

1 2 3 4 5 6

Marker Combination

O czar—I‘m]
7

 

 

 

Figure 2.2. A summary of the molecular marker composition of the 93 F3 lines selected
from the field (East Lansing, 2003). SU91 is a SCAR marker linked to a C88 resistance
QTL on bean linkage group 88. BC420 is a SCAR marker linked to a C88 resistance
QTL on bean linkage group 86. SASl3 is linked to the Middle American anthracnose
resistance gene Co-42. Marker combinations (1-8) correspond to the following: 1: -/-/-,
2= SU91+/-/-, 3= SU91+/BC420+/-, 4= SU91+/-/SAS13+, 5= -/BC420+/SAS13+, 6= -

/BC420+/-, 7= -/-/SASl3+, 8= SU91+/BC 420+/SASl3+.

106

DISCUSSION

The potential of combining multiple sources of resistance to C88 and
anthracnose was examined using a system of indirect selection in common bean. The
combination of resistance to both diseases in a common genetic background would
facilitate the development of commercial bean cultivars with an improved spectrum of
disease resistance. Since resistance to C88 in common bean is controlled quantitatively,
the level of resistance conferred by a single QTL is partial (Nodari et al., 1993; Eskridge
and Coyne, 1996; Jung et al., 1996). Because of the availability of more molecular
markers linked to quantitative traits, the pyramiding of multiple QTL is now possible.
Pyramiding using direct screening was not previously feasible, due to epistatic
interactions that prohibited the ability to distinguish the effects of distinct genetic loci
(Ribaut and Hoisington, 1998). Development of breeding lines with pyramided QTL
from various resistance sources could help increase the levels of partial resistance beyond
that conferred by a single QTL (Castro et al., 2003).

A group of F3 lines with differing combinations of the C88 resistance QTL
SU91 and BC420 were evaluated for resistance to C88 in the field. Neither marker was
associated with increased levels of C88 resistance in the field, which was unexpected,
based on the results of previous studies (Chapter One). The results obtained in this field
experiment were possibly due to inadequate establishment of disease (infection levels
were low even in susceptible checks). Very little variation was observed in C88 ratings
(most notably in pods), which may be a reason why it was difficult to discriminate
between resistance and susceptibility. This field study was space planted to facilitate

selection, which results in a lower planting density. The greater distance between plants

107

may have played a role in the inability to establish CBB pressure in this experiment, as
the movement of bacteria from plant to plant may have been reduced in comparison to
higher planting densities. The mean CBB leaf rating in this field experiment was only
3.05, on a scale of 1-9, and the maximum CBB leaf rating was 7.0. It is possible that the
level of infection was not sufficient to effectively discriminate between susceptible and
resistant plants.

The potential for linkage between CBB resistance and agronomic traits was also
studied in the field experiment. Negative correlations between both markers for C88
resistance (SU91 and BC420) and flowering were observed in this experiment, which
agreed with previous reports of delayed maturity linked to C88 resistance (Tar’an et al.,
2001). BC420 was also associated with delayed maturity in this study. The effect of
SU91 and BC420 on yield was not determined, as yield data was not available for this
experiment, because of insufficient seed for replications. Based on previous results
(Chapter One), it seems likely that early-maturing, high-yielding, CBB resistant lines can
be recovered from this selection program, despite the negative associations with
resistance loci.

The results of the tWo greenhouse direct inoculation experiments contrasted the
field study. SU91 was correlated with enhanced levels of C88 resistance in leaves and
pods in both greenhouse experiments. The presence of BC420 was also correlated with
C88 resistance in leaves (experiment two) and pods (experiment one). Previous
experiments have revealed that BC420 confers differing levels of resistance in pods and
leaves (with pod resistance being higher). This may account for the differing results of

