mi:

255$.

 

.‘

2.2m; «

sufiéwu “
. it 1

.

h u ‘1 vV
. . .. 4 .
r17". , “3.31 (if):
$3, in. afimsxxnzai Til
ville-v.13!)- ‘5 ,
. , 1illylbvv 3.1...)

.i ..
i. -
a. . .2,

.1-
- l I ||.|.V\ nv‘ ..D|.ya.v|\
II .‘Illrn ~r\ I‘

.3.
.1.
T. . . . 5

{if 10...!
‘1 .. I v .\I.§ .Cvfit:

: .2 a . sq £h$$§~fli $55an

.h‘

"

 

'Ir

we"

j) LIBRARIES
:1 RSITY
' ' MICHIGAN STATE UNIVE
‘ " - ’ 7‘ EAST LANSING. MICK 48824-1048
933, CL» 037/ k

This is to certify that the
dissertation entitled

DISSECTION OF R-GENE MEDIATED ANTHRACNOSE
RESISTANCE IN PHASEOLUS VULGAFIIS

presented by

Veronica Vallejo

has been accepted towards fulfillment
of the requirements for the

 

Ph.D. degree in Plant Breeding and Genetics

. /

Major Professor’s Signature 7
/ 4/ / /5' ., a 5

Date

MSU is an Affin'native Action/Equal Opportunity Institution

 

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

 

JAN 0 4 20m

 

 

 

 

 

 

 

 

 

 

 

 

 

 

# ‘
2/05 c:/CIRC/DateDue.indd—p.15

 

DISSECTION OF R-GENE MEDIATED ANTHRACNOSE RESISTANCE IN
PHASEOLUS VULGARIS

By

Veronica Vallejo ‘

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Plant Breeding and Genetics Program
Department of Crop and Soil Sciences

2005

ABSTRACT

DISSECTION OF R-GENE MEDIATED ANT HRACNOSE RESISTANCE IN

PHASEOLUS VULGARIS

Veronica Vallejo

Anthracnose, caused by the fungal pathogen Colletrotrichum lindemuthianum, is
one of the most economically important diseases of common bean (Phaseolus vulgaris)
worldwide. A comprehensive molecular characterization of genomic and transcriptional
factors which contribute to resistance is necessary to facilitate breeding for anthracnose
resistance. In this study two anthracnose resistance genes from the Andean and Middle
American bean gene pools were characterized using molecular marker and traditional
genetic techniques. In addition, expression analysis was used to identify genes which are
constitutively expressed or induced during the incompatible interaction that may
contribute to the resistance.

An amplified fragment length polymorphism (AFLP) marker (EACTMCCA-108)
was identified, linked at 9.9 cM from the Andean resistance locus, Co-Iz. The AFLP
marker was converted to an STS marker (SEACTMCCA) and mapped to linkage group B1
of the integrated bean linkage map. Additionally, it was determined that the Andean
cultivar Jalo EEP558 carries the Co-I locus using molecular markers, genetic

characterization and inoculation with various races. The C0-7 anthracnose resistance

locus, originally from the Middle American landrace G 2333, was isolated in a breeding
line named MSU7. The resistance spectrum of the C0-7 gene was determined and an
AFLP marker linked to the gene at 4.9 cM was identified.

Several constitutively expressed genes involved in primary metabolism were
identified that have a probable role in the ability of the bean plant to mount C0-42
mediated resistance. Additionally, several genes, differentially regulated between
inoculated and mock-inoculated treated plants in the incompatible interaction, were
identified. Changes in gene expression were detected as early as 1 hour post inoculation.
Among the genes identified was a putative serine/threonine receptor kinase, a putative
lysine decarboxylase-like enzyme and a NIMl-like protein. The products of these genes
have previously been correlated with the defense response in a number of other taxa. I
speculate that the combination of constitutive expression differences in primary
metabolism processes and genes involved in active defense contribute to resistance

mediated by the C042 gene in common bean.

DEDICATION

To my family, who has sacrificed so much to get me to where I am today, and to my
husband, Justin, for his unconditional love and encouragement.

iv

ACKNOWLEDGMENTS

I would like to thank my committee for all of their critical evaluation and support. I
would especially like to acknowledge my advisor/mentor Dr. James Kelly for believing
in me and providing me with this wonderful opportunity. Also, I would like to thank all
of the bean lab people, past and present, for their help, support and companionship, but
especially Pat O’Boyle and Eveline Caixeta for their friendship. I would like to give a
special thanks to Dr. Maeli Melotto for everything that she has done over the years to
promote my professional development. Lastly, I would like to acknowledge the GAANN
(Graduate Assistance in Areas of National Needs) fellowship for financial support and

especially the directors: Rebecca Grumet and Steve Triezenberg.

TABLE OF CONTENTS

LIST OF TABLES ......................................................................... viii
LIST OF FIGURES .......................................................................... ix
GENERAL INTRODUCTION ............................................................. 1
GENERAL OBJECTIVES ........................................................ 20
LITERATURE CITED ............................................................. 26
CHAPTER ONE:

MOLECULAR TAGGING OF THE C 0-1 2 LOCUS AND GENETIC
CHARACTERIZATION OF THE ANTHRACNOSE

RESISTANCE IN ANDEAN BEAN CULTIV AR J ALO EEP558 ................... 33
ABSTRACT ......................................................................... 33
INTRODUCTION .................................................................. 35
MATERIALS AND NIETHODS ................................................. 37
RESULTS AND DISCUSSION ................................................... 41
CONCLUSIONS .................................................................... 45
LITERATURE CITED ............................................................. 56

CHAPTER TWO:

MOLECULAR AND GENETIC CHARACTERIZATION OF THE C 0-7
LOCUS CONDITIONING RESISTANCE TO ANT HRACNOSE IN

COMMON BEAN AND BREEDING LINE MSU7 .................................... 59
ABSTRACT ......................................................................... 59
INTRODUCTION .................................................................. 61
MATERIALS AND NEETHODS .................................................. 63
RESULTS AND DISCUSSION ................................................... 65
CONCLUSIONS ..................................................................... 69
LITERATURE CITED .............................................................. 75

CHAPTER THREE:

ANALYSIS OF CONSTITUTIVELY EXPRESSED AND INDUCED
GENES IN THE RESPONSE OF PHASEOLUS VULGARIS TO
COLLETOTRICH UM LINDEMUTHIANUM ...............................

ABSTRACT ......................................................................... 77

vi

INTRODUCTION .................................................................. 79

MATERIALS AND METHODS .................................................. 81

RESULTS AND DISCUSSION ................................................... 85

CONCLUSIONS .................................................................... 99

LITERATURE CITED ............................................................ 104
APPENDICES:

APPENDIX A1: INITIAL DISSECT ION OF THE
ANTHRACNOSE RESISTANCE IN THE LANDRACE
CULTIVAR G 2338 ............................................................... 110

APPENDIX A2: UNEXPECTED RESISTANCE GENES
UNCOVERED ...................................................................... 1 15

vii

LIST OF TABLES

Table 1.1 Disease reaction of J alo EEP558 and three differential cultivars
to 10 races of C. lindemuthianum ................................................... 48

Table 1.2 RAPD markers located near the mapped location of the Co-I
locus tested on bean cultivars with different anthracnose
resistance genes ........................................................................ 49

Table 1.3 SSR markers evaluated for putative linkage with the Co-I locus
using cultivars Cardinal and Kaboon and R and S bulks ......................... 50

Table 1.4 Molecular markers linked to known anthracnose resistance
genes tested on J alo EEP558 ......................................................... 51

Table 1.5 F2 populations inoculated with C. lindemuthianum ............................... 52

Table 2.1 Primer sequences for the SCAR marker SAB3400 linked to the
C0-5 locus .............................................................................. 71

Table 2.2 F2 Populations inoculated with C. lindemuthianum race 7 ........................ 72

Table 2.3 Disease reaction of MSU7 lines derived from G 2333 and
SEL 111 and SEL 1360 inoculated with various races of
C. lindemuthianum ..................................................................... 73

Table 3.1 Analyses of TDF sequence homology using BLASTX and
BLASTN and their expression pattern from constitutive
experiment ............................................................................. 100

Table 3.2 Analyses of TDF sequence homology using BLASTX and their
expression pattern from cDNA-AFLP analysis of inoculated
nd mock-inoculated plants. TDFs 11.1 and 11.4 were isolated
from the same band. TDFs 18.1 and 18.2 were also isolated from

the same band ......................................................................... 101
Table Al.l RAPD markers linked to anthracnose resistance genes ........................ 113
Table A1.2 SCAR markers linked to the Co-4 locus or specific alleles ................... 113
Table A2.1 F2 populations inoculated with race 7 of C. lindemuthianum ................ 118

viii

LIST OF FIGURES

Figure l The disease cycle of anthracnose in P. vulgaris ...................................

Figure 2 Common bean linkage map. Genomic distribution of genes with
a biochemical function (mostly disease response genes), major

...22

genes coding for phenotypic traits, and QTL in common bean ..................... 23

Figure 3 Pictoral representation of the 5 classes of R-genes. Xa21: Oryza
sativa gene for Xanthomonas oryzae. Cf-9: Lycopersicon

esculentum gene for Cladosporiumfidvum resistance...................................

Figure 1.1 Amplification products of the AFLP marker EACTMCCA-108/ 107
from various bean genotypes using A) (1) Cardinal, (2) Kaboon,
(3) susceptible bulk, (4) resistant bulk, (5-9) individuals of the
resistant bulk, (10-14) individuals of the susceptible bulk,
B) (1) 110 bp fragment of the 10 bp ladder, (2) Kaboon, (3) Cardinal,

(4) J alo EEP558, (5) BAT93, (6) MDRK, (7) Perry Marrow .................

Figure 1.2 Amplification products of the STS marker, SEACTMCCA from
various bean genotypes using (1) MDRK (Co-1), (2) BAT93 (Co-9),
(3) Jalo EEP558 (Co-I), (4) Perry Marrow (Co-13), (5) Kaboon (Co-12),

(6) Cardinal (no known anthracnose resistance genes) .........................

Figure 1.3 The STS marker SEACTMCCA was mapped to bean linkage group

.25

...53

...54

B1 using the BAT93/Jalo EEP558 RIL population ............................... 55

Figure 2.1 Partial gel of amplification products produced by AFLP
primer pair EACCMCGT tested on parents and individuals of
resistant and susceptible bulks. Lanes (1) MSU7-3, (2) MDRK,
(3-8) individuals from resistant bulk, (9) 100 bp ladder, (10-15)
individuals from susceptible bulk. Upper arrow indicates the lower
band as the marker linked to Co-7. Lower arrow indicates the 200 bp

fragment of the 100 bp ladder ........................................................ 74

Figure 3.1 cDNA-AFLP expression profiles across time points 0, 0.5, l, 2,
4, 6, 8, 12, 16 and 20 hpi for inoculated and mock-inoculated
treatments. (A) TDF 11.1 and 11.4 (B) TDF 18.1 and 18.2
(C) TDF 22 and 23, although TDFs 22 and 23 were isolated from
different bands, their expression pattern is identical (D) TDF 31, note

that time point 16h is missing ...................................................... 102

Figure 3.2 Sequence alignment generated by ClustalW. Putative protein

ix

domains detected by the BLASTP program are boxed. A)
Alignment with gi|509l6205 and TDF 18.1 B) Alignment
with gi|49182284 and TDF 18.2 ................................................... 103

GENERAL INTRODUCTION

Anthracnose. Colletotrichum lindemuthianum, the causal fungus of bean
anthracnose, has global distribution affecting not only common bean but also cowpea
(Vigna uniguiculata) and tepary bean (P. acutifolius var. latifolius) among other less
economically important species of Phaseolus and Vigna (Tu, 1982). Anthracnose is the
most economically important disease of common bean worldwide (Melotto et al., 2000
b). Yield losses of up to 95% have been reported (Guzman et al., 1979). The two most
important environmental components which affect infection by C. lindemuthianum and
the expression of disease symptoms are temperature and humidity (Pastor-Corrales and
Tu, 1989). The conidia are produced in acervuli, and are encased in a hydrophilic
mucilage called the spore matrix (Bailey et al., 1983). The optimum temperature range
for conidia development is 18 to 20°C. A relative humidity of at least 92% and moisture
on the foliage for 12 hours post conidia deposition are important for germination (Tu,
1982). The symptoms usually appear on the cotyledonary leaves as small, dark brown to
black lesions. Later, the lesions may appear on veinlets on the abaxial surface of the
leaves, and on the petiole and larger leaf veins. Lesions can also appear on stems and
lead to plant girdling and death. The pathogen often invades the pod, and mycelia and
conidia can infect the cotyledons or seed coat of the developing seeds where it can lie
' dormant. Pod infection appears as flesh to rust-colored lesions that develop into sunken
cankers with a slightly raised black ring border. Severely infected young pods are often
aborted. The infected seed may also contain dark brown to black cankers making them
not suitable for marketing (Pastor-Corrales & Tu, 1989). Sporulation that occurs within

lesions produces secondary inoculum. C. lindemuthianum spores are splash-dispersed

over short distances and wind-dispersed over longer distances (Tu, 1992). Seedlings
grown from infected seed often result in damping off as their hypocotyls tend to break off
from diseased areas (Tu, 1982). In addition to the highly efficient transmission of
anthracnose through seed, another problem which can lead to an outbreak is that the
fungus can survive in seed or infected cr0p residues for many years under low moisture
conditions in the form of conidia or sclerotia. The overall result of disease is yield loss,

marketability of seed and seed quality (Tu, 1982).

Control strategies. Chemical, cultural and genetic control strategies are available
for bean anthracnose with varying effectiveness. Spraying with fungicides at flower
initiation, late flowering and pod fill, however, has had limited success in controlling
anthracnose (Pastor-Corrales and Tu, 1989). In addition, the high cost of repeated
chemical application makes this control strategy impossible for growers in developing
countries, where farming is primarily subsistence and beans are a critical component of
the diet (Broughton et al., 2003). The efficacy of seed treatment is also marginal due to
variable location of the pathogen within the seed. Cultural controls, such as the use of
anthracnose-free seed and crop rotation are highly recommended. Anthracnose-free seed
can be produced in environmental conditions, such as in a hot and semi—arid climate,
which are unfavorable for the development of the pathogen. Two or three year crop
rotation is suggested because of the ability of the pathogen to survive in crop debris
(Lenné, 1992). In developing countries the production and distribution of anthracnose-
free seed is not generally feasible, therefore, the use of resistant cultivars remains the

most practical control strategy. The extensive use of resistant cultivars, anthracnose-free

seed and crop rotation has considerably reduced the importance of anthracnose in North

America, Europe and Australia (Pastor-Corrales and Tu, 1989).

C. lindemuthianum. Colletotrichum lindemuthianum is an ascomycete and is the
causal agent of bean anthracnose. C. lindemuthianum has a perfect and imperfect stage.
Because the telomorph is very rare in nature, and even in culture, the fungus is known by
its anamorph name, Colletotrichum lindemuthianum. The telomorph was first identified
as Glomerella cingulata f.sp. phaseoli but was then classified by Shear as Glomerella
lindemuthianum. Since the imperfect stage is very rare in nature, anthracnose is
primarily caused by the anamorph, therefore, the most common disease cycle is described
in Figure 1. Briefly, conidia, produced in acervuli, that have over-wintered on plant
debris or infected seed, germinate and infect the plant which leads to the development of
symptoms. Secondary inoculum is produced in the infected areas of the plant and is
spread by wind, rain, animals or equipment.

C. lindemuthianum is a hemibiotrophic fungus, that is, it has a biotrophic and a
necrotrophic growth phase. Infection is established through a brief biotrophic phase,
associated with large intracellular primary hyphae. Later, the fungus switches to the
necrotrophic phase which is associated with narrower secondary hyphae which ramify
throughout the host tissue (TeBeest et al., 1997). During the biotrophic phase, the host
plasma membrane invaginates, surrounding the infection vesicle and primary hyphae. An
interfacial matrix is deposited at the host/pathogen interface. Twenty-four hours post
penetration, functional integrity of the host plasma membrane is lost and the host cell

begins to degrade and die. Cell death is confined to infected cells at this point and not

associated with extensive host cell wall dissolution. No macroscopic symptoms are
observed since dead tissue does not become brown. As the primary hyphae colonize new
cells, biotrophy is established in each newly infected cell, thus biotrophy and necrotrophy
may occur simultaneously. The appearance of secondary hyphae announce the start of
the highly destructive necrotrophic phase. The secondary hyphae are narrow and move
both inter- and intracellularly. During this stage, host cells are rapidly killed in advance

of fungal growth. Host cells are extensively degraded and the typical symptoms of

anthracnose become visible (TeBeest et al., 1997).

Pathogen variability. C. lindemuthianum has worldwide distribution, having
been reported on all continents (Pastor-Corrales and Tu, 1989). The classification of C.
lindemuthianum isolates into races was first made by Barrus (1911). Due to the rare
occurrence of sexual recombination and the belief that dispersal over great distances was
limited, initially, pathogenic variability was thought to be low (Beebe and Pastor-
Corrales, 1991). Pathogen variability was difficult to assess because of the lack of a
universal system of race classification. Currently, virulence variability in C.
lindemuthianum is assessed based on the reaction of a given isolate on a standard
differential series of 12 common bean genotypes and a binary system based on the
position of each cultivar within this series (Pastor-Corrales, 1991). The adoptation of this
standard procedure permits the comparison of data from various research groups.
Throughout this dissertation I will sometimes refer to C. lindemuthianum “strains” and
other times “races”. This is an unfortunate consequence of some researchers not adopting

the universal nomenclature used to characterize races.

The combination of virulence and molecular analyses of C. lindemuthianum
populations has expanded the breadth of understanding of variability within and between
fungal populations. Intraspecific diversity has been analyzed using ribosomal DNA
polymorphism (Balardin et al., 1999). To assess intraspecific diversity among 57
isolates, the authors used PCR primers specific to the rDNA region comprising the two
internal transcribed spacers (IT 81 and IT S2) and the 5-8S rRNA gene in C.
lindemuthianum, and digested the resulting amplicons with endonucleases (PCR-RFLP
analysis). The isolates tested divided into two groups based on the PCR-RFIP analysis.
Although Group I consisted mainly of Middle American races (65%), and group II
consisted mainly of Andean races (85%), neighbor-joining and parsimony analyses of the
sequence data did not support an association of any particular ITS genotype with host
gene pool, virulence or geographic origin of races, suggesting that virulence can arise in
different geographic regions at different times, independent of genetic background
(Balardin et al., 1999). Intra-race polymorphisms were observed among isolates of races
7, 17, 31 and 73 collected in various countries using PCR-RFLP of the ITS regions.
These results support a greater level of molecular variability within C. lindemuthianum
than that determined by virulence analysis alone and suggests that specific virulence
patterns evolved independently. More recently, Mahuku and Riascos (2004) assessed
genetic variability of 200 C. lindemuthianum isolates collected from Andean and Middle
American varieties and regions using virulence on the differential series, DNA sequence
of repetitive-elements (Rep-PCR) and random amplified microsatellites (RAMS).
Pathotypic (90 pathotypes identified) and genetic (0.97) diversity among the 200 isolates

revealed a high level of diversity. This diversity, however, does not cluster by bean gene

pools and the authors suggest that the high diversity observed in the Middle American

region indicates that C. lindemuthianum originated from this region.

Host resistance. There are three general categories of the host response to
infection with C. lindemuthianum: extreme resistance, extreme susceptibility and
intermediate. Extreme susceptibility is characterized by the development of fungal
infection vesicles within living epidermal cells, followed by further colonization of host
cells by intracellular primary hyphae and a transient biotrophic relationship is re-
established in each newly colonized cell. Six to seven days post inoculation, brown,
water-soaked lesions appear and the fungus switches to necrotrophic growth, that results
in the breakdown of plant cell walls and the death of plant cells in advance of fungal
growth (O’Connell and Bailey, 1988). Extreme resistance is characterized by the rapid
and localized death of epidermal cells soon after penetration. Infection vesicles are not
formed and the growth of the fungus is restricted to the penetrated cell (O’Connell et al.,
1985). The gene-for-gene (GFG) relationship between C. lindemuthianum and P.
vulgaris was proposed as expressed as either the survival of the initially infected
epidermal cell or its rapid death (Bailey, 1983), although, most studies involving the
inheritance of anthracnose resistance evaluate host response at the plant level rather than
at the single cell level. The intermediate category of host response to C. lindemuthianum
was created to accommodate other less extreme host responses. This category ranges
from small groups of dead cells, which appear as flecks, to large areas of necrosis,

classified as limited lesions. Plants which are placed in the intermediate response

category, often survive and yield well, and consequently are considered to be resistant by
some research groups and partially resistant by others.

Resistance to bean anthracnose, primarily conditioned by major genes (Co-I
through Co-IO), is believed to function in a GFG manner with pathogenicity in C.
lindemuthianum, although some quantitative trait loci (QTL) for partial resistance have
also been reported (Geffroy et al., 2000). The GFG theory, proposed by Flor (1947),
describes when one gene in the plant corresponds with one gene in the pathogen to
mediate the compatible/incompatible reaction. The gene in the plant which mediates this
response is referred to as a resistance gene (R-gene). R-genes are inherited, most often,
in a single dominant fashion, although, examples of recessively inherited resistance 3 not
uncommon (Deslandes et al., 2002; Huang et al., 1997 ; Johansen et al., 2001). This GFG
relationship between P. vulgaris and C. lindemuthianum is inferred since there have not
been studies conducted on the inheritance of pathogenicity in the fungus due to problems
with ascospore viability (Bryson et a1, 1992). Each anthracnose resistance gene
conditions resistance to numerous races of C. lindemuthianum and several genes (Co-1,
Co-3 and C0-4) are reported as complex loci with multiple alleles. This is in accordance
with genetic studies that revealed that genes efficient against different strains of a specific
pathogen were often located at complex loci displaying a multiallelic structure and/or a
cluster of linked genes responsible for different specificities (Crute and Pink, 1996; Pryor
and Ellis, 1993).

As an alternative to major genes which condition resistance to anthracnose,
Geffroy et al. (2000) focused on the identification of QTL that contribute to anthracnose

resistance. Geffroy et al. (2000) identified 10 putative QTL scattered across the genome

(Figure 2) contributing from 11-76% of the phenotypic variation. Some of the QTL are
strain-specific and co-localize with other anthracnose resistance genes or defense
response genes. In this study, the selection of C. lindemuthianum strains and the
resistance source were critical to detecting QTL. This analysis was conducted using two
strains of C. lindemuthianum that produce differential symptoms on the parental material
where only partial resistance was observed. Tissue- and strain-specific QTL were
detected. A critical assessment of this study is made difficult because the authors did not

characterize their fungal strains by the universal anthracnose differential series.

Resistance genes. Gene—for-gene disease resistance in plants involves two basic
processes: the perception of the pathogen and the response by the plant to limit disease.
Based on the combination of structural motifs, most cloned R-genes fall into five classes
(Jones, 2001) (Figure 3). Class 1 contains just one member, Pto (tomato), which contains
a serine-threonine kinase (STK) catalytic domain. Additionaly, Pto has a N-terminal
myristilation motif. The second class is made up of proteins which have leucine-rich
repeats (LRRs), a putative nucleotide binding sit (NBS), and an N-terminal putative
leucine-zipper (LZ) or other coiled-coil (CC) sequence. Class 3 R-gene proteins are the
same as class 2 but instead of the CC sequence, these proteins have a TIR domain which
is homologous to the N terminus of the 1011 and Intertleukin 1 receptor (IL-1r) proteins.
The notable lack of a transmembrane (TM) domain in proteins belonging to classes 1
through 3, suggests that these are be intracellular proteins (Figure 3). Class 4 R-genes

lack an NBS and have a TM and an extracellular LRR. The fifth class of R-gene proteins

has a single member, Xa21 protein from rice. This protein has an extracellular LRR, a
TM and a cytoplasmic STK domain (Jones, 2001).

No R-genes have been cloned from P. vulgaris, however, several R-gene-like
sequences have been discovered. Melotto and Kelly (2001) identified a 1,110 bp open
reading frame (ORF), closely linked to the C0-42 anthracnose resistance gene. This ORF,
named COX-4, encodes a STK domain, highly similar to the Pto gene in tomato, but with
a highly hydrophobic membrane-spanning region. Melotto et a1. (2004) cloned several
additional copies of the COX-4 gene from the same region, identifying a cluster of STK
sequences. A family of related LRR sequences has been identified in the region of the
Co-2 anthracnose resistance gene (Geffroy et al., 1998). Three of these LRRs were
further characterized and found to be members of the NBS-LRR class of R-gene proteins
(Creusot et al., 1999). Ferrier-Cana et al. (2003) identified four NBS-LRR sequences
which map to bean linkage group B4, co-localizing with previously identified QTL and
major genes (Co-y, Co-z, and Co-9) for anthracnose resistance. Additionally, Rivkin et
al. (1999) identified several NBS containing genes from P. vulgaris with conserved
kinase domains. Some of these NBS sequences mapped near the Ur-6 gene which
confers resistance to Uromyces appendiculatus.

Most cloned R-gene proteins are grouped into either class 2 (NBS-LRR with N-
terminal CC domain) or class 3 (NBS-LRR with N-terrninal TIR domain) type proteins
(Ellis et al., 2000). Class 1 is the only class of R-genes lacking an LRR, therefore, R-
gene proteins appear to rely on a limited number of structural and functional domains, of
which the LRR must be important. The LRR domain contains leucine or other

hydrophobic residues at regular intervals (~24 amino acids in length) (Bent, 1996).

Leucine-rich repeat domains of proteins from yeast, Drosophila and humans are known to
mediate protein-protein interactions (Bent, 1996). Because LRR were known to play a
role in protein-protein interactions in eukaryotes, researchers speculated that the LRR
may function in recognition of the pathogen via binding of an elicitor molecule in plants.
Support for this functional role comes from the identification of mutant alleles of RPSZ
(Bent et al., 1994; Mindrinos et al., 1994) and RPMl (Grant et al., 1995), which are
nonfunctional due to single amino acid changes within the encoded LRR region. In
addition to recognition, some studies suggest roles for the LRR in signaling. For
instance, point mutations in the LRR domain of the Arabidopsis RPSS resistance protein
partially suppresses multiple bacterial and downy mildew resistance genes, suggesting a
dominant negative interaction with a shared signaling component (Warren et al., 1998).
The NBS domain of R-genes, sometimes referred to as a P-loop, is an arrrino acid
motif which is present in a variety of proteins having ATP or GTP binding activity
(Martin et al., 2003). The NBS is part of a larger domain that includes homology
between R-gene proteins and eukaryotic cell death effectors, Apaf-l and Ced4. Together,
this larger domain is referred to as the NB-ARC or Ap-ATPase domain. Because the
NBS domain is highly conserved in some R-genes, the functionality of these proteins is
likely to be dependent on binding nucleotide triphosphate (Martin et al., 2003). The CC
motif of R-gene proteins is a repeated heptad sequence which has hydrophobic amino
acid residues. The LZ is an example of a CC structure (Martin et al., 2003). The LZ,
found in many different proteins, is believed to promote the formation of CC structures
facilitating protein-protein interactions and is better known for its role in the dimerization

of eukaryotic transcription factors (Bent, 1996). In Arabidopsis, the R-gene proteins

10

which contain CC domains, depend on downstream signaling components that are
different from those required by TIR-NBS-LRR proteins. This suggests that the CC
domain may function in signaling rather than in recognition (Martin et al., 2003). Serine-
threonine kinases are proteins involved in phosphorylation and thus are believed to be
involved in signal transduction of the defense response (Hardie, 1999). Both Pto and the
kinase domain of Xa21 are functional STKs (Martin et al., 2003).

Hypersensitive response. Stakman (1915) is generally credited as the first to use
the term ‘hypersensitive’ to describe the rapid and localized plant cell death induced by
rust fungi in rust-resistant cereals. The term ‘hypersensitive response’ was adopted when
it was clear that such cell death was a common expression of disease resistance in plants,
regardless of pathogen (Heath, 2000). This response is defined as a rapid, localized cell
death at the site of pathogen infection (Kombrink and Schmelzer, 2001). Characteristic
of the hypersensitive response (HR) is the presence of brown, dead cells at the site of
infection, although, HR is not restricted to those cells which have been infiltrated by the
pathogen (Heath, 2000).

Most R-genes trigger the HR, although, mutational studies indicate that the HR
also depends on genes which are present in both resistant and susceptible genotypes,
termed ‘required for disease resistance genes’ or RDR genes, even in non-GFG
interactions. For instance, mutations in the NDRI (a putative membrane associated
protein) gene of Arabidopsis, suppresses resistance which is mediated by class 2 but not
class 3 R-genes, whereas, the opposite is the case with mutations in EDS] (a putative
lipase). This suggests that there must be more than one pathway that leads to HR cell

death. The activated pathway may be more dependent on the class of R gene rather than

11

the pathogen (Heath, 2000). In A. thaliana, mutations at the DNDI locus cause plants to
be defective in HR cell death, but retain characteristic responses to avirulent
Pseudomonas syringae such as the production of pathogenesis-related proteins and strong
restriction of pathogen development (Yu et al., 1998). In P. vulgaris, phenylalanine
ammonia-lyase, chalcone synthase, chalcone isomerase and chitinase transcripts
accumulated after infiltration with a Hrp-mutant strain of Pseudomonas syringe pv tabaci
in the absence of a HR. Transcript accumulation occurred in the same temporal pattern
as seen with infiltration with a wild-type strain which elicited the HR (Jakobek and
Lindgren, 1993). These and other studies in which HR cell death was uncoupled from
the induction of defense genes and disease resistance (del P020 and Lam, 1998; Heath,
1998), also suggest that there are unique biochemical events associated with the HR,
distinct from other plant defenses.

Additional characteristic features of R-gene mediated resistance response are the
induction of phytoalexin production, the synthesis of chitinases and glucanases and the
reinforcement of the cell wall. Early events in the signal transduction pathway of the
defense response include calcium ion flux, specific changes in protein phosphorylation,
the generation of activated oxygen species and the production of salicylic acid (Bent,
1996).

Co-evolution of P. vulgaris and C. lindemuthianum. Wild beans are found from
northern Mexico to the north-west region of Argentina (Gepts and Debouck, 1991). P.
vulgaris was domesticated in two distinct regions, Middle America, primarily Mexico,
and along the eastern slope of the Andes in South America, specifically southern Peru,

Bolivia and northwestern Argentina (Gepts et al., 1986). A minor domestication center

12

has been suggested in Colombia, however, it is unclear whether this area represents a
domestication center or a region of gene flow between wild and domesticated bean gene
pools (Gepts and Bliss, 1986; Beebe et al., 1997). The divergence of the two major gene
pools during domestication has been supported by many morphological and molecular
studies: variability in seed size (Evans, 1973), variability in the major seed storage
protein, phaseolin, (Gepts et al., 1986), isozyme (Koenig and Gepts, 1989; Singh et al.,
1991) and DNA marker variability (Becerra-Velasquez and Gepts, 1994; Haley et al.,
1994). In addition to the above mentioned gene pools, recently discovered wild
populations, considered to be ancestral, constitute a third gene pool in the region of
Ecuador and northern Peru (Debouck et al., 1993; Kami et al., 1995; Tohme et al., 1996).
Based on their unique phaseolin type, which is absent from the domesticated gene pools,
these ancestral populations are not believed to have been involved in the domestication of
common bean (Debouck et al., 1993).

The two major gene pools belong to the same species, despite a partial
reproductive isolation caused by complementary lethal genes (Koinange and Gepts,
1992) and thus provide unique opportunities for breeders. A notable example is that of
Beaver and Kelly (1994) who employed recurrent selection strategies to improve the
yield potential of large-seeded red beans. Another good example of utilizing the genetic
differences between gene pools is in breeding for anthracnose resistance.

Anthracnose resistance genes are categorized according to the gene pool origin of
the host cultivar. The races of C. lindemuthianum are also similarly classified depending
on the gene pool of the host cultivar for which each race was isolated (Balardin & Kelly,

1998). Isolates from the Andean region have a narrow virulence range and are more

13

virulent on cultivars of Andean origin, whereas Middle American isolates have a broader
virulence range and predominantly attack Middle American cultivars (Beebe & Pastor-
Corrales, 1991; Pastor-Corrales et al., 1995; Cattan-Toupance et al., 1998; Geffroy et al.,
1999). Several studies on various host/pathogen interaction have shown that plants are
more resistant to allopatric pathogens than to sympatric ones (Parker, 1985; Nevo, 1986;
Lawrence and Burdon, 1989). This is consistent with theoretical predictions for
conditions when either migration rates are low for both host and pathogen or migration
rate of the host is lower than that of the pathogen (Gandon et al., 1996).

If resistance to anthracnose is controlled via a GFG relationship between host and
pathogen, this would imply that bean plants carry R-genes corresponding to Avr genes
from C. lindemuthianum races which may have originated in the opposite gene pool. The
inference is that bean plants must carry many R-genes which confer resistance to
allopatric races. This raises the question: why would a plant maintain R-genes for races
of a pathogen for which it is not exposed (also termed unnecessary R-genes)? Geffroy et
al. (1999) presented several possible explanations: (l) the unneeded specificities might
be maintained because of their linkage to other R-genes, (2) these apparently unneeded
genes might provide a renewable resource for the generation of new R-genes by genetic
re-assortment events, (3) these R-genes may have other roles in the biology of the plant
and might thus be involved in its fitness, (4) these genes may function locally against
other pathogens or races delaying epiderrrics, (5) these genes may be maintained because
they force the pathogen to limit the number of virulence factors it can use. Although, this
may suggest a co-evolution of host and pathogen leading to parallel gene pools, when one

considers pathogen studies, the literature on the pathogenic variation of C.

14

lindemuthianum is conflicting. Several research groups have found evidence to support
the co—evolution of host and pathogen (Pastor-Corrales, 1996; Sicard et al., 1997;
Balardin and Kelly, 1998, Ansari et al., 2003) yet others have determined that there is no
association between virulence phenotype and geographic region or host gene pool (Fabre
et al., 1995; Balardin et al., 1997; Balardin et al., 1999; Mahuku and Riascos, 2004). The
conflict in the literature may be a product of the use of bred genotypes by some
researchers. Using genotypes which result from breeding programs, the identity and
origin of the resistance genes are not known and might confound the classification of
isolates into distinct groups congruent with the diversity in P. vulgaris. Another possible
reason for the conflict is that the separation and classification of pathotypes is dependent
on the differential series. The present differential series contains only four Andean
genotypes, all of which carry the Co-I gene, and eight Middle American genotypes, some
of which have multiple genes which condition resistance to anthracnose. To clarify the
question of co-evolution of C. lindemuthianum and P. vulgaris, a thorough analysis based
on a large set of Andean and Middle American wild genotypes and pathotypes is required
before strong conclusions about co-evolution can be made. Geffroy et al. (1999) carried
out a similar study to analyze the molecular evolution of R-genes and the host-pathogen
co-evolution process at the population level. The authors did not, however, assess
molecular diversity of the C. lindemuthianum strains used. They collected 48 wild P.
vulgaris plants from the three centers of diversity of the host species (Middle American,
South Andean and North Andean) and cross inoculated them with 26 strains of C.
lindemuthianum. They found that most of the resistance specificities (R-genes) were

overcome in sympatric situations, indicating an adaptation of the pathogen to the local

15

host. In contrast, plants were generally resistant to allopatric strains which suggests that
R-genes that were efficient against exotic (allopatric) strains but had been overcome
locally were maintained in the plant genome. The authors concluded that co-evolution
between P. vulgaris and C. lindemuthianum has led to a differentiation for resistance in
the three centers of diversity of the host. Additionally, they mapped phenotypic
resistance and several resistance gene analogs (RGAs) and found a cluster comprising
both Andean and Middle American resistance specificities, suggesting that this locus
existed prior to the separation of the two major gene pools of P. vulgaris.

Comparative studies of wild and cultivated populations of P. vulgaris have shown
greater genetic diversity in wild populations, suggesting a founder effect of domestication
(Gepts, 1990; Sonnante et al., 1994). The domesticated Andean gene pool is diverse in
plant and seed morphology as well as agroecological adaptation, however, based on
molecular analyses has a narrow genetic base as compared to Middle American
gerrnplasm (Beebe et al., 2001). AFLP analysis revealed that wild Andean bean
populations have undergone more geographic isolation from each other as compared to
most Middle American wild beans. Therefore, less genetic “rrrixing” has resulted in
discrete populations in southern Peru, Bolivia and northern Argentina (Tohme et al.,
1996). Interestingly, only one Andean anthracnose resistance gene has been
characterized (Co-1) and four putative Andean resistance genes have been identified (C0-
w, Co-x, Co-y and Co-z) (Kelly and Vallejo, 2004). The low number of Andean
anthracnose resistance genes identified, relative to the number of Middle American (Co-2
through Co-IO) genes, is not surprising considering that the Andean gene pool is lower in

genetic diversity than the Middle American.

16

Breeding for anthracnose resistance. Despite the extensive literature on
resistance to anthracnose, bean breeders still struggle with the decision as to which
gene(s) to deploy in resistance breeding programs. Independence of the nine anthracnose
resistance genes (Co-3 is believed to be allelic to Co-9) distributed across the genome
(Kelly and Vallejo, 2004) offers bean breeders the unique opportunity to pyramid
complementary resistance genes (Duvick, 1996) and, in certain cases, based on it’s
resistance spectrum, to choose the most effective allele at different loci, as the most
prudent strategy in breeding for durable resistance to anthracnose (Kelly and Miklas,
1998). When breeding for anthracnose resistance in a particular geographic area, careful
consideration should be taken in choosing which genes to combine. Ideally, each gene of
a pyramid should, if deployed singly, condition resistance to all known races in that
specific geographic region (Kelly and Miklas, 1998). For example, the combination of
the Co-5 and Co-6 anthracnose resistance genes would meet this criterion in North
America (Young and Kelly, 1996). To expedite the process of gene pyramiding, a
combination of marker-assisted selection, and confirmation of successful gene
introgression by inoculation with select races, is the best strategy for anthracnose
resistance breeding (Kelly et al., 2003). Given the recent activity in mapping and gene
tagging, new information on the location of most major genes controlling resistance to
anthracnose is now available (Kelly et al., 2003). The availability of markers linked to
specific anthracnose resistance genes, and alleles, in the case of multi-allelic loci such as
Co-I, Co-3, C04 and Co-9 provide essential tools for the construction of pyramids with

genes of complementary resistance spectra. The identification of markers linked to

17

anthracnose resistance genes is particularly important when it is not possible to confirm
introgression by inoculation with discriminating races. This is the case when one gene is
hypostatic to the other and a discriminating race cannot be identified.

Although most anthracnose resistance genes are inherited independently of each
other, they are also in some cases clustered with other resistance genes in the bean
genome (Kelly et al., 2003). Gene clusters of anthracnose and Ur-rust resistance genes
are located on linkage groups B1, B4 and B11 (Figure 2). The Andean resistance genes
for anthracnose, Co-I, and rust, Ur-9, cluster on B1, whereas Middle American genes,
Co-3/Co-9, Ur-5, and Andean genes Co-y, Co-z cluster on B4. Other Middle American
genes, Co-2 and Ur-3/Ur-11, cluster on B11 (Kelly et al., 2003).

In the case of anthracnose resistance genes, clusters have been reported at the Co-
2 and Co-9 loci (Creusot et al., 1999; Ferrier-Cana et al., 2003; Geffroy et al., 1999;
Mendez-Vigo et al., 2005). The breakdown of resistance conferred by Co-2, a gene that
confers resistance to a wide array of races, may have been aggravated by the disruption of
the resistance gene cluster. Similarly, resistance conferred by Co-2 could have broken
down due to recombination events within the locus thus disrupting the functionality of
closely linked resistance genes (Kelly and Vallejo, 2004). Although multiple alleles at
other anthracnose resistance loci have been reported, the possibility that there are
multiple closely linked genes in a cluster is not excluded by the allelism test. Therefore, it
is very likely that gene clusters do exist at loci where multiple alleles have been
previously identified. Allelism studies on populations of limited size (n=100), detect only
linkage between two loci being tested, therefore one can only conclude whether two loci

are less than or greater than 50 cM apart. This test for independence does not detect

18

 

recombination events that occur within the putative cluster, because only one anthracnose
race is typically used in the analysis. Despite the evidence of resistance gene clusters,
practical breeding for resistance may not be greatly facilitated by this knowledge. The
value of mapping for breeders is not the associated traits that might be carried along in
crossing, but the knowledge of independence of resistance genes should facilitate the
pyrarrriding of different genes to enhance resistance. Although prudent pyrarrriding of
resistance genes and deployment of appropriate resistance gene combinations has proven
to be an adequate method of anthracnose control in many parts of the world, the addition
of QTL that confer partial resistance offers breeders additional tools in selecting for

durable resistance particularly in areas where pathogen diversity is extremely high.

19

GENERAL OBJECTIVES

Colletotrichum lindemuthianum is a highly variable and widespread pathogen, which
makes breeding for durable anthracnose resistance a serious challenge. Although many
major anthracnose resistance genes have been characterized and quantitative resistance is
being investigated, Andean sources of resistance remain under characterized and utilized.
New sources of resistance, both Andean and Middle American must be characterized to
provide breeders various options to exploit the best combination of genes for the release
of locally resistant bean cultivars. Tools which facilitate the introgression of these
resistance traits are required to expedite the selection process. In this dissertation, the
objectives were to develop molecular markers linked to the Andean Co-Iz gene and
characterize a new Andean source of resistance, to characterize the Middle American
anthracnose resistance gene Co-7 and lastly, to investigate the expression of genes
involved in the early resistance response to C. lindemuthianum. The research project is
divided into three major sections, each of which is represented by a chapter in this
dissertation. The objectives in the first chapter were to develop a molecular marker
linked to the Co-IZ gene from the highly resistant Andean cultivar Kaboon, and to
characterize the anthracnose resistance present in the Andean cultivar J alo EEP558, used
as a parent in the bean integrated linkage map. The second chapter objectives were to
develop bean lines which are homozygous for the Co-7 gene, characterize the spectrum
of resistance conditioned by the Co-7 gene and find an amplified fragment length
polymorphism (AFLP) marker linked to this locus. The third chapter objectives were to

identify genes with a putative role in the defense response by investigating constitutive

20

differences between plants which carry the anthracnose resistance gene, Co-42, and plants
which do not, and by investigating changes in gene expression during the incompatible

interaction between C. lindemuthianum and P. vulgaris by the cDNA-AFLP technique.

21

Conidia (n) spread 10 More acervuli that
other part Of plant or contain conidia

and rain storms, or by

people moving through
wet fields

 

    
 

Secondary infection

 

J Conidia germinate and fungus infects Development of symptoms
plant tissue Eg. Leaf, stem, petiole etc

Primary infection

  

Overwinter as mycelium or
conidia on debris or in seed

 

Conidia (n)
In acervulus in host debris or
seed

 

Figure 1. The disease cycle of anthracnose in P. vulgaris.

22

069.095
0...: CL:
Duet

 

IIIIKK
!
i

9....-

Ill
13!
a~-

E

3.13 as:

002‘:

 

I KKK

‘0
Q

:5
.- 5...:
.22....
ant—d);
81:3
83!.
.3...
.1...-
... ...u
I. ...a
3.2...
sifuv.
83.3
In:
3.3.3.
8....
.5:
3m
=>
3...:
8......3
5...
l. .3.
.85 7&3
5:.
.34..