lines carrying BC420 in greenhouse experiments. When the markers were combined

108

within a single genotype, however, the resistance levels were intermediate to the
resistance conferred by either marker alone and that of genotypes with no known
resistance loci. The difficulty in establishing a consistent method of evaluating common
bean for resistance to C88 using direct inoculations has been documented (Singh and
Munoz, 1999). The unreliability of direct inoculations could be one reason why the
results obtained in this study were so unexpected. In the 1" greenhouse experiment, the
susceptible check ‘Midland’ was rated a mean CBB leaf rating of only 4.75 (using the
previously described 1-9 scale), although the other susceptible check ‘Othello’ was rated
a mean CBB leaf rating of 7.3. Additionally, the mean CBB leaf rating of greenhouse
experiment 1 was only 2.8. In the 2nd greenhouse experiment, which was conducted
under similar conditions, ‘Midland’ had a mean CBB leaf rating of 7.25, while ‘Othello’
received a mean rating of 5.75. The mean CBB leaf rating for the 2nd greenhouse
experiment was 3.3. These results illustrate the difficulty of establishing high levels of
disease pressure using inoculations with the bacterium. There is variation in disease
reaction even within the cultivars included as checks, possibly due to fluctuations in
greenhouse environmental conditions.

The presence of both markers in this genetic background does not appear to be
beneficial, based on the results obtained in the greenhouse. The potential failure of QTL
pyramiding in this particular situation warrants further investigation. The potential
benefits of pyramiding resistance loci have been previously documented, contradicting
the results obtained in this research (Castro et al., 2003). QTL pyramiding allows
breeders to combine the genetic components of a quantitative trait and have a larger

impact on a phenotype that can be achieved by use of a single QTL. Future field

109

 

experiments need to be conducted to gain additional information regarding the apparent
failure of QTL pyramiding in this study.

The results of the greenhouse inoculations provide further evidence that the field
results from this study are less reliable, as SU91 was significantly associated with C88
resistance in leaves and pods (in two separate greenhouse experiments). SU91 was also
associated with C88 resistance in previous field studies (Chapter One), also contrasting
the field results in this experiment. Further greenhouse and replicated field evaluations
using various isolates in different environmental conditions will be needed to confirm the
C88 resistance potential of these 93 lines.

In conclusion, 93 F4 lines with various combinations of resistance to common
bacterial blight and anthracnose, were generated from this study. The molecular marker
combination present in each genotype was determined, and the disease reaction of each
genotype to direct inoculation with Xap was observed. Although it would be useful to
confirm the results of MAS by screening the selected lines for anthracnose resistance, no
races of the pathogen are currently available in our laboratory that would discriminate the
effects of Co-I‘ and Co-4‘.

The selections derived from this material may be useful to introgress pyramided
C88 and anthracnose resistance into other market classes of beans. No commercially
available cultivar in any market class has adequate levels of C88 resistance for
production in Michigan. Introgressing CBB resistance into any of the major market class

of common beans would be beneficial.

110

REFERENCES

Ariyarathne, H.M., D.P. Coyne, G. Jung, P.W. Skroch, A.K. Vidaver, J.R. Steadman,
P.N. Miklas, and M.J. Bassett. 1999. Molecular mapping of disease resistance
genes for halo blight, common bacterial blight, and bean common mosaic virus in
a segregating population of common bean. J. Amer. Soc. Hort. Sci. 124:654-662.

Amaud-Santana, 8., DP. Coyne, KM. Eskridge, and AK. Vidaver. 1994. Inheritance;
low correlations of leaf, pod, and seed reactions to common blight disease in
common beans; and implications for selection. J. Amer. Soc. Hort. Sci.
119:116-121.

Bai, Y., TE. Michaels, and KP. Pauls. 1997. Identification of RAPD markers linked to
common bacterial blight resistance genes in Phaseolus vulgaris L. Genome
40:544-551.

Balardin, RS, and JD. Kelly. 1998. Interaction between Colletotrichum lindemuthianum
races and gene pool diversity in Phaseolus vulgaris. J. Amer. Soc. Hort. Sci.
123:1038-1047.

Beattie, A., TE. Michaels, and KP. Pauls. 1998. An efficient, reliable method to screen
for common bacterial blight (CBB) resistance in Phaseolus vulgaris L. Ann.
Rept. Bean Improv. Coop. 41:53-54.

Castro, A.J., X. Chen, A. Corey, T. Filichkina, P.M. Hayes, C. Mundt, K. Richardson, 8.
Sandoval—Islas, and H. Vivar. 2003. Pyramiding and validation of quantitative
trait locus (QTL) alleles determining resistance to barley stripe rust: effects on
adult plant resistance. Crop Sci. 43:2234-2239.