75....
5.: 7...:
I. 8......
82...... h
3.3...
3...:
3......
3......
3.2...
‘.:U
.2...
.2... .k
a. «as
8133.
8.2.3
3:...
8....
8......
‘1:—
is
03.0
. 95.: 2:9...
.2. E .3393
2.... 9.:
ii... n60 >60
2.00
.5
..
3......
8:...e
2......
.1: U
a...
3....
3....
3......
I25... :6 m
i

 

:=§
'3.=§<

~78

8...:
8...... Ni. uttu

lllK

 

I III

E... z o. .6?
.I. «R:
In. :61

”1:21

RI:

.2 38:...
...E
5.93
To».
Shun”... Nuooi—
..n..k Nut:
.0...
to...
.33..
5.: «1:.
.3! .w
3...:
833 0.60
83...: l
I...
and...
.3... 0.00
a 2.. ..8
2.1... 8...:
in
8...)... t.
.34... EU 3
83.3.
is... A E 5...: 0019:
8:...- N23
5%
.5:
.82..
8......
“to
.82... ~—
83......
8:3 «.01..
82...: E 8......
R.u.¢.~§..§..ltu
.'
§~N=~ ~“0 £U~Z<
.25...
8...... 39m ...53-
...33 .
ES «20
8...; «it:
3.... 3....
8...... ”an“: 5‘
5.... a and;
8%... 3.2.3
85.3.. E .23 T...“
.3...
8...... a...&
3...... I}...
~Q= .700 8.43 ”M
.243. a...- :0
.0. x60
3......
.558: 5.... TOO
~fl

23

.958 092:: 85 .o 8288 8088 on. .6 8:82 on. E Em... 2t 8 850% 2 A20 5880M 8 83898 88w 8.88:
88 .8 8:828 .82 2:. .8382. 2 88 25 823 808 88.88 wmmmmm ofifimah<m 05 8 8:88 3888 8: 083 888 8
6888088 2 88m .88 We :2883 2:88:88 888;. 8 8:88 @733 088858 .8 480 0.8 A288». 8.88 Em: 08
on. @8888 .888... .8 08w x 2 m3. .888 :88: .28 88 0:: N 88 .N d 8% 8:... .8... 2:. 88m .28 88 8: .2 US. .5. .D
.Eui .m 288.8 88... 8m .8: .8888an 2 33289.. $288.22. .8 88w 0.8 ..w «in. .8... .388 088888 887888
..2588 82x8- U8 .8 8.68 802 888 on. 2 H 8.2:. A3 :85 2:82 8800 88m 2 88328. .8 88m 3688.
8.. men 28 N78 2-8 :8. 8:828. 8.882. a 2 0:8 ~ BC. .88888 88— 353: .8 3-3% 2: 88.88 82 8:828.
“we. 8: 82 .b och 82 8:828. 08888:: 8.... 82 eb 05. A882 8283 88m 88 29:08:: 8.88 8:: 3:8 Swan:
88m 3880805 05 n8. 9:8 .2883 8888 828208 2.53088. 05 8e deem 888: 88 mo 82 08 oh .803 :8 8 cm;
-8836 em :0 8382 08mm 08 8 8:88 :89 08: $69 88 we... 6:: $88 .8 8 88:0 .3 hm 2 8:88 83 new ”88:8
8232—8 2.: 5.3 860$ 2.2:; .8: zzovu 808 2 88 8.5. .88 8888 8 APO 8a .88. 299828 .8 w:€8 88m 888

.388 8:88. 0888 >388V 88:8 88.8829 a £3, 88w .0 8888.6 28080 .88 888: 88 88800 .N 0.5!.—

24

Cf-9 Xa21
Class 4 Class 5

 

Plasma
membrane
—

cytoplasm

Pto
Class 1

 

RPSZ N, L6,
RPMl RPPS
Class 2 Class

Figure 3. Pictoral representation of the 5 classes of R-genes. Xa21: Oryza sativa gene
for Xanthomonas oryzae. Cf-9: Lycopersicon esculentum gene for Cladosporium fulvum
resistance. Pto: Lycopersicon pimpinellifolium gene for Pseudomonas syringae
resistance. N: Nicotiana tabacum gene for tobacco mosaic virus. RPSZ, RPM]:
Arabidopsis thaliana resistance to Pseudomonas syringae. L6: Zea mays gene for
Melampsora lini resistance. RPPS: Arabidopsis thaliana gene for Peronospora

parasitica.

25

 

LITERATURE CITED

Ansaii, K.I., Palacios, N., Araya, C., Langin, T., Egan, D., and Doohan, RM. 2003.
Pathogenic and genetic variability among Colletotn'chum lindemuthianum isolates of
different geographic origins. Plant Path. 53:635-642.

Bailey, J .A. 1983. Biological perspectives of host-pathogen interactions. Pages 1-32. In:
The Dynamics of Host Defense. J .A. Bailey and BJ. Deverall (eds.). Academic Press,
Sydney, Australia.

Bailey, J.A., O’Connell, R.J., Pring, R.J., and Nash, C. 1992. Infection strategies of
Colletom'chum species. Pages 88-120. In: Colletotrichum Biology, Pathology and
Control. J .A. Bailey and M.J. Jeger (eds.). CAB. International, Wallingford, UK.

Balardin, R.S., Jarosz, A., and Kelly, JD. 1997. Virulence and molecular diversity in
Colletotrichum lindemuthianum from South, Central and North America.
Phytopathology 87: 1 184-1 191.

Balardin, RS. and Kelly, J .D. 1998. Interaction between Colletotrichum lindemuthianum
races and gene pool diversity in Phaseolus vulgaris. HortScience 123:1038-1047.

Balardin, R.S., Smith, J .J ., and Kelly, J .D. 1999. Ribosomal DNA polymorphism in
Colletotrichum lindemuthianum. Mycological Research 103:841-848.

Barrus, M.F. 1911. Variation of cultivars of beans in their susceptibility to anthracnose.
Phytopathology 1:190-195.

Beaver, J.S., and JD. Kelly. 1994. Comparison of two selection methods for the
improvement of dry bean populations derived from crosses between gene pools. Crop
Sci. 34:34—37.

Becerra—Velasquez, V.L. and P. Gepts. 1994. RFLP diversity of common bean
(Phaseolus vulgaris) in its centers of origin. Genome 37:256—263.

Beebe, SE. and Pastor-Corrales, M.A. 1991. Breeding for disease resistance. Pages 561-
617. In: Common Beans: Research for Crop Improvement. A. Van Schoonhoven and
O. Voyest (eds.). CAB. International, Wallingford, UK, and CIAT, Cali, Colombia.

Beebe, S., Toro, 0., Gonzalez, A.V., Chacon, M.I., and Debouck, D. 1997. Wild-weed-
crop complexes of common bean (Phaseolus vulgaris L. Fabaceae) in the Andes of

Peru and Colombia, and their implications for conservation and breeding. Gen. Res.
Crop Evol. 44:73-91.

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

26

 

Bent, A.F., Kunkel, B.N., Dahlbeck, D., Brown, K.L., Schmidt, R., et al. 1994. RPS2 of
Arabidopsis thaliana: a leucine-rich repeat class of plant disease resistance genes.
Science 265: 1856—1860.

Bent, A.F. 1996. Plant disease resistance genes: Function meets structure. Plant Cell 8:
1757—1771.

Broughton, W.J., Hernandez, G., Blair, M., Beebe, S., Gepts, P., and Vanderleyden, J.
2003. Beans (Phaseolus spp.) — model food legumes. Plant and Soil 252:255.

Bryson, R.J., Caten, C.E., Hollomon, D.W., and Bailey, J.A. 1992. Sexuality and genetics
of Colletotrichum. Pages 27-46. In: Colletotrichum: Biology, Pathology and Control.
J .A. Bailey and M.J. Jeger (eds.). C.A.B. Intemational,Wallingford, UK.

Campa., A.C., Rodriguez-Suarez, A., Paneda, R., Giraldez, J ., and Ferreira, J. 2005. The
bean anthracnose resistance gene C0-5, is located in linkage group B7. Annu Rept.
Bean Improv. Coop. 48:68-69 (in press).

Cattan-Toupance, I., Michalakis, Y., and Neema, C. 1998. Genetic structure of wild bean
populations in their South-Andean center of origin. Theor. Appl. Genet., 96:844-851.

Creusot, F., Macadre, C., Ferrier—Cana, E., Riou, C., Geffroy, V., Sevignac, M., Dron,
M., and Langin, T. 1999. Cloning and molecular characterization of three members of
the NBS-LRR subfamily located in the vicinity of the Co-2 locus for anthracnose
resistance in Phaseolus vulgaris. Genome 42:254-264.

Crute, LR, and Pink, D.A.C. 1996. Genetics and utilization of pathogen resistance in
plants. Plant Cell 8: 1747-1755.

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

Deslandes, L., Olivier, J., Theulieres, F., Hirsch, J., Feng, D.X., Bittner-Eddy, P.,
Beynon, J ., and Marco, Y. 2002. Resistance to Ralstonia solanacearum in
Arabidopsis thaliana is conferred by the recessive RPSI-R gene, a member of a novel
family of resistance genes. Proc. Natl. Acad. Sci. USA 99:2404-2409.

Duvick, D.N. 1996. Plant breeding, an evolutionary concept. Crop Sci. 36:539-548.

Ellis, J ., Dodds, P., and Pryor, T. 2000. Structure, function and evolution of plant disease
resistance genes. Curr. Opin. Plant Biol. 3:278—84.

Evans, A.M. 1973. Plant architecture and physiological efficiency in the field bean. Pages

279—284. In: Potentials of Field Bean and Other Food Legumes in Latin America. D.
Wall (ed.). CIAT, Cali, Colombia.

27

 

Fabre, J .V., Julien, J ., Parisot, D., and Dron, M. 1995. RAPD, rDNA and pathogenicity
analyses describe two different groups of strains in Colletotrichum lindemuthianum,
the causal agent of anthracnose of common bean. Mycological Research 99:429—435.

Ferrier—Cana, E., Geffroy, V., Macadre, C., Creusot, F., Irnbert-Bollore, P., Sevignac, M.,
and Langin, T. 2003. Characterization of expressed NBS-LRR resistance gene
candidates from common bean. Theor. Appl. Genet. 106:251-261.

Flor, H.H. 1947. Host parasite interaction in flax rust — its genetics and other
implications. Phytopathology 45:680-685.

Gandon, S., Capowiez, Y., Dubois, Y., Michalakis, Y., and Olivieri, I. 1996. Local
adaptation and gene for gene co-evolution in a metapopulation model. Proc. Roy.
Soc. Lond. B 263:1003-1009.

Geffroy, V., Creusot, F.J., Falquet, J ., Sevignac, M., Adam-Blondon, A.F., Bannerot, H.,
Gepts, P. and Dron, M. 1998. A family of LRR sequences in the vicinity of the Co-2
locus for anthracnose resistance in Phaseolus vulgaris and its potential use in marker-
assisted selection. Theor. Appl. Genet. 96:494-502.

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

Geffroy, V., Sevignac, M., de Oliveira, J .C.F, Fouilloux, G., Skroch, P., Thoquet, P.,
Gepts, P., Langin, T., and Dron, M. 2000. Inheritance of partial resistance against
Colletotrichum lindemuthianum in Phaseolus vulgaris and co-localization of
quantitative trait loci with genes involved in specific resistance. Mol. P1. Microbe
Interact. 13:287-296.

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

Gepts P., Osborn, T.C., Rashka, K., and Bliss, RA. 1986. Phaseolin protein variability in
wild forms and landraces of the common bean (Phaseolus vulgaris): evidence for
multiple centers of domestication. Econ. Bot. 40:451-468.

Gepts, P. 1990. Biochemical evidence bearing on the domestication of Phaseolus
(Fabaceae) beans. Econ. Bot. 44:28-38.

Gepts, P. and Debouck, D.G. 1991. Origin, domestication and evolution of the common
bean, Phaseolus vulgaris. Pages 7-53. In: Common Beans: Research for Crop
Improvement. A. van Schoonhoven and O. Voysest (eds). CAB. International,
Wallingford, UK. '

28

Grant, M.R., Godjard, L., Straube, E., Ashfield, T., Lewald, J ., et al. 1995. Structure of
the Arabidopsis RPM] gene enabling dual specificity disease resistance. Science
269:843—846.

Guzman, P., Donado, M.R., and Galvez, GE. 1979. Economic losses due to bean
(Phaseolus vulgaris) anthracnose in Columbia. Turrialba 29:65-69.

Haley, S.D., Afanador, L.K., and Kelly, JD. 1994. Selection for monogenic resistance
traits with coupling- and repulsion-phase RAPD markers. Crop Sci. 34: 1061—1066.

Heath, M.C. 1998. Apoptosis, programmed cell death and the hypersensitive response.
Eur. J. Plant Path. 104:117—124.

Heath, M.C. 2000. Hypersensitive response-related death. Plant Molecular Biology
44:321—334.

Huang, N., Angeles, E.R., Domingo, J., Magpantay, G., Singh, S. Zhang, G.,
Kumaravadivel, N., Bennett, J ., and Khush, GS. 1997. Pyramiding of bacterial blight
resistance genes in rice: marker-assisted selection using RFLP and PCR, TAG Theor.
Appl. Genet. 95:313-320.

Johansen, I.E., Lund, O.S., Hjulsager, OK, and Laursen, J. 2001. Recessive resistance in
Pisums ativum and potyvirus pathotype resolved in a gene-for-cistron correspondence
between host and virus. Journal of Virology 75:6609-6614.

Jakobek, J .L. and Lindgren, PB. 1993. Generalized induction of defense responses in
bean is not correlated with the induction of the hypersensitive reaction. Plant Cell
5:49-56.

Jones, J .D.G. 2001. Putting knowledge of plant disease resistance genes to work. Curr.
Opin. Plant Biol. 4:281—287.

Kami, J., Velasquez, V.B., Debouck, D.G., and Gepts, P. 1995. Identification of
presumed ancestral DNA sequences of Phaseolin in Phaseolus vulgaris. Proc. Natl.
Acad. Sci. USA 92:1101-1104.

Kelly, J .D. and Miklas, RN. 1998. The role of RAPD markers in breeding for disease
resistance in common bean. Mol. Breed. 4:1-11.

Kelly, J.D., Gepts, P., Miklas, RN, and Coyne, DP. 2003. Tagging and mapping of

genes and QTL and molecular marker-assisted selection for traits of economic
importance in bean and cowpea. Field Crops Res. 82: 135-154.

Kelly, JD. and Vallejo, VA. 2004. A comprehensive review of the major genes
conditioning resistance to anthracnose in common bean. Hort Sci. 39: 1 196-1207.

29

Koenig, R., and Gepts, P. 1989. Allozyme diversity in wild Phaseolus vulgaris: further
evidence for two major centers of genetic diversity. Theor. Appl. Genet. 78:809-817.

Koinange, E.M.K., and Gepts, P. 1992. Hybrid weakness in wild Phaseolus vulgaris L. J.
Hered. 83: 135—139.

Kombrink, E. and Schmelzer, E. 2001. The Hypersensitive Response and its Role in
Local and Systemic Disease Resistance. Eur. J. Plant Path. 107:69-78.

Lawrence, G.J. and Burdon, J .J. 1989. Flax rust from Linum marginale: variation in a
natural host-pathogen interaction. Can. J. Bot. 67:3192-3198.

Lenne, J .M. 1992. Colletotrichum disease of legumes. Pages 289-307. In: Colletotrichum:
Biology, Pathology and Control. J.A. Bailey and M.J. Jeger (eds.). CAB.
International, Wallingford, UK.

Mahuku, GS. and Riascos, J.J. 2004. Virulence and molecular diversity within
Colletotrichum lindemuthianum isolates from Andean and Mesoamerican bean
varieties and regions. Eur. J. Plant Path. 110:253-263.

Martin, G.B. et al. 2003. Understanding the functions of plant disease resistance. Annu.
Rev. Plant Biol. 54:23—61.

Melotto, M., Vallejo, V.A., Awale, H., and Kelly, JD. 2000. Use of RAPD and AFLP
analysis to tag the Co-I gene conditioning resistance to bean anthracnose. Annu.
Rept. Bean Improv. Coop. 43:84-85.

Melotto, M. and Kelly, JD. 2001. Fine mapping of the C0-4 locus of common bean
reveals a resistance gene candidate, COX-4 that encodes for a protein kinase. Theor.
Appl. Genet. 103:508-517.

Melotto, M., Coelho, M.F., Pedrosa—Harand, A., Kelly, J .D., and Camargo, L.E.A. 2004.
The anthracnose resistance locus Co-4 of common bean is located on chromosome 3

and contains putative disease resistance—related genes. Theor. Appl. Genet. 1092690-
699.

Méndez-Vigo, B., Rodriguez-Suarez, C., Paneda, A., Ferreira J., and Giraldez R. 2005.
Molecular markers and allelic relationships of anthracnose resistance gene cluster B4
in common bean. Euphytica (submitted).

Mindrinos, M., Katagiri, F., Yu, G.L., and Ausubel, FM. 1994. The A. thaliana disease

resistance gene RPSZ encodes a protein containing a nucleotide-binding site and
leucine-rich repeats. Cell 78: 1089—1099.

30

Nevo, E. 1986. Genetic resources of wild cereals and crop improvement: Israel, a natural
laboratory. Israeli J. Bot. 35:255-278.

O’Connell, R.J. and Bailey, J .A. 1991 Hemibiotrophy in Colletotrichum lindemuthianum.
Pages 211-222. In: Electron Microscopy of Plant Pathogens. K. Mendgen and DE.
Leseman (eds.). Springer, New York.

O’Connell, R.J., Perfect, S.E., Hughes, H.B., Carzaniga, R., Bailey, J .A. and Green, J .R.
2000. Dissecting the cell biology of Colletom'chum infection processes. Pages 57-77.
In: Colletotrichum - Host Specificity, Pathology, and Host—Pathogen Interaction. D.
Prusky, S. Freeman, and M. Dickman (eds.). A.P.S. Press, St. Paul, Minnesota.

Parker, M.A. 1985. Local population differentiation for compatibility in an annual
legume and its host-specific fungal pathogen. Evolution 39:713-723.

Pastor-Corrales, M.A. and Tu, J.C. 1989. Anthracnose. Pages 77-104. In: Bean
Production in the Tropics. H.F. Schwartz and M.A. Pastor-Corrales (eds.). CIAT,
Cali, Colombia.

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

Pastor-Corrales, M.A., Otoya, M.M, Molina, A., and Singh, SP. 1995. Resistance to
Colletotrichum lindemuthianum from Middle America and Andean South America in
different common bean races. Plant Dis. 79:63-67.

del P020, 0. and Lam, E. 1998. Caspases and programmed cell death in the
hypersensitive response of plants to pathogens. Curr. Biol. 8:1129—1132.

Pryor, T. and Ellis, J. 1993. The genetic complexity of fungal resistance genes in plants.
Adv. Plant Pathol. 10:281-305.

Rivkin, M.I., Vallejos, CE, and McClean, RE. 1999. Disease-resistance related
sequences in common bean. Genome 42:41—47.

Sicard, D., Michalakis, Y., Dron, M., and Neema, C. 1997. Genetic diversity and
pathogenic variation of Collerotrichum lindemuthianum in the three centers of
diversity of its host, Phaseolus vulgaris. Phytopathology 87:807-813.

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

Sonnante, G., Stockton, T., Nodari, R.O., Becerra—Velasquez, V.L., and Gepts, P. 1994.

Evolution of genetic diversity during the domestication of common bean (Phaseolus
vulgaris L.). Theor. Appl. Genet. 89:629-635.

31

TeBeest, D.O., Correll, J.C., and Weidemann, G.J. 1997. Speciation and population
biology in Colletotrichum lindemuthianum. Pages 157-165. In: The Mycota V — Plant
Relationships (Part B). Carroll and Tudzynski (eds.). Springer-Verlag, Berlin,
Germany.

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

Tu, J .C. 1982. Effect of temperature on incidence and severity of anthracnose on white
bean. Plant Dis. 66:781-783.

Tu, J.C. 1992. Colletotrichum lindemuthianum on beans: population dynamics of the
pathogen and breeding for resistance. Pages 204-224. In: Colletotrichum - Biololgy,
Pathology and Control. J.A. Bailey and M.J. Jeger (eds.). CAB. International,
Wallingford, UK.

Young, RA. and Kelly, JD. 1996. Characterization of the genetic resistance to
Colletotrichum lindemuthianum in common bean differential cultivars. Plant Dis.
80:650-654.

Yu, I.C., Parker, J ., and Bent, A.F. 1998. Gene-for-gene disease resistance without the

hypersensitive response in Arabidopsis dndI mutant. Proc. Natl. Acad. Sci. USA
95:7819—7824.

32

CHAPTER ONE

MOLECULAR TAGGING or THE 00-12 LOCUS AND GENETIC
CHARACTERIZATION OF THE ANTHRACNOSE RESISTANCE IN ANDEAN
BEAN CULTIVAR JALO EEP558

ABSTRACT

The anthracnose resistance gene, Co-I, is a complex multiallelic locus, with 4
characterized alleles (Co-1 - Co-I‘) that has been previously mapped to bean linkage
group B1 on the core bean (Phaseolus vulgaris) integrated linkage map. The objectives
of this study were to develop a molecular marker linked to the Co-IZ gene, from the
Andean anthracnose differential cultivar Kaboon, and to characterize the anthracnose
resistance in the Andean cultivar Jalo EEP558. Two strategies were employed to find a
marker linked to the Co-Iz locus (1) screen random AFLP primer pair combinations
using bulked segregant analysis, (2) screen RAPD and SSR primers located on linkage
group B1 for linkage to Co-I. The AFLP marker, EAchCCA-108/107 was identified,
linked at 9.9 cM from the Co-Iz allele. An STS marker, SEAchCCA, was derived from
EACFMCCA-108/ 107. This marker is not allele specific, but the STS and AFLP markers
appear to be gene pool specific. The low level of polymorphism observed near the Co-I
locus could be due to its close proximity to a block of highly conserved domestication
genes. The anthracnose resistance in the Andean cultivar J alo EEP558 was characterized
using a combination of traditional and molecular techniques. Inoculation with various
races of C. lindemuthianum revealed that Jalo EEP558 carried an allele of the Co-I locus,
most likely the Co-I allele. The inoculation of the F2 population, Jalo EEP558/Perry

Marrow (Co-I3), with race 357, which yields an R/S (resistant/susceptible) reaction in the

33

parents, revealed that resistance to race 357 in Jalo EEP558 is conditioned either by one
dominant and one recessive gene with epistasis or one dominant gene. The allelism test
was conducted in the Jalo EEP558/MDRK F2 population which was inoculated with race
73. This population of 200 individuals did not show segregation for resistance suggesting

that resistance to race 73 is conditioned by an allele of the Co-I locus.

34

INTRODUCTION

Anthracnose, caused by the hemibiotrophic fungus Colletotrichum
lindemuthianum, is considered the most economically important disease of common bean
worldwide. This is due, in part, to the seed-home nature and pathogenic variability of C.
lindemuthianum. Genetic resistance is the best form of control, especially in areas of
subsistence farming, commercial seed production and the certified seed industry.
Pyrarniding of anthracnose resistance genes of Andean and Middle American origin has
been proposed as a strategy for broad-based durable resistance breeding (Young and
Kelly, 1996a; Young and Kelly, 1996b). Many Middle American anthracnose resistance
genes have been characterized, Co-Z through Co-IO, (Kelly and Vallejo, 2004), however,
only Co-I, Co-x and Co-y are from Andean origin. New sources of resistance, therefore,
are always being sought to provide genetic resources for anthracnose resistance breeding.

Co-I is a complex locus, with 4 characterized alleles (Kelly and Vallejo, 2004),
each conferring resistance to a different range of C. lindemuthianum races. The Andean
differential cultivar, Kaboon, was shown to be highly resistant to anthracnose, only
defeated by two of 40 races tested (Balardin et a1, 1997). The major dominant gene
conditioning resistance in Kaboon is reported to be Co-Iz, an allele of the Co-I gene
(Melotto and Kelly, 2000). The Andean bean cultivar, Jalo EEP558, used as a parent in
the mapping population of the bean integrated map, also demonstrates resistance to
anthracnose (Freyre et al., 1998; Geffroy et al., 1999) and may provide a new source of

resistance in the Andean gene pool.

35

Molecular markers linked to the Co-I locus have been previously identified
(RAPD: Young and Kelly, 1997a; AFLP: Mendoza et al., 2001). These markers are
linked in repulsion phase and have proven to be difficult to reproduce between
laboratories. Mendez de Vigo (2001) used the repulsion marker F10530 to map the Co-I
locus to bean linkage group B1. With the advent of the linkage map position of the Co—I
locus, a series of RAPD (Freyre et al., 1998) and SSR (simple sequence repeat) (Blair et
al., 2003) markers located near the locus can now be evaluated for linkage to Co-I. The
availability of other more reproducible markers linked in coupling phase would greatly
facilitate the introgression of the Co-I gene into anthracnose resistance gene pyramids
using marker-assisted selection (MAS).

Previously, a total of 679 random primers tested in the Cardinal
(susceptible)/Kaboon F2 p0pulation, only 15 were polymorphic between DNA bulks and
none were found to be linked to the Co—I gene (Melotto et al., 2000b). Based on this low
level of polymorphism that was found with RAPD analysis in this population, the AFLP
technique (Vos et al., 1995) was chosen to develop a molecular marker linked to the Co-
I2 gene. It is possible that the Co-I gene is flanked by tightly conserved regions with low
levels of polymorphisms, which could not easily be detected by the RAPD technique.
Additionally, studies have indicated that AFLP analysis is more sensitive to lower levels
of polymorphisms than RAPD analysis (Sharma et al., 1996; Mackill et al., 1996). To
specifically target the genomic region containing the Co-I gene, bulked-segregant
analysis (BSA) (Michelmore et al., 1991) was used. The objectives of this study were to

identify a molecular marker linked to the Co-I locus using RAPD, AFLP, SSR and BSA

36

analyses and to characterize the anthracnose resistance in the Andean cultivar Jalo

EEPSSS.

MATERIALS AND METHODS

Plant material and C. lindemuthianum races. A population segregating for Co-
12 was used from the cross between Cardinal, a universally susceptible genotype, and
Kaboon; this population was designated C/K. Seeds were planted in the greenhouse and
tissue was collected from the seedlings at the primary leaf stage for DNA extraction.
Approximately ten days after germination, seedlings were spray inoculated with race 73.

The Andean cultivar Jalo EEP558 was inoculated with a series of C.
lindemuthianum races (Table 1.1). Additionally J alo EEP558 was crossed with Michigan
Dark Red Kidney (MDRK) which carries the Co-I gene and Perry Marrow (PM) which
carries Co-13. The J alo EEP558/PM segregating F2 population was inoculated with race
357, which yields an R/S (resistant/susceptible) reaction in the parents, to determine the
number of anthracnose resistance genes in Jalo EEP558. The segregating F2 population
from the Jalo/MDRK cross was used to test for allelism by inoculation with race 73

which is avirulent on both parents.

Inoculum preparation. All races utilitzed in this study were grown from
monosporic cultures maintained on filter papers and stored at -20° C. Inoculum was
prepared by growing the fungus from a filter paper on either potato dextrose agar (PDA)

or Mathur’s agar (dextrose (8 g/L), MgSO4. 7 H2O (2.5 g/L), neopeptone (2.4 g/L), yeast

37

extract (2 g/L), and agar (16 g/L)) at 22°C in the dark. When the cultures began to
sporulate, spores were re-plated in replicate to generate sufficient inoculum. Spores were
suspended in a solution of water and 0.01% Tween 20. Spore concentration was
estimated using a hemacytometer and the concentration was adjusted to 1.2x106
spores/mL. The prepared inoculum was then sprayed onto the abaxial and adaxial
surfaces of the leaves. Inoculated seedlings were incubated at approximately 100%
relative humidity for 48 hours. Seedlings were rated for disease 5 days post inoculation
on a scale described by Balardin et al. (1997). All races used in this study were first

inoculated on the anthracnose differential series to confirm their identities.

AFLP and BSA analyses. Contrasting DNA bulks were constructed by pooling
equal volumes of fluoremetrically standardized DNA concentrations from five resistant
individuals and five susceptible individuals from the segregating population. The DNA
bulks will be referred to as R and S bulks, for resistant and susceptible respectively. The
AFLP analysis was conducted as described Hazen et al. (2002). Genomic DNA from
parents and bulks was first digested with EcoRI and MseI. Adapters which served as
subsequent priming sites were ligated to the sticky-ends followed by two PCR
amplifications; pre-amplification was performed with one selective base on each primer
and selective amplification was performed using primers with three selective nucleotides.
Final PCR products were separated on 6% polyacrylamide gels and visualized with silver

staining (Promega, Madison, WI).

38

STS development. The band of interest from each of the parents were cut from a
fresh gel using a sterile scalpel. The gel slice was then placed in a microcentrifuge tube
with 50 ul of water and incubated overnight at 4°C to diffuse the DNA template.
Selective PCR was done with the original selective primer combination using 5 pl of the
template DNA to re-amplify the fragment. Various volumes (3, 5, and 101.11) of template
DNA were tested to optimize amplification (data not shown). PCR products were then
separated on a 4% NuSieve GTG agarose gel (FMC BioProducts, Rockland, ME) made
with lXTBE and an ethidium bromide concentration of 0.2 mg/mL. The band was re-
isolated from the agarose gel to purify the fragment using a QIAquick gel extraction kit
(QIAGEN, Valencia, CA). The purified fragment was then cloned using the TOPO TA
cloning kit (Invitrogen, Carlsbad, CA). Plasmid DNA was extracted from transformed
bacterial colonies using the Wizard plus SV miniprep kit (Promega, Madison, WI). The
plasmid DNA was digested using EcoRI and run on a 6% polyacrylamide gel for size
selection. Selected colonies containing the inserts of appropriate size were then
sequenced at the Genomics Technology Support Facility located at Michigan State
University. Sequences were aligned using ClustalW (Thompson et al., 1994) and co-
dominant sequence specific primers were designed. The primers were tested on the
parents, individuals from the segregating population in addition to other genotypes
containing various alleles of the Co-I gene and the parents of the mapping population
Bat193 and Jalo EEPSS8. The PCR reaction was as described by Melotto et al. (1996).
The thermal cycler profile used was: 3 min at 94 ° C; 30 cycles of 10 s at 94° C, 30 s at

52° C, 2 min at 72° C; 5 min at 72° C.

39

RAPD analysis. RAPD markers which map near the Co-I locus on linkage
group B1 were screened on various Co-I carrying genotypes (Table 1.2) to determine if
any were candidate markers putatively linked to Co-I. The RAPD analysis was similar
that described by Haley et al. (1993) using Taq DNA polymerase (Invitrogen, Carlsbad,
CA) and Operon primers (Operon Technologies, Inc., Alameda, CA). The thermal cycler
profile used was 3 cycles of 1 min at 94° C, 1 min at 35° C, 2 min at 72° C; 34 cycles of
10 s at 94° C, 20 s at 40° C, 2 min at 72° C; 5 min at 72° C. PCR products were run on

1.4% agarose gels using 1X TAE and 0.2 mg/mL of ethidium bromide.

SSR analysis. SSR markers located on linkage group B1 (Table 1.3) were tested
on parents and R and S bulks. The PCR reaction contained 1.5 mM MgCl2, 0.2 mM
dNTP mix, 0.1 M of each primer, 1 U Taq DNA polymerase (Invitrogen, Carlsbad, CA)
and 50 ng DNA. The thermal cycler profile used was 5 min at 94° C; 30 cycles of 1 min
at 94° C, 1 min at 47° C, 1 min at 72° C; 5 min 72° C. PCR products were separated on

6% polyacrylamide gels as described in AFLP analysis above.

Molecular Mapping. Candidate AFLP markers and the STS marker,
SEAchCCA, were tested for linkage to resistance on the C/K segregating population
using MAPMAKER (Lander et al. 1987). Linkage distance in cM was calculated using
Kosambi’s mapping function with a LOD threshold value of 2.0. SEAchCCA was also
mapped to the bean integrated linkage map using 65 recombinant inbred lines (RILs).
These RILs are a subset of those used originally to generate the integrated map from the

F2 population BAT93/Jalo EEP558 (Freyre, 1998).

40

 

RESULTS AND DISCUSSION

RAPD, SSR and AF LP analyses. One method employed to identify markers
linked to the Co-I locus was to test markers located near the previously mapped F1053o
RAPD marker (Young and Kelly, 1997) linked to the Co-I locus. Nine RAPD and seven
SSR markers were tested for putative linkage to the Co-I locus using parents, R and S
bulks in addition to several genotypes carrying varying alleles at the Co-I locus. None of
the RAPD and SSR markers tested, however, appeared to be associated with the presence
of the Co-I locus (Data not shown). A total of 232 AFLP primer pair combinations were
tested for linkage with the Co-I locus. 22 primer combinations (10.5% of all primers
tested) were polymorphic between parents and R and S bulks, however, only one,
EAchCCA, showed putative linkage when tested on the individual members of the R and
S bulks (Figure 1.1A). Two bands of interest are produced among the PCR products
which result from the EACTMCCA primer combination. The larger band is 108 bp in
length and is present in Kaboon and the R bulk, suggesting a putative linkage to the
resistance conferred by the Co-Iz gene. The smaller band is present in the susceptible
parent, Cardinal, and the S bulk indicating that it may be linked to the susceptible allele at
the Co-I2 locus. The linkage analysis revealed that this marker, EAchCCA-108/107 was
linked at a distance of 9.9 cM from the Co-I locus and segregated in a 1:2:1 (p=0.478)
ratio indicating that it is a co-dominant marker. The marker was tested on cultivars
BAT93 and Jalo EEP558, the parents of a mapping population and genotypes with

different Co-I alleles (Figure 1.1B). The band associated with the Co-I locus (108 bp),

41

however, was not amplified in these genotypes possibly due to mutations in the
restriction sites of the enzymes and thus, EAchCCA-108/107 appears to be specific to the

Co-I2 allele of the Co-I locus.

STS marker. AFLP analysis is relatively complicated and costly, requiring
extensive template preparation. Consequently, AFLP markers are not well suited for
marker-assisted selection (MAS) of large populations in breeding programs. To facilitate
the use of MAS to introgress the Co-Iz gene into susceptible cultivars, the AFLP marker
was converted to a sequence-tagged site (STS) marker. Figure 2 shows the sequence
alignment of the resistant and susceptible alleles of the marker. The alignment of the
sequences revealed that the two bands differ only by two changes; a one bp deletion
which results in a fragment length difference and a single nucleotide polymorphism. Due
to the small size of the original AFLP fragments, only 107 and 108 bp, the derived STS
marker, SEAchCCA, was also very small, 79 and 80 bp. The co-dominant nature and the
small fragment length difference between alleles of the SEAchCCA marker requires the
use of polyacrylamide gels as the method to resolve the PCR products. This requirement
and the distance of the marker from the locus, 9.9 cM, makes it less suitable for use in
MAS.

The derived STS marker was also tested on genotypes which carry different
alleles at the Co-I locus (Figure 1.2). The allele of SEAchCCA which is associated with
resistance (108 bp), is also present in cultivars MDRK (Co-1) and Perry Marrow (Co-13).
Thus, in contrast to the AFLP marker, EACTMCCA-108/107, which appears to be allele

specific, the derived STS marker is not allele specific. This is most likely because the

42

STS primers do not include the entire restriction enzyme site, therefore, mutations in
these sites do not affect the amplification of SEACTMCCA. Although EACTMCCA-108/107
and SEACTMCCA are not suitable for MAS, these two markers can be used to initiate the
characterization of Andean sources or anthracnose resistance. SEAchCCA can be used to
detect the presence of resistant alleles at the Co-I locus; EAchCCA-108/107 can be used
to identify the Co-Iz allele within Andean germplasm. No association was detected
between the presence of either marker and the Co-I locus among Middle American
genotypes tested (Data not shown) and thus the use of these markers is believed to be
limited to Andean gerrnplasm.

To confirm the genomic location of the Co-I locus, SEACTMCCA was mapped
using a RIL population derived from the cross between genotypes Bat93 and Jalo
EEP558. No amplification product was observed in BAT93; this is not surprising
considering BAT93 is a Middle American cultivar. Jalo EEP558 presented the 80 bp
fragment, therefore the marker was scored in a dominant fashion rather than co-dominant.
The linkage analysis confirmed the placement of the Co-I locus on linkage group Bl
(Figure 1.3). Based on linkage of Co-I with the RFLP clones used to construct and
integrate the physical map with the genetic linkage map of common bean (Pedrosa et al.,

2003), the Co-I locus is located on the long arm of chromosome 2.

Polymorphism at Co-I genomic region. The region surrounding the Co-I locus
appears to have limited variability at the DNA level. Support for this observation comes
from the small number of polymorphisms detected between the R and S bulks, only 10.5

% of 232 primers tested. One theory is that this is a result of the genomic location of the

43

 

Co-I locus. The Co-I locus is located on the distal end of linkage group B1, adjacent to
the largest block of genes which control the domestication syndrome in the bean genome
(Koinange et al., 1996). Blocks of domestication genes, much like co-adaptive gene
complexes, are highly conserved. The population used in this study was between two
domesticated bean lines, thus little variability between the parents is expected at the block
of domestication genes. Since the Co-I locus maps adjacent to this block of highly
conserved genes, it is not surprising that reduced polymorphism was detected

surrounding this locus.

Characterization of anthracnose resistance in Jalo EEP558. To characterize
the anthracnose resistance in the Andean cultivar Jalo EEP558, Jalo EEP558 was first
screened with markers linked to other known anthracnose resistance genes (Table 1.4).
Only SEACTMCCA 80 was present in J alo EEP558, indicating that Jalo EEP558 carries the
Co-I gene. The disease reaction of Jalo EEP558, when inoculated with 10 diverse races
of C. lindemuthianum, is identical to that of MDRK (Table 1.1). The comparison of the
disease reaction of J alo EEP558 with that of other Co-I containing cultivars supports the
hypothesis that Jalo EEP558 carries the Co-I locus and suggests that it is the same allele
present in the differential cultivar MDRK. To determine the number of anthracnose
resistance genes segregating from Jalo EEP558, 94 plants from the Jalo EEP558/PM F2
population were inoculated with race 357, which gives an R/S (resistant/susceptible )
reaction in the parents (Table 1.5). The segregation ratio of resistantzsusceptible
individuals of the J alo EEP558/PM population fits both a 13:3 ratio (p-value = 0.667) and

a 3:1 ratio (p-value = 0.074). The population size used in this experiment prevents us

from distinguishing between a 13:3 or a 3:1 ratio. A segregation ratio of 13:3 suggests
that J alo EEPSSB carries one dominant and one recessive gene with epistasis. A ratio of
3:1 suggests that only one dominant gene conditions resistance to race 357. Therefore,
we are unable to determine if Jalo EEP558 carries more than one gene which conditions
resistance to anthracnose. Interestingly, Geffroy et al. (1999), identified two putative
Andean anthracnose resistance genes (Co-y and Co-z) from J alo EEP558. These loci map
to a gene cluster which contains the Middle American resistance gene Co-9 on linkage
group B4 of the integrated map. This gene cluster, comprising both Andean and Middle
American anthracnose resistance genes, suggests that this locus existed prior to the
separation of the two major gene pools of P. vulgaris. The authors did not characterize
their fungal strains by the universal anthracnose differential series and thus the results of
this study cannot be compared to other studies.

To test the anthracnose resistance in Jalo EEP558 for allelism with the Co-I
locus, 200 individuals from the J alo EEP558/MDRK F2 population were inoculated with
C. lindemuthianum race 73, which elicits an R/R (resistant/resistant) reaction in the
parents (Table 1.5). No segregation in the progeny was observed indicating that Jalo

EEP558 carries the Co-I gene or a cluster gene family member.

CONCLUSIONS

The Co-I locus offers breeders choices among alleles for anthracnose resistance,

however, other Andean resistance genes must be characterized to provide diverse tools

necessary for developing more durable anthracnose resistance. Although many Andean

45

sources of anthracnose resistance have been characterized, no new resistance loci are
being identified. This begs the question: Why are there so many independent
anthracnose resistance genes in the Middle American gene pool when compared to the
Andean gene pool? The answer to this interesting question may lie in an understanding
of the domestication of P. vulgaris. Common bean is a noncentric crop, having two or
more domestication events in distinct wild populations leading to the formation of two
geographically separated gene pools of domestication. The cultivated Andean gene pool,
however, has a very narrow genetic base when compared to Middle American. Beebe et
al. (2001) suggest that the three Andean bean races share a common origin based on the
multiple correspondence analysis of AFLP genotyping of landrace, wild and cultivated
beans of Andean and Middle American origin. This narrow genetic base suggests that
the domestication of Andean beans occurred within a narrow wild population (Beebe et
al., 2001). This has direct implications on the diversity of independent anthracnose
resistance genes available within the cultivated Andean gene pool. The identification of
multiple resistance genes in the Andean cultivar Jalo EEP558, suggests that
uncharacterized resistance to anthracnose in the Andean gene pool remains to be
discovered.

The apparent deficiency of polymorphism at the molecular level in the genomic
region of the Co-I locus, which may be a result of close linkage to a block of genes
contributing to the domestication syndrome, has made the identification of markers
linked to this locus very difficult. To circumvent this problem a population derived from

the cross of a cultivated by a wild bean should be used. Molecular markers linked to the

46

Co-I locus which can differentiate between all alleles will be useful to facilitate the

introgression of anthracnose resistance conditioned by this complex locus.

47

Table 1.1 Disease reaction3 of J alo EEP558 and three differential cultivars to 10 races of

C. lindemuthianum.

 

 

Race Jalo EEP558 MDRK" PM° Kaboon“
2 s s R R
7 s s s R
38 s s s s
47 s s s s
73 R R R R
80 R R R R
88 R R R R
128 R R R R
357 R R s s
1545 R R R R

 

a Disease reaction: R = resistant, S = susceptible
b Co-I
c Co-13

d Co-I2

48

Table 1.2 RAPD markers located near the mapped location of the Co-I locus tested on

bean cultivars with different anthracnose resistance genes.

 

 

RAPD Markers Bean Cultivars Anthracnose Resistance Gene

14400 Cardinal No known anthracnose resistance genes
D201000 La Victoria No known anthracnose resistance genes
H31 100 Kaboon Co- [2

D8800 MDRK Co-I

A13625 Perry Marrow C0-13

F6400 Cornell 49-242 C0-2

A4700 BAT93 C0-9

D6800 . J alo EEP558 Characterized in current study

F6800

 

49

Table 1.3 SSR markers evaluated for putative linkage with the Co-I locus using cultivars

Cardinal and Kaboon and R and S bulks.

 

SSR markers (expected size bp)
BM146 (281)b

BM 157 (113)b

BMd-IO (139)a

Pv-ag003 (201)C

BM 53 (287)b

EM 200 (221)b

BMd-45 (129)a

 

3‘ Primer sequences are available in Blair et al. (2003).
b Primer sequences are available in Gaitan-Solr’s et al. (2002).

c Primer sequences are available in Yu et al. (2000).

50

 

Table 1.4 Molecular markers linked to known anthracnose resistance genes tested on J alo

 

 

EEP558.

Molecular Marker Linked Anthracnose Locus Reference

SAB3500 C0-5 Chapter Three

SH181200 C0-42 (Awale and Kelly, 2001)
SAS 13950 C0-4 (Young et al., 1998)
SEAchCCA Co-I Present Chapter

 

51

Table 1.5 F2 populations inoculated with C. lindemuthianum.

 

 

 

Population Race Disease Observed Expected p-value
Inoculated Reaction Ratioa Ratio“
Jalo EEP558/PM 357 R/S 78:16 13:3 0.667
3:1 0.074
J alo EEP558/MDRK 73 R/R 200:0 - -

 

3 Ratio of resistant to susceptible individuals

52

 

A)
1234567891011121314

,
4

mm w m 1"“
i .

l
T ‘

 

Figure 1.1 Amplification products of the AFLP marker EAchCCA-108/107 from various
bean genotypes using A) (1) Cardinal, (2) Kaboon, (3) susceptible bulk, (4) resistant
bulk, (5-9) individuals of the resistant bulk, (10—14) individuals of the susceptible bulk,
B) (1) 110 bp fragment of the 10 bp ladder, (2) Kaboon, (3) Cardinal, (4) Jalo EEP558,

(5) BAT93, (6) MDRK, (7) Perry Marrow.

53

80 bp

Figure 1.2 Amplification products of the STS marker, SEAchCCA from various bean
genotypes using (1) MDRK (Co-1), (2) BAT93 (Co-9), (3) Jalo EEP558 (Co-1), (4) Perry
Marrow (Co-13), (5) Kaboon (Co-12), (6) Cardinal (no known anthracnose resistance

genes).