Coyne, DP, and ML. Schuster. 1974. Inheritance and linkage relations of reaction to
Xanthomonas phaseoli (E. F. Smith) Dowson (common blight), stage of plant
development and plant habit in Phaseolus vulgaris L. Euphytica 23:195-204.

Drij flrout E. and W.J. Blok. 1987. Inheritance of resistance to Xanthomonas campestris
pv. phaseoli in tepary bean (Phaseolus acutifolius). Euphytica 36:803-808.

Eskridge, KM. and DP. Coyne. 1996. Estimating and testing hypothesis about the
number of genes using inbred-backcross data. J. Hered. 87:410-412.

Jung, G., D.P. Coyne, P.W. Skroch, J. Nienhuis, E. Arnaud-Santana, J. Bokosi, H.M.
Ariyarathne, J .R. Steadman, J .S. Beaver, S.M. 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.

111

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, ID. and LO. Copeland. 1996. Status of common bacterial blight problems in
Michigan dry beans. Pages. 173-177. ler. Taller Interacional sobre bacteriosis
somun del frijol. Oficina de Coordinacion Regional de Profiijol, Cesda, San
Cristobal, Republica Domincana.

McElroy, J .8. 1985. Breeding dry beans, P. vulgaris L., for common bacterial blight
resistance derived from Phaseolus acutifolius A. Gray. Ph.D. Cornell University,
Ithaca, NY.

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

Miklas, P.N., J .8. Smith, R. Riley, K.F. Grafton, S. Singh, C. Jung, and DP. Coyne.
2000. Marker-assisted breeding for pyramided resistance to common bacterial
blight in common bean. Ann. Rept. Bean Improv. Coop 43:39-40.

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

Park, SJ. and 8N. Dhanvantari. 1994. Registration of common bean blight-resistant
germplasm, HR45. Crop Sci. 34:548.

Park, 80., DP. Coyne, N. Mutlu, G. Jung, and J. Steadman. 1999. Confirmation of
molecular markers and flower color associated with QTL for resistance to
common bacterial blight in common beans. J. Amer. Soc. Hort. Sci.

124:519-526.

Paterson, AH. 1995. Molecular dissection of quantitative traits: progress and prospects.
Genome. 5:321-333.

Pedraza, F ., G. Gallego, S. Beebe, and J. Tohme. 1997. Marcadores SCAR y RAPD para
la resistencia a la bacteriosis comun (CBB). Pages 130-134. In: Taller de
Mejoramiento de F rijol para el Siglo XXI: Bases para Una Estrategia para
America Latina. S. P. Singh and O. Voysest (ed.). CIAT, Cali, Colombia.

Ribaut, J. and D. Hoisington. 1998. Marker-assisted selection: new tools and strategies.
Trends in Plant Science. 32236-239.

112

Silva, L.O., S.P. Singh, and MA. Pastor-Corrales. 1989. Inheritance of resistance to
bacterial blight in common bean. Theor. Appl. Genet. 78:619-624.

Singh, SP, and CG. Munoz. 1999. Resistance to common bacterial blight among
Phaseolus species and common bean improvement. Crop Sci. 39:80-89.

Tar’an, 8., TE. Michaels, and KP. Pauls. 2001. Mapping genetic factors affecting the

reaction to Xantlzomonas axonopodis pv. phaseoli in Phaseolus vulgaris L. under
field conditions. Genome 44:1046-1056.

Young, RA. and J .D. Kelly. 1996. Characterization of the genetic resistance to

Colletotrichum Iindemuthianum in common bean differential cultivars. Plant
Dis. 80:650-654.

Young, RA. and JD. Kelly. 1997. RAPD markers linked to three major anthracnose
resistance genes in common bean. Crop Sci. 37:940-946.

Young, R.A., M. Melotto, R.O. Nodari, and J .D. Kelly. 1998. Marker-assisted dissection
of the oligogenic anthracnose resistance in the common bean cultivar, ‘G 2333’.
Theor. Appl. Genet. 96:87-94.

Yu, K., S.J. Park, and V. Poysa. 2000. Marker-assisted selection of common beans for
resistance to common bacterial blight: efficacy and economics. Plant Breeding
119:411-415.