54

B1

 

_ ..srs
9.8

J —Bngl7la
8.9

— —Bng173
8.6

__J _Bng122
3‘5 — —wwo7.1
7.4

_ w013.2
3'9 _ wv20.3
8.4
5.4 — —WQ03.1

_ —WP01.1
5'4 _ _ww09.2
9.2

_ _DROD13
6.6

_ —D1662
5'5 _ _Dl327
7.1

— —wor6.3
2.0 _ _
4.2 w120.1
5.3 WR20.1

— —W'mr.5
10.7

_ ..DRours

Figure 1.3 The STS marker SEACTMCCA was mapped to bean linkage group Bl using the

BAT93/Jalo EEP558 RIL population.

55

LITERATURE CITED

Awale, HE. and Kelly, J .D. 2001. Development of SCAR markers linked to C0-42 gene
in common bean. Ann. Rep. Bean Improv. Coop. 44: 1 19-120.

Balardin, R.S., Jarosz, A., and Kelly, JD. 1997. Virulence and molecular diversity in
Colletotrichum lindemuthianum from South, Central and North America.
Phytopathology 87:1 184-1191.

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

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

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

Gaitan-Solr’s, E., Duque, M.C., Edwards, K.J., and Tohme, J. 2002. Microsatellite repeats
in common bean (Phaseolus vulgaris L.): isolation, characterization, and cross-
species amplification in Phaseolus ssp. Crop Sci. 42:2128—2136.

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

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

Hazen, S.P., Leroy, P., and Ward, R.W. 2002. AFLP in Triticum aestivum L.: patterns of
genetic diversity and genome distribution. Euphytica 25:89.

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

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

56

 

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

Mackill, D., Zhang, Z., Redona, E., and Colowit, P. 1996. Level of polymorphism and
genetic mapping of AFLP markers in rice. Genome 39: 969-977.

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

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

Melotto, M., Vallejo, V., Awale, H., and Kelly, JD. 2000. Use of RAPD and AFLP
analysis to tag the Co-l gene conditioning resistance to bean anthracnose. Ann.
Rep. Bean Improv. Coop. 43:84-85.

Méndez de Vi go, B. 2001. Mapa genético de Phaseolus vulgaris L. y resistencia a
antracnosis en faba granja asturiana. Ph.D. Thesis, University of Oviedo, Oviedo,
Spain, 119pp.

Mendoza, A., Hernandez, F., Hernandez, S., Ruiz, D., de la Vega, O.M., de la Mora, G.,
Acosta, J ., and Simpson, J. 2001. Identification of Co-I anthracnose resistance and
linked molecular markers in common bean line A193. Plant Dis. 85:252-255.

Michelmore, R.W., Paran, I., and Kesseli, R.V. 1991. Identification of markers linked to
disease-resistance genes by bulked segregant analysis: a rapid method to detect

markers in specific genomic regions by using segregating populations. Proc. Natl.
Acad. Sci. 88:9828-9832.

Sharma, S.K., Knox, M.R., and Ellis, T.H.N. 1996. AFLP analysis of the diversity and
phylogeny of Lens and its comparison with RAPD analysis. Theor. Appl. Genet. 93:
751-758.

Thompson, J.D., Higgins, D.G., and Gibson, T.J. 1994. CLUSTAL W: improving the
sensitivity of progressive multiple sequence alignment through sequence weighting,
position-specific gap penalties and weight matrix choice. Nucleic Acids Res.
22:4673-4680.

 

Vos, R, Hogers, R., Bleeker, M., Reijans, M., Van de Lee, T., Homes, M., Frijters, A., Pot, J.,
Peleman, J ., Kuiper, M., and Zabeau, M. 1995. AFLP: a new concept for DNA

fingerprinting. Nucleic Acid Research. 23:4407-4414.

57

Young, R.A., and Kelly, J.D. 1996a. Characterization of the genetic resistance to
Colletom'chum lindemuthianum in common bean differential cultivars. Plant Dis.
80:650-654.

Young, R.A., and Kelly, J.D. 1996b. RAPD markers flanking the Are gene for
anthracnose resistance in common bean. HortScience 121:37-41.

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

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

Yu, K., Park, S.J., Poysa, V., and Gepts, P. 2000. Integration of simple sequence repeat

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

58

CHAPTER TWO

MOLECULAR AND GENETIC CHARACTERIZATION OF THE C0-7 LOCUS
CONDITIONING RESISTANCE TO ANTHRACNOSE IN COMMON BEAN
AND BREEDING LINE MSU7

ABSTRACT

The most resistant cultivar in the anthracnose differential series is the landrace G
2333. G 2333 is known to carry a three gene pyramid for anthracnose resistance: Co-42,
Co-5 and C0-7. Both Co-5 and Co-42 genes have been well characterized through the use
of lines derived from crosses with G 2333 that carry these individual genes. The
objectives of this study were to characterize the spectrum of resistance conditioned by the
C0—7 gene and to develop a molecular marker linked to this resistance gene. In order to
characterize resistance conferred by this locus, the Co-7 gene had to be separated from
the other anthracnose resistance genes. To this purpose, the F2 population from the cross
Black Magic (susceptible)/Sel 111 (resistant and carries C0-5 and Co-7) was inoculated
with C. lindemuthianum race 7. The resistant individuals would carry Co-7 (homozygous
or heterozygous) and/or Co-5 (homozygous or heterozygous). To select against the Co-5
gene, in the absence of a race to discriminate between C0-5 and Co-7, the SCAR marker,
SAB3 was developed from the RAPD marker AB3450 which is linked to the Co-5 locus.
F3 progeny tests conducted with race 7 identified those individuals which were
homozygous at the Co-7 locus and these were designated MSU7-1 through -6. I
confirmed that MSU7 carried only one gene conditioning resistance to race 7 by crossing

MSU7 with MDRK and inoculating the F2 (3:1 segregation ratio with p—value = 0.82).

59

 

Also, lines MSU7-3, MSU7-4 and MSU7-6 were all resistant to race 7 and susceptible to
races 448 and 73, suggesting that they carry only the Co-7 gene. Inoculation with
additional C. lindemuthianum races revealed that line MSU7-3 carries partial resistance
to races 31 and 47. Based on these results it is clear that Co-7 does not contribute
significantly to the resistance spectrum of G 2333, although epistatic interactions were
not investigated in this study. The main value of the C0-7 gene for breeding for
anthracnose resistance will be in the large-seeded Andean varieties of dry bean which are
generally very susceptible to race 7 of C. lindemuthianum. An amplified fragment length
polymorphism (AFLP) marker was developed, which mapped 4.9 cM from the Co-7

locus, to facilitate the introgression of the Co-7 gene into susceptible bean cultivars.

60

INTRODUCTION

Anthracnose caused by the fungal pathogen Colletotrichum lindemuthianum, is a
serious seed-bome disease of dry beans. Worldwide, yield losses due to anthracnose are
particularly severe in subsistence agricultural systems where beans are a valuable source
of plant protein, and income to the producer. Genetic resistance is the most cost effective
means to control the disease. Ten major anthracnose resistance genes (Co-I — C0-10)
have been characterized in common bean and are available for use in breeding programs
(Kelly and Vallejo, 2004). Achieving durable anthracnose resistance poses a challenge to
bean breeders. Due to the high degree of pathogen variability and the continual
emergence of new races, single gene deployment is not an effective strategy to control
bean anthracnose. The pyramiding of resistance genes which have complementary
spectra of resistance has been suggested as a strategy to circumvent the problem of
pathogen variability (Young and Kelly, 1996a; Young and Kelly, 1996b).

The most resistant Middle American cultivar in the anthracnose differential series
is G 2333. G 2333 is the landrace cultivar, Colorado de Teopisca from Chiapas, Mexico
that possesses a naturally occurring gene pyramid for anthracnose resistance. The first
report on the nature of resistance in G 2333 suggested that two independent dominant
genes were present. A 15:1 (resistantzsusceptible) ratio was observed after inoculation of
an F2 population of G 2333/Pijao with race 521 (Pastor-Corrales et al., 1994). Young and
Kelly (1996a) identified the first gene in G 2333 as Co-5 using breeding line SEL 1360
derived from G 2333 (Young and Kelly, 1998). They inferred that since race 521, used

by Pastor-Corrales et al. (1994), overcomes the C0-5 gene G 2333 must possess three

61

independent dominant resistance genes, one of which is C0-5. The second gene in G
2333 was discovered to be allelic to the C0-4 gene in differential cultivar TO and
designated Co-42 (Young et al., 1998). The breeding line SEL 1308, derived from G
2333, carries only the C0-42 gene from the original gene pyramid. The third anthracnose
resistance gene from G 2333 was designated Co-7 (Young et al., 1998), however, no
further characterization was performed and little is known about the spectrum of
resistance which it confers.

The availability of lines SEL 1308 and SEL 1360 facilitated the identification of
the Co—42 and Co-5 genes respectively in G 2333. These lines permited the
characterization of the anthracnose resistance conditioned by the genes and provided
breeders with information on which to base their gene combinations and deployments
strategies. Additionally, these lines were used to identify markers linked to their
respective anthracnose resistance genes which serve as tools to expedite the introgression
process (Young and Kelly, 1998). To begin to isolate the Co-7 gene from the other
anthracnose resistance genes in the pyramid, Young et al. (1998) used a combination of
marker-assisted selection (MAS) and inoculation with races of C. lindemuthianum to
derive SEL 111, which possessed two anthracnose resistance genes: Co-5 and Co-7. The
goals of the current study were to (1) derive a bean breeding line which only carries the
Co-7 gene for anthracnose resistance using a combination of inoculation with C.
lindemuthianum races and MAS, (2) characterize the resistance spectrum conditioned by
the Co-7 gene and (3) identify a molecular marker linked to the Co-7 gene using

amplified fragment length polymorphism (AFLP) and bulked-segregant (BSA) analyses.

62

MATERIALS AND METHODS

Plant Material and C. lindemuthianum races. Populations developed in this
study are shown in Table 2.1. To confirm that the original breeding line SEL 111 carries
only two major genes for anthracnose resistance, a segregating F2 population was derived
from the cross between the cultivar Black Magic (BM) and SEL 111. The black bean
BM is generally considered universally susceptible to anthracnose. This F2 population of
72 individuals was inoculated with race 7 which produced an S/R (susceptible/resistant)
reaction in the parents (BM and SEL 111 respectively). A molecular marker linked to the
C0-5 locus was used to select against the presence of the Co-5 gene among the progeny
segregating for the Co-5 and Co-7 loci within the BM/SEL 111 population. A F3 progeny
test was conducted to determine which of the F2 individuals, which showed the absence
of the Co-5 locus based on MAS, were homozygous for Co-7 by inoculation of F3
families of 15 individuals each, with race 7. The derived lines were named MSU7-1
through 6 and were advanced through five generations of selfing. Line MSU7-5 was not
used in this study because of insufficient seed. These F25 MSU7 lines were inoculated
with a series of C. lindemuthianum races (7, 31, 47, 73, 357, 448 and 1545) to determine
the resistance spectrum conferred by the Co-7 gene. Additionally, specific F25 MSU7
lines were crossed with the cultivar Michigan Dark Red Kidney (MDRK) and the
resulting F2 population was inoculated with race 7 which yields an S/R reaction in the
parents (MDRK and MSU7 respectively) to determine the number of genes segregating

for anthracnose resistance.

63

Inoculum preparation and inoculation procedures. All races utilitzed in this
study were grown from monosporic cultures maintained on filter papers and stored at -
20° C. C. lindemuthianum races were grown from a filter paper on either potato dextrose
agar (PDA) or Mathur’s agar (dextrose (8 g/L), MgSO4. 7 H2O (2.5 g/L), neopeptone (2.4
g/L), yeast extract (2 g/L), and agar (16 g/L)) at 22°C in the dark. After the onset of
sporulation, spores were re-plated in replicate to generate sufficient inoculum. Spores
were scraped from the media into a solution of water and 0.01% Tween 20. An estimated
spore concentration was determined using a hemacytometer and the concentration was
adjusted to 1.2 x 106 spores/mL. The prepared inoculum was then sprayed onto the
abaxial and adaxial surfaces of the leaves. Inoculated seedlings were incubated for 48
hours at approximately 100% relative humidity. Seedlings were rated for disease 5 days
post inoculation on a scale described by Balardin et al. (1997). All races used in this
study were first inoculated on the anthracnose differential series to confirm their

identities.

MAS analysis and derivation of a SCAR marker. To dissect the Co-5, Co-7
gene pyramid in SEL 111, the BM/SEL 111 segregating population was inoculated with
race 7, and a molecular marker linked to the Co-5 locus was used in a MAS strategy. The
RAPD marker, AB3450, is linked in coupling-phase to Co-5 (5.9 i 1.9 cM) (Young and
Kelly, 1997). This marker was converted into a sequence characterized amplified region
(SCAR) marker, using the protocol described by Melotto et al. (1996), to facilitate its use

in a MAS approach. Primer sequences for SAB3450 are shown in Table 2.2. The thermal

cycler profile used to amplify the SAB3 marker consisted of 34 cycles of 10 s at 94°C, 40

s at 67°C, and 2 min at 72°C; one cycle of 5 min at 72°C.

AFLP and BSA analyses. DNA bulks (Michelmore et al., 1991) were
constructed by pooling equal volumes of fluoremetrically standardized DNA
concentrations from six resistant individuals and six susceptible individuals from the
segregating MDRK/MSU7 population. The DNA bulks will be referred to as R and S
bulks, for resistant and susceptible respectively. The AFLP analysis was conducted as
described Hazen et al. (2002). Briefly, genomic DNA from parents and bulks was
digested with EcoRI and MseI. Adapters which serve as subsequent priming sites were
ligated to the sticky-ends followed by two PCR amplifications, pre-amlification was done
with one selective base on each primer and selective amplification was done using
primers with three selective nucleotides. Final PCR products were separated on 6%

polyacrylamide gels and visualized with silver staining (Promega, Madison, WI).

Linkage analysis. Candidate AFLP markers were tested for linkage to resistance
on the MDRK/MSU7 segregating population using MAPMAKER 3.0 (Lander et al.
1987). Linkage distance in cM was calculated using Kosambi’s mapping function with a

LOD threshold value of 2.0.
RESULTS AND DISCUSSION

Derivation and genetic characterization of MSU7 lines. To confirm that SEL

111 carried two major genes for resistance to anthracnose, SEL 111 was crossed with BM

65

and the resulting F2 population was inoculated with race 7, which yields and S/R
(susceptible/resistant) reaction in the parents (BM and SEL 111 respectively). The
segregation of resistant to susceptible individuals fit a 15 :1 ratio (p = 0.57) indicating that
two dominant anthracnose resistance genes were segregating from SEL 111 for resistance
to race 7 (Table 2.2).

To isolate C0-7 from C0-5, the SCAR marker SAB3400 was developed to select
within the segregating population against the C0-5 locus. The derived SCAR marker
amplified a 400 bp fragment that appears as a single polymorphic band between the
parents (data not shown). Young and Kelly (1997) reported a band of 450 bp linked to
the C0-5 locus, however, I consistently amplified a band 400 bp in size with the RAPD
marker, OAB3450 and the derived SCAR marker, SAB3400.

Using SAB3400 to indirectly select against the Co-5 locus, all of the individuals
from the BM/SEL 111 F2 population which were resistant to race 7 were screened. Those
individuals which did not carry SAB3400 were used for further study. F3 families of these
individuals were subjected to a progeny test in which 15 F23 seedlings were inoculated
with race 7 to determine which F2 individuals were homozygous for Co-7. Those F2
individuals, whose F3 family did not segregate for resistance when inoculated with race7,
were self-pollinated for two additional generations and designated breeding lines MSU7-
1 through 6.

To confirm the existence of a single dominant gene conferring resistance to race
7, lines MSU7-3 and MSU7-4 were used for crossing with MDRK. All F2 seeds from
these crosses were planted as a single population (referred to collectively as

MDRK/MSU7), however, pedigrees were recorded to facilitate the removal of progeny

66

which resulted from self-pollination. The F2 population from the MDRK/MSU7 crosses
was inoculated with race 7. The segregation of resistant to susceptible individuals best
fits a 3:1 ratio (p = 0.82) indicating that only one major dominant gene is conferring

resistance to race 7 in MSU7-3 and MSU7-4 (Table 2.2).

Resistance conditioned by the Co-7 gene. Initial characterization of the MSU7
lines was done by inoculation with C. lindemuthianum races: 7, 73 and 448. The disease
reaction profile of the MSU7 lines was compared with SEL 111 and SEL 1360 (Table
2.3). Inoculation of the MSU7 lines with these three races revealed that MSU 7-1 had the
same disease reaction as SEL 111 and SEL 1360. Lines MSU 7-3, 7-4, and 7-6,
however, were susceptible to races 73 and 448. These results suggest that MSU 7-1 is
more resistant than lines MSU7-3, MSU7-4, and MSU7-6. The fact that the resistance
spectrum of MSU 7-1 is identical to that of SEL 111 (Co—5 and C0-7) and SEL 1360 (C0-
5) suggests that MSU 7-1 carries Co-5 or both Co-5 and C0-7 and is a result of a
recombination event between SAB3400 and the Co-5 locus. MSU7-2 showed segregation
within the line when inoculated indicating possible seed contamination (data not shown),
therefore, this line was removed from this study. Lines MSU7-3, MSU7-4, and MSU7-6
were resistant to race 7 and susceptible to races 73 and 448, indicating that they carry
only C0-7. Support for this finding comes from a study by Alzate—Marin et al. (1998)
who reported a 15:1 segregation ratio of resistant to susceptible individuals in a F2
population from the cross G 2333/Ruda inoculated with race 73. Since Ruda does not
carry major genes for anthracnose resistance, and G 2333 is known to carry three, one of

the genes must have been overcome by race 73. Neither SEL 1308 (Co-42) or SEL 1360

67

(Co-5) confer susceptibility to race 73 (Kelly and Vallejo, 2004), therefore, Co-7 must
then condition susceptibility to race 73.

Further characterization of the anthracnose resistance conditioned by the Co-7
gene was conducted by inoculating MSU7-3, MSU7-4 and MSU7-6 with additional C.
lindemuthianum races: 31, 47, 357 and 1545 (Table 2.3). All MSU7 lines inoculated with
race 1545 were susceptible. Race 357 overcame resistance in all three MSU7 lines tested
but not the resistance in SEL 111 and SEL 1360, indicating that the resistance in SEL 111
to race 357 is conferred by the C0-5 gene and not Co-7. Inoculation with races 31 and 47 '
showed that MSU7-3 possessed partial resistance when compared to MSU7-4 and
MSU7-6 which were susceptible. The author defines partial resistance as the appearance
of very small constricted lesions that do not lead to plant death. This observed difference
between the disease reactions of the MSU7 lines to the same race of C. lindemuthianum

can be explained by differences in the genetic background among the lines

Molecular marker analysis. A total of 147 AFLP primer pair combinations
were tested on the parents and R and S bulks from the MDRK/MSU7 F2 population. Five
candidate markers linked to C0-7 were identified. Candidate markers were defined by
the same polymorphism being present in the parents and contrasting DNA bulks. Only
one primer pair, EAchcor-290, however, was linked to the Co-7 locus (Figure 2.1).
Linkage analysis revealed that the AFLP marker, EAchcor-240. was linked in repulsion
phase to the Co-7 locus at 4.9 cM. The Co-5 and Co-7 loci could not be mapped to the
bean integrated map (Freyre et al., 1998) due to lack of polymorphism between the

parents of the mapping population: BAT93, Jalo EEP558 (data not shown).

68

CONCLUSIONS

The C0-7 gene conditions resistance to race 7, but susceptibility to all other races
tested. This indicates that the C0-7 gene does not contribute significantly to the
resistance spectrum of G 2333, although epistatic interactions have not been tested. Race
7 is a widespread race of C. lindemuthianum and a particular problem in large-seeded
Andean varieties of dry bean. I suggest that C0-7 be used in combination with other
resistance genes with complimentary spectra for anthracnose resistance breeding of the
large-seeded Andean bean varieties such as kidney and cranberry beans.

In this study, the dissection of a gene pyramid was accomplished through a MAS
and inoculation strategy. A caveat of working with molecular markers for the dissection
of gene pyramids is the potential for selecting recombinant individuals, that is,
individuals in which the molecular marker and the resistance allele have been uncoupled
as a result of a recombination event. In the present study, I identified line MSU7-1 which
has the same disease reaction as SEL 111 and SEL 1360 indicating that it still carries the
Co-5 gene and possibly the Co-7 gene, but which does not carry the marker, SAB34oo
linked to the Co-5 locus. Unfortunately, I have not yet tested MSU7-1 for the presence of
the marker, EAchcm-240, to test for the Co-7 gene. MSU7-2 appears to be more
resistant than MSU7-3 and MSU7-6, conferring partial resistance to races 31 and 47, both
races of Andean origin. I speculate that the partial resistance is due to background
genetic effects between the different MSU7 lines. I propose that lines MSU7-3 and
MSU7-6 are homozygous for Co-7 based on molecular marker data and inoculation with

different races of C. lindemuthianum. The availability of the MSU7 lines permitted the

69

characterization of the anthracnose resistance conditioned by the Co-7 gene and the
identification of a marker linked to this locus. This marker and the MSU7 lines provide
breeders tools to expedite the introgression of the Co-7 gene into a complementary gene
pyramid. Additionally, since the Co-7 gene has never been fully tested for independence
of all known anthracnose resistance loci, the availability of the MSU7 lines should

facilitate the full characterization of the Co-7 locus.

70

Table 2.1 Primer sequences for the SCAR marker SAB3400 linked to the C0-5 locus.

 

Forward primer sequence“ 5'-TGGCGCACACATAAG'ITCTCACGG-3'

Reverse primer sequence“ 5'-TGGCGCACACCATCAAAAAAGGTT-3'

 

“ The underlined sequences are the original RAPD primer.

71

Table 2.2 F2 Populations inoculated with C. lindemuthianum race 7.

 

 

Population Parental Population Observed Expected P-value
Reaction Size ratio ratio

BM/SEL 111 S/R 155 147:8 15:1 0.57

MDRK/MSU7“ S/R 55b 42:13 3:1 0.82

 

“ Lines MSU7-3, MSU7-4 were used to develop this population

“ Population size is small because crosses with MSU7-2 had to be removed from

population due to evidence of seed contamination.

72

Table 2.3 Disease reaction of MSU7 lines derived from G 2333 and SEL 111 and SEL

1360 inoculated with various races of C. lindemuthianum.

C. lindemuthianum races

 

 

Breeding line R-gene(s) 7 73 448 31 47 357 1545
MSU7-1 C0-5, C0-7 R R R NT“ NT“ NT“ NT“
MSU7-2 NAb R R 8 NT“ NT“ NT“ NT“
MSU7-3 C0-7 R S S Rc RC S S
MSU7-4 Co-7 R S S S S S S
MSU7-6 Co-7 R S S S S S S
SEL 111 C0-5, C0-7 R R R R R R S
SEL 1360 C0-5 R R R R R R S

 

“ NT: not tested

b NA: information not available

C R: partial resistance

73

123456789101112131415

Resistant Bulk Susceptible Bulk

r

EACCMCG'T‘24O

 

Figure 2.1 Partial gel of amplification products produced by AFLP primer pair
EACCMCGT tested on parents and individuals of resistant and susceptible bulks. Lanes (1)
MSU7-3, (2) MDRK, (3-8) individuals from resistant bulk, (9) 100 bp ladder, (10-15)
individuals from susceptible bulk. Upper arrow indicates the lower band as the marker

linked to C0-7. Lower arrow indicates the 200 bp fragment of the 100 bp ladder.

74

LITERATURE CITED

Alzate-Marin, A.L., Menarim, H., de Arruda, M.C., de Carvalho, G.A., de Paula, Junior,
T. J., de Barros, E. G. and Moreira, M. A. 1998. Inheritance of anthracnose
resistance in common bean differential cultivars TO, G 2333, and AB 136. Annu.
Rept. Bean Improv. Coop. 41: 167-168.

Balardin, R.S., J arosz, A., and Kelly, J .D. 1997. Virulence and molecular diversity in
Colletotrichum lindemuthianum from South, Central and North America.
Phytopathology 87:1184-1191.

Freyre, R., Skroch, P.W., Geffroy, V., Adam-Blondon, A.F., Shirrnohamadali, A.,
Johnson, W.C., Llaca, V., Nodari, R.O., Pereira, P.A., Tsai, S.M., Tohme, J., Dron,
M., Nienhuis, J ., Vallejos, CE, and Gepts, P. 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.

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

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

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

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

Michelmore, R.W., Paran, 1., and Kesseli, R.V. 1991. Identification of markers linked to
disease-resistance genes by bulked segregant analysis: a rapid method to detect

markers in specific genomic regions by using segregating populations. Proc. Natl.
Acad. Sci. 88:9828-9832.

Pastor-Corrales, M.A., Erazo, O.A., Estrada, EL, and Singh, SP. 1994. Inheritance of
anthracnose resistance in common bean accession G 2333. Plant Dis. 78:959-962.

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

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

75

Young, R.A., and Kelly, J.D. 1996b. RAPD markers flanking the Are gene for
anthracnose resistance in common bean. HortScience 121 :37—41.

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

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

76

CHAPTER THREE

ANALYSIS OF CONSTITUTIVELY EXPRESSED AND INDUCED GENES IN
THE RESPONSE OF PHASEOLUS VULGARIS TO COLLETOTRICHUM
LINDEMUTHIANUM

ABSTRACT

The induction of the defense response in plants triggers a rapid multicomponent
series of events. Initial response of the plant can occur within minutes and involve rapid
gene activation. The response of common bean, Phaseolus vulgaris, to an incompatible
race of Colletotrichum lindemuthianum was used in this study. C. lindemuthianum is the
fungal pathogen which causes bean anthracnose, a serious seed-bome disease of dry bean
worldwide. The objectives of this study were to use cDNA-AFLP to determine if any
constitutively expressed differences exist between plants which carry the C0-4“ gene that
may contribute to the resistance response, and (2) identify genes which are differentially
regulated during the early stages of the incompatible interaction between C.
lindemuthianum and P. vulgaris. Several constitutively expressed genes involved in
primary metabolism were discovered which have a possible role in the ability of the bean
plant to mount Co-4“ mediated resistance. Additionally, several genes were identified
which are differentially regulated between inoculated and mock-inoculated treated plants
in the incompatible interaction. Changes in gene expression were detected as early as 1
hour post inoculation. Among the genes identified, was a putative serine/threonine
receptor kinase, a putative lysine decarboxylase-like enzyme and a NIMl-like protein.
The products of these genes have previously been correlated with the defense response in

a number of other taxa. I speculate that the combination of constitutive expression

77

differences in primary metabolism processes and genes involved in active defense

contribute to anthracnose resistance mediated by the Co-4“ gene in common bean.

78

INTRODUCTION

Resistance to pathogens in plants is accompanied by the rapid deployment of a
multicomponent defense response. Individual components of this response include the
hypersensitive response (HR), chemicals with antimicrobial properties such as
phytoalexins and hydrolytic enzymes and structural changes such as the lignification and
strengthening of cell walls (Dixon and Harrison, 1994). Initial response of pathogen-
challenged or elicitor-treated plant cells can occur within rrrinutes and involve rapid gene
activation. The intracellular signal transduction events during the initial stages of the
resistance response include rapid and transient changes in ion fluxes across the plasma
membrane, production of oxidative bursts and changes in the phosphorylation status of
various proteins (Kombrink and Schmelzer, 2001). Thus, extensive reprogramming of
both primary and secondary metabolism occurs at the level of gene expression (Schenk et
al., 2000) and many of the induced proteins have a direct or indirect inhibitory effect on
the pathogen (Hammond-Kosack and Jones, 1996).

The incompatible interaction between C. lindemuthianum, the fungal pathogen
which causes anthracnose on dry bean (Phaseolus vulgaris), has been previously studied
at the level of cytology and gene expression of key enzymes involved in the defense
response (Ryder et al., 1984; Edwards et al., 1985; Crarner et al., 1984; Bell et al., 1986;
Hedrick et al., 1988; Mahé et al., 1992; Franc-Velazquez and Lozoya-Gloria, 2003).
These studies have primarily focused on the induction of the L—phenylalanine ammonia-
lyase (PAL) and chalcone synthase (CHS) genes either in cell suspension or whole plants
in response to fungal inoculation, fungal-derived elicitors or chemical elicitors. PAL and

CH8 are key enzymes in the production of phytoalexins in P. vulgaris (Dixon et al.,

79

1983). From these studies it is apparent that changes in P. vulgaris gene expression are
initiated prior to condial germination, however, very few studies have focused on the
very early changes in gene expression at the whole plant level. Fraire-Velazquez and
Lozoya-Gloria (2003), studying the incompatible interaction between C. lindemuthianum
race 1472 and the cultivar Michigan Dark Red Kidney (MDRK), found that all conidia
germinated 12 hours post-inoculation (hpi) and that PAL transcripts accumulated strongly
within 2hpi. Additionally, they identified three genes which they suggested are involved
in the establishment of the incompatible interaction within 6hpi: beta-glucosidase, Seven
i_n Absentia (sina) and SUMO (Small Ubiquitin related MQdifer).

Genome-wide expression analysis is a valuable tool for determining the functions
of genes and their spatial and temporal expression patterns, as well as elucidating the
genetic networks in which they participate (Schenk et al., 2000). Microarray technology
has been implemented to address many questions involving the interaction of plants with
various biotic and abiotic stresses. This technology is nevertheless restricted to a small
number of organisms for which representative sequence information is available. The
cDNA-amplified fragment length polymorphism (cDNA-AFLP) technique provides an
alternative approach to microarray analysis and is widely available at a lower cost and
does not require prior sequence knowledge (Bachem et al., 1996). The cDNA-AFLP
technique is a PCR-based technology similar to differential display, but with higher
reproducibility. Specifically, the most critical advantage of cDNA-AFLP over
differential display is that cDNA-AFLP is based on linker-ligated PCR whereas

differential display is based on PCR which is arbitrarily primed (Bachem et al., 1996).

80

Anthracnose is a serious seed-bome disease of dry beans. Worldwide, yield
losses due to anthracnose are particularly severe in subsistence agricultural systems
where beans are a valuable source of plant protein, and income to the producer. Genetic
resistance is the most cost effective means to control the disease. Ten major anthracnose
resistance genes (Co-1 — Co-IO) have been characterized in common bean and are
available for use in breeding programs (Kelly and Vallejo, 2004) although none have
been cloned. The recently identified allele at the Co-4 locus, Co-4“, conditions
resistance to 97% of anthracnose races present in North and South America (Balardin and
Kelly, 1998; Young et al., 1998) and provides the broadest-based resistance of all of the
alleles at the C0-4 locus (Kelly and Vallejo, 2004). The objectives of this study were to
use cDNA-AFLP to (1) determine if any constitutively expressed differences exist
between plants which carry the Co-4“ gene that may contribute to the resistance response,
and (2) identify genes which are differentially regulated during the early stages of the

incompatible interaction between C. lindemuthianum and P. vulgaris.

MATERIALS AND METHODS

Plant material and growth conditions. To analyze constitutive differences
potentially associated with C0-4“ mediated resistance, two genotypes which carry Co-42
were compared to two which do not carry this gene. These genotypes are referred to as
+Co-4“ and —C0-4“ respectively. To minimize differences due to genetic background the
plant material used was highly related. The two genotypes used that carried Co-4“ were

the breeding line SEL 1308 and a BC4F2 progeny, designated A51-2, from the cross

81

Jaguar*4/SEL 1308. The two genotypes which do not carry the Co-4“ gene were the
black bean cultivar Jaguar and the breeding line SEL 1360. Since Jaguar is the recurrent
parent used to develop A51-2, this line is a near-isogenic line from Jaguar. Also, SEL
1360 and SEL 1308 are sister-lines derived from the same backcross with G 2333.
Although neither Jaguar nor SEL 1360 carry Co-4“, they both are known to carry other
anthracnose resistance genes. A51-2 was selected among BC4F2 progeny based on the
presence of makers linked to the C04 locus: SASl3 (Young et al., 1998) and SBB14
(Awale and Kelly, 2001). SBB14 is a co-dominant marker and was used to select an
individual which is not heterozygous at the C0-4 locus. The selected BC4F2 progeny was
designated A51-2. Seeds were planted in a growth chamber and provided 12 hours of
light. The chamber temperatures were: daytime at 22° C, nighttime at 20°C. Tissue for
RNA extraction was collected when the plants had fully expanded primary leaves.
Tissues from 6 plants of a single genotype were bulked to eliminate possible single plant
effects.

To analyze the pattern of gene expression in the incompatible interaction during
the early stages of plant/pathogen interaction, the genotype SEL 1308 was used. Plants
were grown in the same conditions as previously described. Tissue was collected at

various time points and bulked as described above.

Inoculum preparation and inoculation. Two treatments were used: inoculated
(I) and mock-inoculated (MI). In the I treatment, C. lindemuthianum race 7 was used.
Mock-inoculation was performed using water and 0.01% Tween 20. The fungal isolate

were grown from a monosporic culture, maintained on filter papers and stored at -20° C.

82

Inoculum was prepared by growing the fungus from a filter paper on potato dextrose agar
(PDA) at 22°C in the dark. When the cultures began to sporulate, spores were re-plated
in replicate to generate sufficient inoculum. Spores were suspended in a solution of
water and 0.01% Tween 20. Spore concentration was estimated using a hemacytometer
and the concentration was adjusted to 1.2 x 106 spores/mL. The prepared inoculum was
then sprayed onto the abaxial and adaxial surfaces of the leaves. Inoculated seedlings
were incubated at approximately 100% relative humidity for 48 hours. Positive and
negative control plants were rated for disease 5 days post inoculation on a scale described
by (Balardin et al., 1997). Race classification of the isolate was confirmed prior to use by

inoculation on the anthracnose differential series to confirm its identity.

cDNA-AFLP analysis. Total RNA was extracted from bulked tissue samples
from time points 0, 0.5, 1, 2, 4, 6, 8, 12, 16, and 20 hpi for both I and MI treated plants.
RNA was extracted using the RNeasy® Plant Mini Kit (Qiagen, Valencia, California).
RNA quality and integrity was assessed using electrophoresis and spectrometery. 250 ug
of total RNA for each genotype was used for mRNA isolation via the PolyATractTM
mRNA System (Promega, Madison, WI). Resulting mRNA samples were speed-
vacuumed and resuspended in 10uL of nuclease-free water. Double-stranded cDNA was
synthesized from the mRNA samples using the Universal Riboclone cDNA Sysnthesis
System (Promega, Madison, WI). cDNA-AFLP analysis was performed as described by
Hazen et al. (2002) with the following modifications. No selective bases were used in the
pre-selective amplification primers and two selective bases were used in the selective

amplification primers. Pre-selective amplification thermal cycler profile used was 2 min

83

at 94° C; 23 cycles of 1 min at 94° C, 1 min at 56°C, 1 min at 72° C; 5 min at 72° C; 4°
soak. The selective amplification thermal cycler profile was: 2 min at 94° C; 12 cycles
of 30 s at 94° C, 30 s at 62° C (-O.5°/cycle), l min 30 s at 72° C; 29 cycles of 30 s at 94°
C, 30 s at 56° C, 45 s at 72° C; 5 min 72° C; 4° C soak. Final PCR products were
separated on 6% polyacrylamide gels and visualized with silver staining (Promega,

Madison, WI).

Transcript-derived fragment (TDF) cloning and sequence analysis. cDNA-
AFLP transcript profiles were analyzed for differences in gene expression. Only
fragments larger than 100 bp were considered. In the analysis of constitutive expression
differences I focused on bands (referred to as TDFs) which were present in genotypes
which have Co-42 mediated resistance and absent in those that do not and visa versa. In
the analysis of expression differences between I and MI treated plants, I focused on TDFs
that were present or absent in two or more consecutive time points. TDFs of interest
were cut from fresh gels using a sterile scalpel. The gel slice was then placed in a
microcentrifuge tube with 50 111 of water and incubated overnight at 4°C to diffuse the
DNA template. Selective PCR was done with the original selective primer combination
using 5 ul of thetemplate DNA to re-amplify the fragment. PCR products were then
separated on a 2.5 % agarose gel made with 1X TAE and an ethidium bromide
concentration of 0.2 mg/mL. TDFs which re-amplified a single product were used for
cloning using the Qiagen PCR Cloning Kit (Valencia, California). The cloning reaction
was used for the transformation of chemically competent DH501 strain of E. coli. The

transformation was canied out as suggested in the Qiagen® PCR Cloning Handbook

84

(April 2001). Sequencing was primed from the T7 or SP6 vector sites. Multiple clones
for each TDF were sequenced, trimmed of vector and AFLP adaptor sequences, and
aligned using ClustalW (Thompson et al., 1994) to identify a consensus sequence among
the clones. Database searches were performed using the BLAST Network Service
(NCBI, National Center for Biotechnology Service). Each TDF was compared against all
sequences in the non-redundant databases using the BLASTX program, which compares
translated nucleotide sequences with protein sequences. TDFs which returned no
significant homology were compared again against the EST databases using the
BLASTN program. The BLASTP program was used on the six-frame translated amino

acid sequence to search for putative protein domains.

RESULTS AND DISCUSSION

Constitutive expression cDNA-AFLP analysis. To gain a better understanding
of Co-4“ mediated anthracnose resistance, I first explored constitutive differences
between genotypes which possessed or do not possess the C0-4“ gene. A total of 149
selective amplification primer pair combinations were tested on the four genotypes:
SEL1308, R51-2, SEL1360 and Jaguar. Only 22 TDFs were identified which were
clearly polymorphic between the genotypes either carrying or not carrying Co-4“. I was
only able to obtain sequence information from 14 of the 22 TDFs due to problems with
band isolation and purification. It should be noted that multiple sequences of the same
size were identified from a single excised band. In these cases, the sequences were
analyzed separately. The multiple sequences could represent different transcripts which

when restricted with EcoRI and MseI, produce fragments of the same size. These

85

fragments would co-nrigrate in the gel and be re-amplified, post excision, by the same
primer combination. Sequence similarity results (Table 3.1) were obtained for nine of the
14 TDFs, although, only six have assigned functions; the remaining TDF sequences were
too small to reveal significant homology. The isolated TDFs represent constitutive
differences in the expression of specific genes between two genotypes which have Co-42
mediated resistance and those that do not. Therefore, these differentially expressed genes
may provide some insight into innate differences which exist between resistant and
susceptible plants. Patterns of differential expression are summarized in Table 3.1.
TDFs C2, C5 and C63 were isolated from genotypes which carry the Co-4“ gene. Based
on homology analyses, TDFs C2 and C5, have putative functions in carbohydrate
metabolism. TDF C2 is homologous to the Malate dehydrogenase enzyme from Brassica
napus. Malate metabolism has been implicated as participating in plant defense based on
fluctuations in the activities of malate-transforrrring enzymes (eg. Malate dehydrogenase)
and the cellular level of malate upon biotic or abiotic stress. Malate oxidation is carried
out by various isoforrns of malate dehydrogenase to release reducing equivalents
(NADH) and CO2 (Schaaf et al., 1995). Gross et al. (1977) speculates that a wall-bound
malate dehydrogenase may play a role in reactive oxygen species (ROS) production by
providing NADH, which is supplied to the cell via a malate oxaloacetate shuttle across
the plasma membrane. In lettuce, peroxidase with NAD(P)H oxidase activity are the
major source of ROS production (Bestwick et al., 1998). Reactive oxygen species result
from the serial reduction of dioxygen. The rapid generation of oxidants has now been
described in many plant/pathogen interactions and is a characteristic feature of HR (Allen

etaL,l999)

86

Although cDNA-AFLP is a useful technique for the large-scale analysis of gene
expression in plants, this technique unlike microarray analysis, does not address the
relative or absolute magnitude of transcript abundance. The cDNA-AFLP technique is
useful for identifying the changes or differences of transcript abundance. Therefore, the
apparent presence and absence of TDF C2 in the +Co-4“ and —C0-4“ genotypes
respectively, indicates that those genotypes which carry the Co-42 gene have more of the
corresponding transcript, which putatively encodes a malate dehydrogenase enzyme, than
those genotypes tested which do not posses Co-4“. Since these plants are capable of
mounting Co-4“ mediated resistance, and they potentially have an increased level of
malate dehydrogenase, which is involved in the production of ROS, I can speculate that
the elevated level of this enzyme may provide a means for these plants to more quickly
produce ROS during an incompatible interaction.

TDF C5 is homologous to the enzyme ADP-glucose pyrophosphorylase (AGPase)
from P. vulgaris. The key regulatory step in starch biosyntheses is catalyzed by AGPase.
Particularly, in leaf and storage tissues in plants, AGPase functions to catalyze the
synthesis of ADP-glucose from glucose-l-phosphate and ATP (Muller-Rober et al.,
1990). TDF C5 had the same expression pattern as TDF C2 (Table 3.1). Among many
metabolic processes, availability of glucose in the cell is important for glycolysis, the
process whereby energy which is stored in carbohydrates is released in the form of ATP
and NADH. The first step in glycolysis is the breakdown of glucose, a six-carbon sugar,
into two three-carbon sugars which are subsequently oxidized and re-arranged to yield
two molecules of pyruvate. The apparent increase in ADP-glucose pyrophosphorylase

transcript in the +Co-42 plants when compared to —Co-4“ plants, suggests that an increase

87

in available ADP-glucose and probable higher levels of ATP and NADH, may confer an
advantage associated with C0-4“ mediated resistance. TDF C63 is homologous to a
putative alpha-mannosidase enzyme from Zea mays. The probable function of alpha-
mannosidase in plants is in the turnover of glycoproteins and in-the processing of
oligosaccharide derivatives, which are implicated in the synthesis of diverse
glycoproteins (Cassab and Varner, 1988). Additionally, Boller (1983) suggests that
alpha-mannosidase may attack specific wall structures of pathogens. The presence of
higher levels of alpha-mannosidase expression in the +C0-4“ genotypes indicates that
there may be a role for this enzyme in defense prior to pathogen attack.

TDFs C7, C17 and C18 represent genes which have higher expression in the
genotypes which lack the Co-42 gene. TDF C7 is homologous to the plastidic 2-
oxoglutarate/malate (malate/0AA) transporter from Zea mays. Plastids perform most of
the plants biosynthetic reactions. Plastid transporters play an important role in the
regulation of metabolism (Dennis and Blakeley, 2000). Additionally, malate is shuttled
into the mitochondria via a malate/0AA transporter. Once inside the matrix, malate can
be oxidized by two enzymes; malate dehydrogenase, to create reducing equivalents
(NADH), or malic enzyme, which generates pyruvates, CO2 and NADH. It is interesting
to note that the expression of the malate dehyrogenase gene (TDF C2) is less in the —Co-
4“ genotypes than in the +C0-4“ genotypes, thus, if the -Co-42 genotypes have increased
malate/0AA transporter expression, it may be likely that malate is oxidized more often
by malic enzyme than by malate dehydrogenase in the —Co-4“ plants. Since malate is
transported into many different plastids for various processes, it is difficult to speculate

on the significance of higher levels of expression in the —Co-4“ plants, however, this data

88

may support the importance of metabolic regulation in the ability of a plant to mount the
defense response. TDF C17 is homologous to a putative lipoate-protein ligase B which
catalyses the formation of an amide linkage between lipoic acid and a specific lysine
residue in lipoate dependent enzymes. Lipoylated proteins use this prosthetic group as a
carrier of intermediates and reducing equivalents during enzymatic reactions. The major
lipoylated proteins are highly conserved and sophisticatedly regulated enzymes of central
metabolism (citric acid cycle, T CA), the pyruvate and alpha-ketoglutarate dehydrogenase
complexes. The enzymatic activity of these complexes requires the binding of several
cofactors including lipoic acid, therefore, the lipoate-protein ligase B, which catalyzes the
addition of this cofactor to the complex is important for it’s function. It is not obvious
what the effect of increased lipoate-protein ligase B (TDF C17) expression is on the
activity of these key enzymes complexes of the TCA cycle. However, it is apparent that
there are defined differences in gene expression between +C0-4“ and -—Co-4“ plants at the
level of central metabolism. TDF C18 is homologous to translation elongation factor-l
(EF-l) from Phaseolus coccineus. Elongation factor-1 is a highly conserved and
ubiquitous protein translation factor, therefore, I am unable to speculate on its possible
role in defense. However, it should be noted that an elongation factor gene has been
mapped to linkage group B7 in P. vulgaris. This elongation factor gene is located
between the phenylalanine ammonia-lyase-2 gene and the anthracnose resistance gene
Co-6 (Gepts, 1999). There is no evidence, however, that TDF C18 is the same elongation

factor which is mapped to this cluster of disease response genes.