113

 

APPENDIX Al

COLLECTION OF Xanthomonas axonopodis pv. phaseoli (Xap)
ISOLATES FROM MICHIGAN DRY BEAN FIELDS

ABSTRACT

Twenty-four samples of common bacterial blight (CBB) infected bean leaf tissue
were collected from two Michigan State University Agricultural Experiment Station
research farms in Saginaw and Montcalm counties, respectively. Xanthomonas
axonopodis pv. phaseoli (Smith) Dye (Xap) cultures were isolated fiom 19 of 24 samples
and plated on MXP media (Claflin et al., 1987). Isolates showed variability in growth
habit (grainy vs. mucoid) and the production of a yellow polysaccharide suspected of
being associated with virulence (Corey and Starr, 1957; Ekpo and Saettler, 1976;
Steadman et al., 2002). Additional Xap isolates were obtained fiom other Universities
throughout the United States. These cultures were maintained for the eventual
characterization of relative virulence. The identification of virulent Michigan isolates

would facilitate the efficient screening of bean genotypes for CBB resistance.

114

 

INTRODUCTION

Common bacterial blight (C88) is a seed-home disease that affects common bean
(Phaseolus vulgaris L.) crops worldwide. The causal agent, Xanthomonas axonopodis
pv. phaseoli (Smith) Dye (Xap), is a gram-negative bacteria (Saettler, 1991). Due to the
ineffectiveness of chemical and cultural controls, the development of genetic resistance is
the preferable method of preventing CBB epidemics. The variability of Xap has serious
implications in the development of a successful CBB resistance screening protocol. The
use of molecular marker-assisted selection (MAS) for disease resistance is a promising
method of identifying resistant genotypes (Yu et al., 2000). When using molecular
markers developed by other breeding programs, however, the markers must be evaluated
to ensure their usefulness in genetic populations different from those in which the marker
was developed (Dudley, 1993). Additionally, because bacterial isolates used in
inoculations may differ between breeding programs, putative resistant plants should be
tested for resistance to local Xap isolates. This would help ensure the potential of novel
resistance sources to confer resistance levels suitable for the desired area of bean
production.

Isolates of Xap differ greatly in their production of a yellow polysaccharide,
ranging from bright yellow to white in color. Those isolates producing copious amounts
of the yellow polysaccharide are referred to as ‘Fuscans’ isolates. This polysaccharide
has been implicated in playing a role in the relative virulence of an isolate (Corey and
Starr, 1957). Additional researchers also observed higher virulence of ‘Fuscans’ isolates
compared to ‘Non-Fuscans’ isolates (Ekpo and Saettler, 1976; Steadman et al., 2002).

When the two types of isolates were used in direct inoculations on a set of C88 resistant

115

’

bean breeding lines, the plants were consistently more virulent to isolates of the ‘Fuscans
category.

The objective of this research was to obtain a collection of Xap isolates for use in
screening potentially CBB resistant materials, with the purpose of developing resistant
varieties for production in Michigan. Recent establishment of a Xap greenhouse
inoculation procedure will allow the future screening of these isolates for relative

virulence.

116

MATERIALS AND METHODS

CBB infected plant tissue was collected from the field for the purpose of isolating
Xap for use in field and greenhouse inoculations. Twelve samples were collected from
diverse genotypes, locations, and market classes at the Saginaw Valley Bean and Beet
Farm. An additional 12 samples were collected at the Michigan State University
Montcalm Potato Research Station.

The protocol used for bacterial isolation was as described by Claflin et al. (1987),
with some minor modifications. All isolates collected from Saginaw and Midland,
Michigan, and Washington State were recovered using the method previously described
in Chapter One. The isolates received from Colorado State University (8434, B439,
B455, and 845 8) were sent as active bacterial cultures in petri plates with YDC media.
The isolate from The University of Nebraska (9712-3) was sent as a lyophilized culture.
Prior to use, approximately 0.008 grams of the stock sample was rehydrated for 10-20

minutes on YDC media, using 50 pl of sterile water.