89

Expression pattern of early events in the incompatible interaction. A total of
65 primer pair combinations were tested on cDNA templates prepared from 10 time
points of I and MI treatments within a 20 sampling period. These time points were
selected to target the earliest changes in gene expression in response to inoculation with
an incompatible pathogen. Based on previous studies using various races and bean
cultivars, germination of spray inoculated C. lindemuthianum conidia occurs within 14hpi
(O’Connell et al., 1985; Fraire-Velazquez and Lozoya-Gloria, 2003). Fraire-Velazquez
and Lozoya-Gloria (2003) observed single plant cell death and severe conidia
degredation at 24hpi of bean plant with an incompatible race. In addition, the fungal
structures appeared similar at 48hpi as those observed at 24hpi in the incompatible
interaction.

Thirty-four TDFs were identified which appeared differentially expressed
between 1 and MI treatments. Differentially expressed TDFs had to be present or absent
in at least two consecutive time points in a treatment to be considered for further analysis.
Twenty-two TDFs were successfully re-amplified, cloned and sequenced. The
expression patterns of the differentially expressed TDFs are summarized on Table 1.2.
Multiple clones per TDF were sequenced and redundancy was eliminated by using
CLUSTALW. Multiple sequences from the same TDF were detected in some cases and
were treated separately for subsequent analyses. Nine of the 22 TDF sequences had
significant similarity with genes of known or putative function (Table 3.2).

TDFs 11.1 and 11.4 represent non-redundant sequences obtained from the same
excised band (Figure 3.1). Upon BLASTX analysis, however, they both have similar

putative kinase functions. TDF 11.1 is homologous to a putative CRKl protein from

90

Oryza sativa. The CRK class of protein is a subclass of the calcium-dependent protein
kinase (CDPK) family; CRK stands for CDPK-related kinase (Furumoto et al., 1996).
CDRKs are calcium-regulated protein kinases. They contain a single polypeptide which
has a catalytic domain for serine/threonine phosphorylation, a junction involved in
autoinhibition, and a C-terrninal calmodulin (CaM)-like domain with four calcium
binding EF hands. CRKs are widely conserved among angiosperms and constitute a
small gene family. They, however, are distinct from typical CDPK by their degenerate
sequence in the Cam-like domain including four calcium-binding EF-hands. Furumoto et
al. (1996) suggest that CRKs actually make up a distinct subfamily of calcium-dependent
serine/threonine kinases which may not actually require calcium for their activity. The
PlantsP database (http://plantsPsdsc.edu) reports that most CRKs contain transmembrane
domains. In addition, CRKs have a myristoylation motif, necessary for targeting to the
plasma membrane (Day et al., 2002). No specific function has been assigned to CRKs in
plant cells, however, the expression pattern of TDF 11.1 (Figure 3.1), indicates that this
gene is more highly expressed in the I treatment when compared to the MI treatment.
This indicates that gene is upregulated in response to the pathogen, starting at lhpi.

TDF 11.4, isolated from the same band as TDF 11.1, also shows homology to a
protein with putative serine/threonine kinase activity from Oryza sativa. However, this
protein is a receptor serine/threonine kinase (classified as a receptor-like kinase; RLK).
Some examples of the role of RLKs in defense are the Xa21 gene in rice and the Pto gene
of tomato. The Xa21 gene, which conditions resistance to bacterial leaf light
(Xanthomonas oryzae pv. oryzae), encodes a membrane-bound RLK with an extracellular

leucine-rich repeat (LRR). Although no ligand is known which binds with this gene

91

 

product, the presence of the extracellular LRR suggests a receptor role in pathogen
recognition (Hardie, 1999). The Pto gene of tomato, which confers strain-specific
resistance to Pseudomonas syringae, encodes a serine/threonine protein kinase with N-
terminal myrostoylation sites, but has no extracellular or transmembrane domains. To
condition resistance, Pto requires the presence of Prf, which encodes a LRR containing
protein. The Pto gene is classified as a RLK due to the sirrrilarity at the kinase domain
(Hardie, 1999). The upregulation of TDF 11.4 in response to inoculation with C.
lindemuthianum suggest that this receptor serine/threonine kinase may be involved in
perception of the pathogen. The only known resistance gene whose expression is induced
by inoculation is the X01 gene in rice which conditions resistance Xanthomonas oryzae
pv. oryzae but does not contain a kinase domain (Yoshimura et al., 1998).

Since both TDF 11.1 and 11.4 were isolated from the same band, and the
homology analysis revealed that they are similar to two different putative proteins, it is
not possible to conclusively say that the expression pattern observed by cDNA-AFLP
corresponds to both. To confirm that these two TDFs are upregulated in response to
inoculation, northern analysis using each TDF as a separate probe is necessary. Because
the TDF 11.1 and 11.4 sequences are so small, only 156 and 167 bases respectively, and
both TDFs encode putative serine/threonine kinase proteins, it is possible that they are
two pieces of the same gene; perhaps each represents different domains of the same
protein. An additional caveat is the ubiquitous nature of kinases. Kinase proteins are
involved in the regulation of many cellular processes such as cell division, metabolism
and responses to biotic and abiotic stresses and approximately 4% of Arabidopsis genes

encode for typical kinase proteins (Hrabak et al., 2003).

92

TDF 12 encodes for a beta-mannosidase. This enzyme belongs to the family 1
glycosidases. Among this family of enzymes, the beta-glucosidase protein has been
previously reported to be upregulated in response to a host/pathogen interaction in Z.
mays/B. maydis (Simmons et al., 2001) and in P. vulgaris/C. lindemuthianum (Fraire-
Velazquez and Lozoya-Gloria, 2003). In both of these studies the beta-glucosidase gene
was suggested to have a role in the defense response although no biochemical evidence is
presented. However, Hsieh and Graham (2001) suggest that beta-glucosidase is involved
in the release of two isoflavone-conjugated compounds which have a central role in the
defense response in Glycine max. Differential expression of the beta-glucosidase gene
identified by Fraire-Velazquez and Lozoya-Gloria (2003) was detected between 1 and 6
hpi of intact bean plants, spray inoculated with an incompatible race of C.
lindemuthianum. Because they bulked the tissue samples from time points 1h — 6hpi for
the subtractive hybridization procedure, I was unable to determine the first time point of
detection and that of greatest accumulation. Despite this, the expression of TDF 12
detected between time points 6 through 12 hpi does not contradict these finding.
Therefore, I believe that the beta-mannosidase, also a family 1 glycosyl hydrolase,
identified in the current study is under similar regulation as the beta-glucosidase detected
by Fraire-Velazquez and Lozoya-Gloria (2003) and may also have a role in the defense
response. Fraire-Velazquez and Lozoya-Gloria observed that cytological differences
were obvious in pathogen development between the incompatible and compatible
interactions after only 14 hpi. The authors used the bean cultivar Michigan Dark Red
Kidney, known to carry the Co-I locus for anthracnose resistance. The fungal conidia in

the incompatible interaction were severely damaged by 24 hpi and they observed fungal

93

cell debris around the collapsed and aggregated spores. Dead single plant cells were
observed at 24 hpi in the incompatible interaction, indicating HR. Fraire-Velazquez and
Lozoya-Gloria suggest that the beta-glucosidase activity could have contributed to the
severe degredation of the conidia.

The expression pattern of TDFs 18.1, 18.2, 22 and 23 (Figure 3.1) are strikingly
similar in that transcript accumulation increased between time points lb and 6hpi,
decreased between 8h and 12hpi and increased again between 16h and 20hpi. A very
similar temporal accumulation pattern was reported for the L-phenylalanine ammonia-
lyase (PAL) gene in the incompatible interaction between P. vulgaris and C.
lindemuthianum. PAL is a key enzyme in the production of phytoalexins in P. vulgaris
through the phenylpropanoid biosynthetic pathway (Franc—Velazquez and Lozoya-Gloria,
2003). Phytoalexins are low molecular weight, lipophilic, antimicrobial compounds that
accumulate rapidly around sites of the incompatible pathogen infection (Smith, 1996).
This would suggest that the TDFs which I identified, are regulated similarly to PAL, and
therefore, may play a role in the response of P. vulgaris to an incompatible race of C.
lindemuthianum. TDF 12, may also be similarly regulated since its pattern of
accumulation is directly opposite that of TDFs 18.1, 18.2, 22 and 23 (Table 3.2) , and it
too seems to be related to a gene involved in the defense response of P. vulgaris to
incompatible races of C. lindemuthianum.

The expression of TDFs 18.1 and 18.2 in the MI treated plant appears to be
constitutive, however, it should be noted that time point 0h in the MI treated plants is
more intense than that of time point Oh of the I treated plants. The expression of this time

point between the two treatments should be the same, therefore, this observed difference

94

 

may represent an increased amount of RNA used in the initial steps of the cDNA-AFLP
analysis. TDF 18.1 is homologous to a putative lysine decarboxylase-like protein from
Oryza sativa (Table 3.2). BLASTP analysis revealed that TDF 18.1 contains a lysine
decarboxylase domain (Figure 3.2). Lysine decarboxylase is present in some bacteria,
but also several higher plants, particularly Graminae, Legurrrinoseae and Solanaceae.
The function of the lysine decarboxylase enzyme is in the catalysis reaction of L-lysine to
cadaverine and CO2 (Crozier et al., 2000). Cadaverine is a polyarrrine, which is a low
molecular mass polycation that has antiherbivory properties and has also been implicated
in molecular signaling events in plant pathogen interactions (Martin-Tanguy, 2001). The
activity of cadaverine is positively correlated with the accumulation of quinolizidine
alkaloids in the aerial parts of the lupin plant where their defensive properties are most
effective (Wink and Hartmann, 1982). Additionally, lysine decarboxylase activity was
found to be positively correlated with chlorophyll levels (Martin-Tanguy, 2001). The
conjugation of polyamines to phenolics has often been described as a defense mechanism
against infection of several higher plants by viruses and fungi (Fleurence and Negrel,
1989; Louis and Negrel, 1991). Therefore, the lysine decarboxylase enzyme may play a
role in the early response of P. vulgaris to C. lindemuthianum in the incompatible
interaction.

TDF 18.2 encodes a NIMl-like protein 1 from Helianthus annuus. NIMl, stands
for Non-inducible Munityl (Delaney et al., 1995), and is also called NPRl, which stands
for Non-expressor of Pathogenesis-Related genesl (Cao et al., 1994). The transduction
of the salicylic acid (SA) signal to activate the production of pathogenesis-related (PR)

proteins, requires the function of NlMl. It is a regulatory protein that was identified

95

 

through screening systenric-acquired resistance compromised mutants of Arabidopsis
(Delaney, 1997). In addition to local defense responses such as HR, recognition of
avirulent pathogens can trigger a secondary defense response which gives the plant a
long-lasting, broad-spectrum disease resistance throughout the entire plant which is
referred to as systemic-acquired resistance (SAR) and is accompanied by the production
of PR proteins (Ryals et al., 1996). NIMl expression in plants is constitutive, however,
upon induction of SAR, it is translocated to the nucleus to act as a modulator of PR gene
expression but does not bind DNA directly. NIMl enhances the DNA binding of a basic
leucine zipper (bZIP) family of transcription factors, referred to as TGAs, to the SA-
responsive promoter elements in the Arabidopsis PR-I gene. The NIMl gene contains
both a BTB/POZ domain and ankyrin repeats (Cao et al., 1997). BLASTP analysis
revealed that TDF 18.2 contains a BTB domain (Figure 3.2). The BTB/POZ domain is
an evolutionarily conserved region of about 120 arrrino acids, originally named for the
Drosophila Broad-Complex, Iramtrack and Bric-a—brac proteins (BTB) (Zollman et al.,
1994), in which the 29x virus and Zinc finger proteins can be found, (Bardwell and
Treisman, 1994). Ankyrin repeats are tandemly repeated amino acid modules of about 33
residues, and are considered the most common protein interaction motif (Sedgwick and
Smerdon, 1999). The expression pattern of TDF 18.2, a NlMl-like gene, suggests a
possible role for that gene in the regulation of genes involved in the defense response.
TDFs 22 and 23 both encode enzymes involved in the biosynthesis of chlorophyll.
Their pattern of expression, which is the same as that for TDFs 18.1 and 18.2, may at first
appear diumally regulated; however, the constitutive expression of these genes in the MI

treated plants suggests that the change in transcript accumulation is not the result of the

96

20h sampling period of this study. Although there is not an obvious link for these
proteins in the defense response, similar results have been reported in other studies which
suggest crosstalk between the light and defense signal transduction pathways. In the
microarray analysis of the defense response of Arabidopsis, Schenk et al. (2000) found
that SA treatment induces the expression of genes encoding chlorophyll a/b-binding
proteins (CABS). In contrast, they found that the application of methyl jasmonate down-
regulated the gene for ribulose bisphosphate carboxylase. It should be noted that the
RNA samples for this study were taken at 24 hours post chemical treatment. Genoud and
Metraux (1999) showed that the Arabidopsis phytochrome A and B signaling mutant,
psi2, had elevated levels of PR gene expression and suggested crosstalk between the light
signal transduction and PR gene signaling pathways.

TDF 31, which encodes a zinc finger (C3H-type) family protein from
Arabidopsis, has a unique expression pattern among the TDFs reported in this study.
TDF 31 expression appears generally repressed in the I treated plants and expressed in
the MI plants. Plants from both treatments at time point thi have expression of the
gene. This gene is quickly down-regulated in both the I and MI treated plants, however,
this gene remains repressed in the I treated plants, whereas, expression is increased again
in the MI treated plants between 2h and 4hpi. Expression is not detected at time point
6hpi in the MI treated plants, but then increases between 8h and 20hpi. The zinc finger
motif is involved in protein binding to DNA as well as protein-protein interaction. They
are classified into C2H2, C3H, C4 and ring finger types based on the arrrino acid
sequences comprising the zinc finger domain (Garrett and Grisham, 2002). In a

microarray experiment addressing the analysis of gene expression during the defense

97

response in Arabidopsis, Schenk et al., (2000) found that the treatment of plants with SA
or methyl jasmonate down-regulated the expression for a zinc-finger transcription factor.
They also found that inoculation with an incompatible fungus and treatment with
ethylene also down-regulated two zinc-finger proteins. The zinc-finger protein from
Arabidopsis, LSDl (lesion-simulating disease resistance), is a negative regulator of the
hypersensitive response (Dietrich et al., 1997), therefore, the involvement of genes
encoding transcription factors with zinc-finger/binding domains in regulating plant
defense responses appears very likely. The repressed expression of TDF 31 supports a

putative role as a negative regulator of the defense response.

Comparison of constitutive and induced differences in plants which can
undergo R-gene mediated resistance. All of the TDFs which were identified in the
analysis of constitutive differences were neither induced nor repressed in response to
inoculation with an incompatible race of C. lindemuthianum. Despite the fact that they
are not inducible does not signify that they are not important for the defense response to
take place. These differences represent innate differences in basic metabolism which
may provide an advantage or a disadvantage in a plants ability to more quickly trigger the
defense response. The TDFs isolated from the analysis of the incompatible interaction
were not differentially expressed between —C0-42 and +C0-4“ plants in the study of
constitutive expression differences, supporting the idea that these TDFs represent an

active response in the plant to alter gene expression.

98

CONCLUSIONS

The cDNA-AFLP technique enabled the discovery of genes which may be
involved in the defense response at the constitutive and induced expression levels. Eight
genes were identified with putative protein functions previously associated with various
aspects of the defense response. I was particularly interested in identifying genes which
play a critical role in the early stages of the plant/pathogen interaction which may be
determinants for the final response of the plant. In the study of the incompatible
interaction, changes were observed in gene expression as early as lhpi suggesting that
these changes occur possibly earlier than pathogen recognition, assumed to occur after
penetration of the fungus into the plant cell. The genes identified may represent genes
which are regulated by the interaction or possible regulators of downstream genes as in
the case with the zinc-finger protein. However, the expression pattern of the identified
genes must be confirmed by either Northern analysis or by real-time PCR to validate
these results. Determining the genomic location of these genes on the bean integrated
linkage map (Freyre et al., 1998), and their relative proximity to known anthracnose
resistance loci, may provide linked molecular markers for use in the marker-assisted
breeding of anthracnose resistant bean cultivars. The demonstration that specific changes
in gene expression occur at an early stage in the plant defense mechanism, may lead to a
better understanding of signal-response coupling mechanisms in plant-pathogen
interactions, and the development strategies to manipulate and enhance the response of

plant cells to biological stress (Schenk et al., 2000).

99

chm/3m $5: 80538 me he ween—«5 320:82 e
Khmer—mu. wan: wagon—Vow me he mama—«ca $2255 a
th<qm win: 80533 me he women—mam $225: a

«You 5053, mombocow E :5on moan—ma meow- ” N960 methane moabocow 5 E085 33 9:3 05 mocewa NV-oD+ a

 

 

 

2% N88- 836588 .5 385 8:883 8:285; Reggae. o2 m8
8-6m 84-8- 898588 .8: m 8&5 5889286: 83865 SmeEa we to
MW.2 84-8- .265 NV 8:88:88 8%2838286& 66:88: 638814.. 5 B
2-8 N4.8+ 88.68 .9 8828585298 2585 8885a. mm: 30
4-3 N4.8+ .9283; 5 asbofimoioi 88:389. 2 A 33%.. RN 3
2-2 N4.8+ 82:8: .3 A 6.85858 8%: 888%» m3 8
E2883». 2% >553 .0: an:
029 m 25w mo 88:3 0353 c2383.. fiwcoq me
.Eoficoaxo

3:35:00 Bot E033 cofimoaxo :05 new Z mefim use wager—m was: awe—CEO: oocoscom me we mug—«~24 fin 039—.

100

575898 E 838% @8865 a £23055 5 03865 82865 e $38.3 wean gauges—am 3 E033 scumbag a

 

 

 

“.8 £523 a 95:85 .9 589a base 83-59 swam =N 6889;» m2 S
8.8 8-9 a .225 a .35 a 988 .9 88326 828826 .m 5826 meageea _afieozo 83838 N: as
2-2 some .528 A IN a 8558: .5 H mm .985 896a 3238} _afieozu 88.93% N2 8
83 c.4358 .9 386.5 8.522 439319. oz. .2:
2.8 :58 a .225: 6-5. A 88.2% .9 5265 85-683858 25.: 63883 8888:» as. 3:

8-8 8-52 a fire a .31: A $583.86 .9 6835 886852-88 _.:Rm._<<_n_w g .2
2-“: Agrzfiu. .Qv 530.5 Eutom woumooou o>§wusm whmfigfimrw h©~ v.2
2-3 ESE e €828. .9 £868 Emu 258$ magnesia. 02 a:

56855 25w .o: 33

2:? m «meow ho E033 35:03 9685 56803.. Ewan; me

653 083

05 Bob uBSo£ 02a 80% NM: ecu :2 mmnE. 6:3 2:2 05 88... c8309 89$ v.2 USN _.: man; .Bcflm 8338:7988 ES

@3238“ mo max—«SN %-<ZQ0 Soc 50:3 58855 :05 EB Xhmfim wEms 320:8: 8:258 me no woman—«ER NM 935.

101

Inoculated Mock-inoculated
" 8

  

   
   
  

 

' 1; fl go o 0.5 1 ea... .11.: as s- 33. “.16 Egg

 

2.;4 6.81216 20

la. w «rm- “3'???“

 

"6 g 12 '16 '20 o 7635 .1

         

Figure 3.1 cDNA—AFLP expression profiles across time points 0, 0.5, 1, 2, 4, 6, 8, 12, 16
and 20 hpi for inoculated and mock-inoculated treatments. (A) TDF 11.1 and 11.4 (B)
TDF 18.1 and 18.2 (C) TDF 22 and 23, although TDFs 22 and 23 were isolated from
different bands, their expression pattern is identical (D) TDF 31, note that time point 16h

' "‘1" 1:13;

 

is missing.

102

A)

   

 

1.0 2.0 3.0 9° 6.0 6.0
1 gm Hmonzmm-m° cm 6543 ' G. GlSIHGL 1-5091620
1 118.1
61 I _"y ' ‘ ' veg-41111111. i141 quwrgaucp g1_6091620
1 - " git/111411011“ MA RRDlAl-‘EIAL‘PGGY TDF18.1

 

 

    
 

ago 190 11,0 1350 110 teo
531.331: 1. v1 : " , . . 3511;01:11511-SPJK {ilVSAf'i 31-5091620
G-LI‘UEE L: 1. Hanan ‘ sitiiAVEEm-:;-;Pi‘~nta211151131 roman

  

 

gi_5091620
TDF18.1

190
181 :1. 1)ng
102911..sz F 0.41
[3] non conserved
I similar
conserved
all match

60
raga 1_4913228
a. . . 518.2

 

 

     

 

 

 
  

  

79 . 190 110 12
61 DINYEm‘ ' ‘ MN C = ~.wg'rtxi'detstmi‘ 313918228
12 H SCGR ......... H I E . :1; Km {SDLLPY TDF18.2
l t§0 11o 1§o 190 120 1§0
121 Al-"L._ -‘st‘{'ig '1 avzarcwoc " .--v.'-_:::FA:> 'ITYABSV‘IFVP gi_4918228
63 chv‘vvvn KLKPSP WiéTCVDHV "-41 0D I'M-‘18.:

  

 

non conserved
| similar

B conserved

fl all match

E] BTB domain
Figure 3.2 Sequence alignment generated by ClustalW. Putative protein domains
detected by the BLASTP program are boxed. A) Alignment with gi|509l6205 and TDF

18.1 B) Alignment with gi|49182284 and TDF 18.2.

103

LITERATURE CITED

Allen, L.J., MacGregor, K.B., Koop, R.S., Bruce, D.H., Kamer, J., and Bown, A.W.
1999. The relationship between photosynthesis and a mastoparan-induced
hypersensitive response in isolated mesophyll cells. Plant Physiol. 119: 1233-1242.