117

RESULTS

Cultures of Xap isolates were extracted from 24 samples of disease plant tissue.
These isolates, which were collected from infected dry beans at Saginaw and Montcalm,
varied greatly in colony growth habit and color. Samples were cultured on the semi-
selective media MXP (Claflin et al., 1987). Viable bacterial cells were recovered from
19 of the 24 samples. Six additional samples were received from other locations, and
were cultured according to the procedure described above. F our of these isolates are
from Colorado State University. One sample was also collected from a C88 infested dry
bean field in Washington. A final isolate was received fi'om The University of Nebraska,
however, it was isolated from CBB infected seeds of the pinto cultivar ‘Tomahawk’
originating from a Michigan bean field. Collection details and notes regarding isolate
characteristics are summarized in Table A1.1.

Isolates collected were predominately of the F uscans-type colonies, with mucoid
growth and copious production of a yellow polysaccharide. Of the 20 samples from
which live bacterial cultures were isolated, 15 yielded isolates of the F uscans-type. The
other samples resulted in the recovery of bacteria that were either Non-Fuscans (white

and grainy) or intermediate in growth.

118

Table A1.1 Summary of information regarding Xap isolates collected from infected bean

tissue at Saginaw and Montcalm research farms, in Michigan, in 2002.

 

 

Sample‘i Sourcei Pigment§ Notes
81 G00540 F uscans Runny
S2 G97905 Unknown Did not survive culture
S3 802581 Fuscans Turned media brown
S4 802525 Fuscans Turned media brown
S5 N01504 Unknown Did not survive culture
S6 N02233 Unknown Did not survive culure
S7 N00810 Intermediate Possibly two distinct isolates
88 802554 F uscans Bright yellow, runny
S9 102522 F uscans Dull yellow, turned media brown
810 802540 Fuscans Gradient, yellow to dull yellow
Sll P00233 F uscans Bright, viscous
S 12 P99120 F uscans Runny
Ml Montcalm Non-Fuscans White, grainy
M2 Montcalm Unknown Did not survive culture
M3 Montcalm Fuscans Grainy, similar to non-fuscans
M4 Montcalm Fuscans Bright, viscous, turned media black
M5 Montcalm Unknown Did not survive culture
M6 Montcalm Fuscans Thick growth
M7 Montcalm Non-Fuscans Off-white, grainy
M8 Montcalm Intermediate Growth similar to non-fuscans
M9 Montcalm F uscans Bright, shiny yellow
M10 Montcalm Fuscans Turned media dark
M11 Montcalm Intermediate Growth similar to non-fuscans
M12 Montcalm Fuscans Grainy, similar to non-fuscans
8434 Azufrado Fuscans Shiny, collected from Colorado
8458 Bill Z Fuscans Shiny, collected from Colorado
8439 Burke Fuscans Shiny, collected from Kansas
8455 Enola Fuscans Shiny, collected from Nebraska
W1 Washington F uscans Collected from WA state
9712-3 Tomahawk F uscans Received from Nebraska, collected from MI

 

1’ = Isolates designated with “S” were collected from the Saginaw Valley Bean and Beet Research Farm.
Isolates designated with “M” were collected from the Montcalm County Research Station.
Isolates designated with “B” were received from Colorado State University.

1; = The variety from which the isolate was collected is presented, when available. Information regarding
the variety from which an isolate was collected is unavailable for the isolates from Montcalm and

Washington.

§ = “Fuscans” isolates produce a yellow pigment on media, “Non-Fuscans” isolates are off-white in color.

119

DISCUSSION AND POSSIBLE FUTURE RESEARCH

The development of C88 resistant cultivars requires the establishment of a
successful disease resistance screening procedure, using isolates that will adequately
identify genotypes with potential CBB resistance. A basic characterization of the isolates
collected in this research will facilitate the development of an optimal CBB screening
protocol. The isolates described above reflect a sample of local Xap populations present
in infected Michigan bean fields. The identification of a small number of virulent
isolates would be useful in both greenhouse and field screening. In combination with
MAS, this would increase the efficiency of selection by reducing the frequency of
susceptible genotypes.

Comparison of isolate relative virulence would be most easily studied under
controlled greenhouse conditions. Conducting a study of this nature in the field could
result in complications due to bacterial movement from weather events. Greenhouse
inoculations allow the relative containment of bacterial isolates, provided that plants
inoculated with different isolates are not grown immediately adjacent to each other. A
study of this nature was not initiated as planned, due to unforeseen complications in the
development of a reliable greenhouse screening protocol. The virulence of these isolates
should be evaluated and compared on a small number of known highly resistant (VAX 5
and XAN 159) and highly susceptible (‘Midland’ and ‘Othello’) genotypes is a possible
direction of future research. Virulent isolates identified from such an analysis would be

useful for direct inoculations in the field and greenhouse.

120

REFERENCES

Claflin, L.E., A.K. Vidaver, and M. Sasser. 1987. MXP, a semi-selective medium for
Xanthomonas campestris pv. phaseoli. Phytopathology 772730-734.

Corey, RR, and MP. Starr. 1957. Colony types of Xanthomonas phaseoli. J. Bact.
74:137-140.

Dudley, J .W. 1993. Molecular markers in plant improvement: Manipulation of genes
affecting quantitative traits. Crop Sci. 33:660-668.

Ekpo, 8.1., and AW. Saettler. 1976. Pathogenic variation in Xanthomonas phaseoli and T
X. phaseoli var. fuscans. Plant Dis. Reptr. 60:80-83.

Saettler, AW. 1991. Diseases caused by bacteria. In: R. Hall (ed). Compendium of bean
diseases. APS Press: St. Paul. p 29-32.

 

Steadman, J .R., M.A. Pastor-Corrales, and J .8. Beaver. 2002. An overview of the 3rd bean t.
Rust and 2nd bean common bacterial blight international workshops, March 4-8,
2002 Pieterrnaritzburg, South Afiica. Ann. Rept. Bean Improv. Coop. 45:52-53.

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

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

121

APPENDIX A2

CHARACTERIZATION OF RESISTANCE TO COMMON
BACTERIAL BLIGHT (CBB) BETWEEN THE KIDNEY BEAN
CULTIVARS ‘MONTCALM’ AND ‘REDHAWK’

ABSTRACT

Observations in the field have suggested that differences in resistance to common

 

bacterial blight (CBB) exist between the kidney bean cultivars ‘Montcalm’ and
‘Redhawk’. DNA was extracted from these cultivars and screened for previously
described molecular markers linked to C88 resistance. Only the SCAR marker SAP6
was polymorphic between these two cultivars. Greenhouse inoculations with the causal
agent Xanthomonas axonopodis pv. phaseoli (Smith) Dye (Xap) were used to examine
the possible differences in resistance reported from field observations. Greenhouse
inoculations confirmed that ‘Montcalm’ does have significantly higher resistance to C88
(p= 0.0067) than ‘Redhawk’. This has possible implications for kidney bean growers
who have experienced problems due to C88 infections and are seeking cultivars with

higher levels of resistance to C88.

122

INTRODUCTION

Common bacterial blight (C88) is a destructive seed-bome disease of common
bean, caused by the bacterial pathogen Xanthomonas axonopodis pv. phaseoli. Most
cultivars grown commercially in Michigan display no resistance to C88. Such cultivars
would be highly beneficial to bean growers and seed producers. Several Michigan bean
farmers have noted a difference in field resistance between the kidney bean cultivars
‘Montcalm’ and ‘Redhawk’ (Dr. James Kelly, personal communication).

The purpose of this research was to examine the level of C88 resistance in
‘Montcalm’. Of the four SCAR markers linked to C88 resistance that were screened
(BC409, BC420, SAP6, and SU91), only SAP6 was polymorphic between the two
cultivars. This marker is linked to a QTL for CBB resistance that has been mapped to
bean linkage group 810 (Miklas et al., 1998; Miklas et al., 2000). This QTL was
identified in the C88 resistant breeding line GNN #1 Sel. 27, with the resistance

originating from within Phaseolus vulgaris (Miklas et al., 2003).

123

 

MATERIALS AND METHODS

DNA was extracted from ‘Montcalm’ and ‘Redhawk’ cultivars according to the
protocol described in Chapter One of this text. Both cultivars were screened with the
following SCAR markers linked to resistance loci for CBB: BC420, SAP6, BC409, and
SU91. The PCR profile of these markers is listed in Table 1.1.

Greenhouse inoculations were used to compare the level of C88 resistance
between the two cultivars. Eight plants of each cultivar were planted in 6-inch pots.
Inoculations followed the protocol described in Chapter One. Visual ratings were taken
on three leaves per plant at 14 days after inoculation. The previously described 1-9 scale
(1= no infection, 9: severe infection) was used for comparisons of C88 resistance in this

study. Statistical analysis was conducted using the PROC GLM function of SAS.