Awale, HE. and Kelly, J .D. 2001. Development of SCAR markers linked to C042 gene
in common bean. Annu. Rept. Bean Improv. Coop. 44: 1 19—120.

Bachem, C.W., van der Hoeven, R.S., de Bruijn, S.M., Vreugdenhil, D., Zabeau, M., and
Visser, R.G. 1996. Visualization of differential gene expression using a novel method
of RNA fingerprinting based on AFLP: analysis of gene expression during potato
tuber development. Plant 9:745—753.

Balardin, R.S., Jarosz, A., and Kelly, JD. 1997. Virulence and molecular diversity in
Colletotrichum lindemuthianum from South, Central and North America.
Phytopathology 87:1184-1191.

Bardwell, V]. and Treisman, R. 1994. The POZ domain: a conserved protein-protein
interaction motif. Genes Dev. 8: 1664-1677.

Bell, J.N., Ryder, T.B., Wingate, V.P.M., Bailey, J.A., and Lamb, CJ. 1986. Differential
accumulation of plant defense gene transcripts in a compatible and an incompatible
plant—pathogen interaction. Mol. Cell. Biol. 6: 1615—23.

Bestwick, C.S., Brown, [R., and Mansfield, J.W. 1998. Localized changes in peroxidase
activity accompany hydrogen peroxide generation during the development of a non-
host hypersensitive reaction in lettuce. Plant Physiol. 118:1067-1078.

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

Cassab, 6.1. and Vamer, J.E. 1988. Cell wall proteins. Ann. Rev. Plant Physiol. Plant
Mol. Biol. 39:321-353.

Cao, H., Bowling, S.A., Gordon, S., and Dong, X. 1994. Characterization of an
Arabidopsis mutant that is nonresponsive to inducers of systemic acquired resistance.
Plant Cell 6:1583-1592.

Cao, H., Glazebrook, J., Clarke, J.D., Volko, 8., and Dong, X. 1997. The Arabidopsis
NPRl gene that controls systemic acquired resistance encodes a novel protein
containing ankyrin repeats. Cell 88:57-63.

Cramer, C. L., Ryder, T. B., Bell, J .N., and Lamb, C. J. 1984. Rapid switching of plant
gene expression induced by fungal elicitor. Science 227: 1240-43.

104

Day, 1.8., Reddy, V.S., Ali, GS, and Reddy, A.S.N. 2002. Analysis of EF-hand-
containing proteins in Arabidopsis. Genome Biology 3:56.1-56.24.

Delaney, T.P., Friedrich, L., and Ryals, J.A. 1995. Arabidopsis signal transduction
mutant defective in chemically and biologically induced disease resistance. Proc.
Natl. Acad. Sci. USA 92:6602-6606.

Delaney, T.P. 1997. Genetic dissection of acquired resistance to disease. Plant Physiol.
113:5-12.

Dennis, D.T. and Blakely, SC. 2000. Carbohydrate Metabolism. Pages 630-675. In:
Biochemistry & Molecular Biology of Plants. B. Buchanan, W. Gruissem, and R.
Jones (eds.). American Society of Plant Physiologists, Rockville, Maryland.

Dietrich, R.A., Richberg, M.H., Schmidt, R., Dean, C., and Dang], J.L. 1997. A novel
zinc finger protein is encoded by the Aribidopsis SDI gene and functions as a
negative regulator of plant cell death. Cell 88:685-694.

Dixon, R.A., Dey, P.M., Lawton, M.A., and Lamb, CJ. 1983. Phytoalexin induction in
French bean. intercellular transmission of elicitation in cell suspension cultures and
hypocotyl sections of Phaseolus vulgaris. Plant Physiol. 71:251—256.

Dixon, R.A., Harrison, M.J., and Lamb, C.J. 1994. Early events in the activation of plant
defense responses. Annu. Rev. Phytopathol. 32:479—501.

Edwards, K., Cramer, C.L., Bolwell, G.P., Dixon, R.A., Schuch, W., and Lamb, CJ.
1985. Rapid transient induction of phenylalanine ammonia-lyase mRNA in elicitor-
treated bean cells. Proc. Natl. Acad. Sci. USA 82:6731—6735.

Fleurence, J. and Negrel, J. 1989. Partial purification of tyramine feruloyl transferase
from TMV inoculated tobacco leaves. Phytochemistry 28:733-736.

Fraire-Velazquez, A. and Lozoya-Gloria, E. 2003. Differential early gene expression in
Phaseolus vulgaris to Mexican isolates of Colletotrichum lindemuthianum in
incompatible and compatible interactions. Physiological and Molecular Plant
Pathology 63:79-89.

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

Furumoto, T., Ogawa, N., Hata, S., and Katsura, I. 1996. Plant calcium-dependent protein

kinase-related kinases (CRKs) do not require calcium for their activities. FEBS
Letters 396:147-151.

105

 

Garrett, RH. and Grisham, C.M. Nucleotides and nucleic acids. Pages 241-291. In:
Principles of Biochemistry with a Human Focus. S. Kiselica and B. Boehme (eds.).
Harcourt College Publishers, Orlando, FL.

Genoud, T. and Metraux, J.-P. 1999. Crosstalk in plant cell signaling: structure and
function of the genetic network. Trends in Plant Science Perspectives 4:503-507.

Gepts, P. and Debouck, D.G. 1991. Origin, domestication and evolution of the common
bean, Phaseolus vulgaris. Pages 7-53. In: Common Beans: Research for Crop
Improvement. A. van Schoonhoven and O. Voysest (eds). CAB. International,
Wallingford, UK.

Gepts, P. 1999. Development of an integrated linkage map. Pages 53-91. In:
Developments in Plant Breeding Vol. 7. Common bean Improvement in the
Twenty-First Century. S.P. Singh (ed.). Kluwer Acad. Pub., Dordrecht, The
Netherlands.

Gross, G.G. 1977. Cell wall-bound malate dehydrogenase from horseradish.
Phytochemistry 16:319—321.

Hammond-Kosack, KB. and Jones, J .D. 1996. Resistance gene-dependent plant defense
responses. Plant Cell 8: 1773-1791.

Hardie, D.G. 1999. Plant protein serine / threonine kinases: classifications and functions.
Annu. Rev. Plant Physiol. Plant Mol. Biol. 50:97-131.

Hedrick, S.A., Bell, J .N., Boller, T., and Lamb, C]. 1988. Chitinase cDNA cloning and
mRNA induction by fungal elicitor, wounding and infection. Plant Physiol. 86:182—
186.

Hrabak, E.M., Chan, C.W.M, Bribskov, M., Harper, J.F., Choi, J .H., Halford, N., Kudla,
J ., Luan, S., Nimmo, H.G., Sussman, M.R., Thomas, M., Walker-Simmons, K., Zhu,
J .K., and Harman, AC. 2003. The Arabidopsis CDPK-SnRK superfamily of protein
kinases. Plant Physiol. 132:666-680.

Hsieh, M.C. and Graham, TL. 2001. Partial purification and characterization of a
soybean beta-glucosidase with high specific activity towards iso-flavone conjugates.
Phytochemistry 58:995—1005.

Kelly, JD. and Vallejo, V.A. 2004. A comprehensive review of the major genes
conditioning resistance to anthracnose in common bean. Hort.Science 39:1196-1207.

Kombrink, E. and Schmelzer, E. 2001. The hypersensitive response and its role in local
and systemic disease resistance. Eur. J. Plant Pathol. 107269—78.

106

 

 

Louis, V., and Negrel, J. 1991. Tyramine hydroxicinnamoil transferase in the roots of
wheat and barley seedlings. Phytochemistry 30:2519-2522.

Mahe, A., Grisvard, J., Desnos, T., and Dron, M. 1992. Bean-Colletotrichum
lindemuthianum compatible interactions: time course of plant defense responses
depends on race. Mo]. Plant—Microbe Interact. 5:47 2—478.

Martin-Tanguy, J. 2001. Metabolism and function of polyamines in plants: recent
development (new approaches). Plant Growth Regulation 34:135-148.

Muller-Rober, B.T., Kossmann, J ., Hannah, L.C., Willmitzer, L., and Sonnewald, U.
1990. One of two different ADP-glucose pyrophosphorylase genes from potato
responds strongly to elevated levels of sucrose. Mol. Gen. Gent. 224: 136-146.

O’Connell, R.J., Bailey, J.A., and Richmond, D.V. 1985. Cytology and physiology of
infection of Phaseolus vulgaris by Colletotricum lindemuthianum. Physiol. Plant
Pathol. 27:75—98.

Ryals, J .A., Weymann, K., Lawton, K., Friedrich, L., Ellis, D., Steiner, H.Y., Johnson, J.,
Delaney, T.P., Jesse, T., Vos, R, and Uknes, S. 1997. The Arabidopsis NIMl protein
shows homology to the mammalian transcription factor inhibitor IKB. Plant Cell
92425-439.

Ryder, T.B., Cramer, C.L., Bel], J .N ., Robbins, M.P., Dixon, R.A., and Lamb, CJ. 1984.
Elicitor rapidly induces chalcone synthase mRNA in Phaseolus vulgaris cells at the
onset of the phytoalexin defense response. Proc. Nat]. Acad. Sci. USA 81:5724-5728.

Schaaf, J ., Herbert, W., and Hess, D. 1995. Primary metabolism in plant defense. Plant
Physiol. 108:949-960.

Schenk, P.M., Kazan, K., Wilson, I., Anderson, J .P., Richmond, T., Somerville, SO, and
Manners, J .M. 2000. Coordinated plant defense responses in Arabidopsis revealed by
microarray analysis. Proc. Natl. Acad. Sci. USA 97:11655—11660.

Sedgwick, SO. and Smerdon, SJ. 1999. The ankyrin repeat: a diversity of interactions on
a common structural framework. Trends Biochem. Sci. 24:311-316.

Simmons, C.R., Grant, 8., Altier, D.J., Dowd, P.F., Crasta, O., Folkerts, O., and Yalpani,
N. 2001. Maize rhml resistance to Bipolaris maydis is associated with few
differences in pathogenesis-related proteins and global mRNA profiles. Mol. Plant—
Microbe Interact. 14:947—954.

Smith, C]. 1996. Accumulation of phytoalexins: defense mechanism and stimulus
response system. New Phytol. 132: 1-45.

107

 

 

Thompson, J.D., Higgins, D.G., and Gibson, T.J. 1994. CLUSTAL W: improving the
sensitivity of progressive multiple sequence alignment through sequence weighting,
position-specific gap penalties and weight matrix choice. Nucleic Acids Res.
22:4673-4680.

Wink, M. and Hartmann, T. 1982. Localization of the enzymes of quinolizidine alkaloid
biosynthesis if leaf chloroplasts of Lupinus polyphyllus. Plant Physiol. 70:74-77.

Yoshimura, S., Yamanouchi, U., Katayose, Y., Toki, S., Wang, Z., Kono, I., Kurata, N .,
Yano, M., Iwata, N., and Sasaki, T. 1998. Expression of Xa1, a bacteria] blight-
resistance gene in rice, is induced by bacterial inoculation. Proc. Natl. Acad. Sci.
USA 95: 1663-1668.

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

Zollman, S., Godt, D., Prive, G.G., Couderc, J.-L., and Laski, EA. 1994. The BTB
domain, found primarily in zinc finger proteins, defines an evolutionary conserved
family that includes several developmentally regulated genes in Drosophilia. Proc.
Nat]. Acad. Sci. USA 91:10717-10721.

108

P?

_——‘t ————---

 

APPENDICES

109

 

 

APPENDIX A1

INITIAL DISSECTION OF THE ANTHRACNOSE RESISTANCE IN THE
LANDRACE CULTIVAR G 2338

INTRODUCTION

Anthracnose, is a seedbome disease of common bean (Phaseolus vulgaris),
caused by the hemibiotrophic fungus, Colletotrichum lindemuthianum. The devastating
effect of anthracnose on yield, particularly in tropical production areas, makes it one of
the most economically important diseases of bean. Several disease management
strategies are available, however, planting genetically resistant cultivars is the most
effective and least expensive way for growers to prevent yield loss due to anthracnose
(Pastor-Corrales et al., 1995). The best breeding strategy to generate resistant cultivars is
by pyrarrriding anthracnose resistance genes with complementary spectra of resistance
(Kelly and Vallejo, 2004). The highly variable nature of the pathogen, however, requires
that new sources of resistance are constantly being characterized. The landrace cultivar
G 2333, from Chipas, Mexico, is the most resistant cultivar of the anthracnose differential
series and carries three independent resistance genes (Co-42, C0-5 and Co-7). Recently, a
highly virulent C. lindemuthianum race of Middle American origin (Costa Rica) has been
identified which overcomes the resistance in G 2333. Interestingly, the landrace G 2338,
also from Chiapas, Mexico is resistant to this race (J. Kelly personal communication).

Since both landraces are from the same region of Mexico, we hypothesize that the

110

 

 

 

differential resistance is conferred by a small difference in their allelic composition,
therefore, we postulate that G 2338 carries a similar anthracnose gene pyramid. The
objective of this study was to initialize the molecular and genetic characterization and

dissection of the anthracnose resistance genes carried by G 2338.

MATERIALS AND METHODS

Initial characterization of the anthracnose resistance in G 2338 was carried out
using molecular markers linked to known anthracnose resistance genes (Table A1.l).
Additionally, G 2338 was previously tested for the presence of several SCAR markers
linked to the Co-4 locus and the Co-42 allele (Table A1.2) (Awale and Kelly, 2001). The
number of genes conditioning resistance to anthracnose in G 2338 was determined by
inoculating a F2 population of 201 individuals from the cross La Victorie (susceptible) x
G 2338 (resistant) with race 7. Based on the results of the initial characterization with
molecular markers linked to anthracnose resistance genes, a SCAR marker was used to

begin the molecular dissection of the gene pyramid in G 2338.

RESULTS AND DISSCUSSION

The initial characterization of the anthracnose resistance genes carried by G 2338,
based on analysis with molecular markers linked to various known genes, indicated that
G 2338 carries the Co-5 and C0-4 loci (Table Al.l). Since allele specific primers exist

for the Co-4 locus, we wanted to determine if the G 2338 carries the same allele as G

111

 

2333 at the C0-4 locus. The results displayed in Table A1.2 indicate that G 2338 does
not carry the same allele at the Co-4 locus as G 2333. To determine the number of genes
segregating for resistance to anthracnose a F2 population (La Victorie x G 2338) was
inoculated with race 7 which yields a SxR reaction in the parents. Out of the 201 F2
individuals inoculated, 9 were lost to lethal combinations of dwarfing genes due to the
intergene pool nature of the cross. The remaining progeny segregated in a 63:1 ratio of
resistant to susceptible indicating that three genes are segregating for resistance to race 7
(p-value= 0.244). Based on our preliminary gene characterization using molecular
markers we believe that two of the genes are C0-5 and C0-4. To begin the molecular
dissection of this gene pyramid, we used the SCAR marker SAB3 linked to the C0-5
locus to select against this locus among the surviving progeny of the F2 La Victorie x G
2338 p0pulation. Unfortunately, we were not able to elucidate the identity of the third
gene conditioning resistance to anthracnose in G 2338. However, based on the similar
origin of both of these landraces and the fact that they both have three genes conditioning
resistance to anthracnose two of which are the same, we speculate that the third gene in G
2338 is very likely to be Co-7. Given the allele complexity at the Co-4 locus we believe
that the resistance differential between G 2333 and G 2338 is due to differences in the

allelic composition at the C0-4 locus.

112

 

 

Table A1.1 RAPD markers linked to anthracnose resistance genes

 

 

Molecular marker Linked resistance gene Linkage phase +/-al
F10530 Co-I repulsion -
Q41440 C0-2 coupling -
H20500 Co-2 coupling -
AB3400 C0-5 coupling +
AH1730 Co-6 coupling -
AK20390 C0-6 repulsion -
AS 13950 C 0-4 coupling +
SB 12 C0-9 coupling -

 

3 Presence (+) or absence (-) in G 2338

Table A1.2 SCAR markers linked to the C0-4 locus or specific alleles.

 

 

SCAR marker Locus G 23333 G 2338
SH18 €0-47- + -
SAS 13 C0-4 + +
53314 C0-4 + +

 

113

REFERENCES

Awale, HE. and J .D. Kelly. 2001. Development of SCAR markers linked to Co-42 gene
in common bean. Annu Rept. Bean Improv. Coop. 44:119-120.

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

Pastor-Corrales MA, Otoya MM, Molina A, and Singh SP. 1995. Resistance to

Colletotrichum lindemuthianum isolates from Middle America and Andean South
America in different common bean races. Plant Dis 79: 63—67.

114

APPENDIX A2

UNEXPECTED RESISTANCE GENES UNCOVERED

Genetic improvement in dry bean is achieved through a combination of traditional
and molecular selection techniques. The resulting cultivars are pure lines which are
improved for one or more specific traits. Despite the homozygous nature of bean
cultivars, sources of genetic variation may be mutations, insect cross pollination or the
cultivars’ heterogeneity (Traka-Mavrona et al., 2000). In the absence of direct selection,
many pure-line cultivars may be heterogeneous for major resistance genes. The
following are case studies in which new resistance genes were unexpectedly uncovered in

pure line cultivars.

Black Magic

The black bean cultivar, Black Magic (BM), was reported to carry resistance to
races beta, gamma and delta of C. lindemuthianum (Kelly et al., 1987). In the currently
used nomenclature for race classification, these are races 130, 102 and 23 respectively
(Melotto et al., 2000). Black Magic, however, unexpectedly segregated for resistance to
race 7 (Melotto, 1999). Although BM carries resistance to anthracnose races 23, 102 and
130, anthracnose resistance was not a specific goal in cultivar development (Kelly,

personal communication). Our hypothesis is that the resistance to race 7 in BM is

115

conferred by a single dominant gene. To determine the number of genes conferring
resistance to race 7 in BM a F2 population was generated from the cross BM (resistant to
race 7; R7) x BM (susceptible to race 7; S7). The resulting population was inoculated
with race 7 (Table A21). The F2 population, BM (R7) x BM (S7), segregated in a 15:1
ratio of resistant to susceptible individuals (p-value = 0.936) (Table A2.1) indicating that
the resistance to race 7, in the cultivar BM, is conditioned by two dominant resistance
genes. Allelism tests must next be canied out to determine if these genes are independent
of the other previously described anthracnose resistance genes. We believe that this
unexpected segregating resistance to anthracnose is a result of the cultivars natural
heterogeneity which exists for specific traits which are not under direct selection during

cultivar development.

Mexico 222

The anthracnose differential cultivar Mexico 222 is believed to carry a single
dominant gene, C0-3 which conditions resistance to anthracnose (Kelly and Vallejo,
2004). To determine if the anthracnose resistance in the MSU breeding line, MSU7-4,
was allelic to Co-3, a F2 population was generated from the cross Mexico 222 x MSU 7-
4. The resulting population was inoculated with race 7, which results in a RxR reaction
in the parents (Table A21). The Mexico 222 x MSU7-4 F2 population segregated for
resistance at a ratio of 63:1 (p-value = 0.979), resistant to susceptible individuals. This
indicates that there are three genes segregating for resistance to race 7 in this population.
Since race 7 elicits a RxR reaction in the parents (Table A21), and it was previously

determined that MSU7-4 carries only one gene conditioning resistance to race 7 (chapter

116

two of this dissertation), we conclude that two anthracnose resistance genes must be
segregating from the cultivar Mexico 222. One of the anthracnose resistance genes in
Mexico 222 is known to be Co-3 and the other remains unknown. Neither of these
anthracnose resistance genes are allelic to the gene carried by MSU7-4. Heterogeneity
within the cultivar can be used to explain this unexpected discovery of an additional
anthracnose resistance gene in Mexico 222. Alternatively, with the abundance of
different C. lindemuthianum races available, and the implicit gene-for-gene relationship
between resistance in the plant and pathogenicity in the pathogen, it is not surprising if
some anthracnose resistance genes go undetected in certain population/race inoculation
experiments. Additional characterization of the second anthracnose resistance gene in
Mexico 222 is needed to determine if it is independent of the previously reported

anthracnose genes and what spectrum of resistance it offers.

117

Table A2.1 F2 populations inoculated with race 7 of C. lindemuthianum.

 

 

Population or Race Disease Observed Expected p-value
Genotypes Inoculated Reaction Ratio“ Ratioa

BM (R7) x BM (s7)b 7 - 87:6 15:1 0.936
Mexico 222 x MSU7-4 7 - 249:4 63:1 0.979
Mexico 222 7 R - - -
MSU7 7 R - - -

 

“ Ratio of resistant to susceptible individuals

b R7 = resistant to race 7, S7: susceptible to race 7

118

REFERENCES

Kelly J.D., Adams, M.W., Saettler, A.W., Hosfield, G.L., Uebersax, M., Ghaderi, A.
1987. Registration of ‘Domino’, ’Black Magic’ tropical black beans. Crop Sci 27:
363

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

Melotto, M. 1999. Genetic inheritance and Molecular cloning of anthacnose resistance
genes in common bean. Ph. D. dissertation pp. 102-107. Michigan State University

Melotto, M., R.S. Balardin, and J .D. Kelly. 2000. Host-pathogen interaction and
variability of Colletotrichum lindemuthianum, p. 346-361. In: D. Prusky, S. Freeman,
and MB. Dickman (eds.). Colletotrichum Host specificity, pathology, and host-
pathogen interaction. APS Press St. Paul MN.

Traka—Mavrona, E., Georgakis, D., Koutsika-Sotiriou,M. and Pritsa, T. 2000. An

integrated approach of breeding and maintaining an elite cultivar of snap bean.Agron
J. 92: 1020-1026.

119

   

llllllllllllllllllllllllllllll

1293 0273