124

RESULTS AND DISCUSSION

The SAP6 primer amplified a band that was polymorphic between the cultivars
‘Montcalm’ and ‘Redhawk’. Because SAP6 is linked to a major C88 resistance QTL on
810 (Miklas et al., 2000), it would be expected that ‘Montcalm’ would display higher
resistance to C88.

Greenhouse inoculations with X. axonopodis pv. phaseoli confirmed higher levels
of resistance in ‘Montcalm’, compared to ‘Redhawk’. The mean leaf rating for
‘Montcalm’ was 3.0, compared to 4.5 in ‘Redhawk’. The p-value for this comparison
was 0.0067. SAP6 was responsible for 17.5 % of the total phenotypic variation in C88
leaf ratings.

This research suggests that higher levels of C88 resistance are present in the
kidney bean cultivar ‘Montcalm’, compared to ‘Redhawk’. It would appear that this
resistance is due to the presence of the C88 resistance QTL linked to the SCAR maker
SAP6, on bean linkage group 810. One of the Great Northern Nebraska #1 lines (with
C88 resistance derived from Montana No. 5) was a parent of ‘Montcalm’ (Copeland and
Erdmann, 1977). This would explain the presence of the QTL on bean linkage group
810 in ‘Montcalm’, as Montana No. 5 carries this QTL for CBB resistance. ‘Montcalm’
may be useful as a parent for introducing this resistance into other large-seeded market
classes of dry bean. It also may be a useful alternative to cultivars with less CBB

resistance in areas where severe CBB infections have occurred.

125

 

REFERENCES

Copeland, L.O., and M.H. Erdmann. 1977. Montcalm and Mecosta, halo blight tolerant
kidney bean varieties for Michigan. Michigan State University Extension Bulletin
957.

Miklas, P.N., R. Delorrne, V. Stone, C.A. Urrea, J .S. Beaver, and J .R. Steadman. 1998. A
RAPD map of disease resistance traits in common bean. Ann. Rept. Bean Improv.
Coop. 41:93-94.

Miklas, P.N., J .R. Smith, R. Riley, K.F. Grafton, S.P. Singh, G. Jung, and DP. Coyne.
2000. Marker-assisted breeding for pyramided resistance to common bacterial
blight in common bean. Ann. Rept. Bean Improv. Coop. 43:39-40.

Miklas, P.N., D.P. Coyne, K.F. Grafton, N. Mutlu, J. Reiser, D.T. Lindgren, and SP.
Singh. 2003. A major QTL for common bacterial blight resistance derives from
the common bean great northern landrace cultivar Montana No. 5. Euphytica
131:137-146. .1:

 

126

APPENDIX A3

DAYS TO FLOWERING DATA FOR 38 F5 SELECTIONS FROM
SAGIN AW, MI 2002.

Table A3.1 Summary of marker genotype (SU91+ or SU91-) and days to flowering for

38 F5 selections from Saginaw, MI 2002.

 

Genotype’r SU91: Flowering§

02T19111 + 49
803622 + 45
803625 + 43
803628 + 44
803630 + 49
803631 + 49
803632 + 45
803633 + 50
803634 + 49
803637 + 50
803639 + 48
803642 + 45
803643 + 51
N03611 + 48
N03614 + 47
N03617 + 44
803612 - 46
803613 - 46
803615 - 48
803619 - 47
803620 - 44
803621 - 45
803623 — 45
803624 - 48
803626 - 50
803627 - 45
803629 - 44
803635 - 44
803636 - 55
803638 - 49
803640 - 47
803641 - 46

127

Genotypet SU91: Flowering§

803644 - 51
803645 - 46
803646 - 51
803647 - 45
N03616 - 44
N03618 - 45

 

T = “8” indicates black bean genotypes, “N” indicates navy bean genotypes, 02T—91 11 is a black bean
genotype.

1. = “+” denotes presence of the SU91 marker in the genotype.
“-“ denotes absence of the SU91 marker in the genotypes.

§ = days to flowering was recorded as the number of days from planting at which a minimum of 50% of the
plants of the genotype had at least one open flower.

128

APPENDIX A4

DAYS TO FLOWERING DATA AND RE-GREENIN G DATA FOR F3
LINES, EAST LANSING 2003.

Table A4.1 Summary of marker genotype (SU91+, SU91-, BC420+, and BC420-), days

to flowering, and re-greening data for 38 F 5 selections from Saginaw, MI 2002.

 

Genotypet SU91: BC420: Flowering§Re-greening1]

03T-9007 + + 53 +
03T-901 3 + + 50 +
03T-901 9 + - 51 *-
03T-9026 + - 50 +
03T-9029 + + 50 +
DST-9036 + + 50 +
03T-9037 - - 48 -
03T-9046 + + 50 +
03T-9048 + 52 -
03T-9049 + - 56 -
03T-9050 + + 45 +
O3T-9051 + + 47 +
03T-9055 + + 53 +
03T-9056 - - 47 -
03T-9058 + - 50 +
03T-9059 + + 47 +
03T-9060 + + 58 +
03T-9061 + - 49 -
03T-9062 + - 51 -
03T-9067 + - 52 -
03T-9076 + - 48 -
03T~9079 + - 48 -
O3T-9082 + - 51 +
03T-9090 + + 51 -
03T-9092 + - 50 -
03T-9093 + — 51 -
03T-9098 - - 51 -
03T-9099 + - 50 -
03T-9104 + - 49 +
03T-91 1 1 + 51 -
03T-91 12 + + 47 +
03T-91 13 + - 48 *-
03T-91 15 + - 48 -

129

Genotypet SU91: BC420: Flowering§Re-greening1l

03T-91 16 + + 48 +
03T-91 19 + - 48 +
03T-9124 + - 53 -
03T-9126 + - 47 +
03T-9128 + + 46 -
03T-9132 + - 48 +
03T-9133 + - 48 -
03T-9136 + + 47 +
03T-9137 + - 47 +
03T-9139 + - 47 +
03T-9141 + + 46 +
03T-9142 + - 47 +
03T-9146 + - 51 -
03T-9148 + - 46 -
03T-9153 + - 48 +
03T-91 54 + - 47 -
03T-91 56 + - 48 -
03T-9163 + - 51 -
03T-9165 - - 52 -
03T-9166 + - 52 +
03T-9169 + - 48 -
03T-9172 + - 53 -
03T-9178 + - 50 +
03T-9180 + - 48 +
03T-9184 + - 45 -
03T-9185 + - 50 -
03T-9186 + - 46 +
03T-9187 + + 45 +
03T-9188 + + 47 -
03T-9189 + + 47 +
CST-9190 + + 45 +
O3T-9191 + + 46 +
03T-9195 - + 47 +
DST-9199 - + 47 +
03T-9201 + + 46 -
03T-9202 + + 47 -
03T-9204 - + 51 +
03T-9207 + + 47 +
03T-921 7 + - 48 +
03T-9222 + + 48 +
03T-9241 + 45 +
03T-9245 + - 47 +
03T-9247 + - 46 +
DST-9250 + - 47 -
03T-9259 + - 47 +
03T-9263 + - 47 +
03T-9267 + - 46 +
03T-9268 + - 47 +

130

 

Genotypet SU91: BC420: Flowering§Re-greening11

03T-9269 + + 47 +
03T-9270 + + 47 +
03T-9271 + + 47 +
03T-9277 + - 45 +
03T-9281 + - 48 +
CST-9288 + - 43 +
03T-9291 - - 51 -
DST-9299 + - 49 -
03T-9305 + + 48 -
O3T-9306 - - 51 +
03T-9309 + + 50 -
03T-931 9 + - 48 +

 

T = Genotypes are F3 selections from the field study (East Lansing, 2003).
I = “+” denotes presence of the marker (SU91 or BC420) in the genotype.
“-“ denotes absence of the SU91 marker in the genotypes.
§ = days to flowering was recorded as the number of days from planting at which a minimum of 50% of the
plants of the genotype had at least one open flower.
1] = “+” indicates that the genotype exhibited post-maturity re-greening.
“-“ indicates that the genotype did not exhibit post-maturity re-greening.

131

 

l«11111111111111iii:

   

A..-- .—.‘.— —_‘

e’__