‘1

8..

v
'1

 

 

             

.t .fifitdohfifih .
. .

«16:2? .5 ..
. away 1%.
A I ‘1 -n3

m2

 

1‘

 

 

'1
'3'
d

i;
"x

    
 

       

           

  

       

       

       

           

  
           

     

      

 

    

 

 

 

 

 
           

XML:
.314.
Vll‘Ils
.MuMA . a
nu“ f5, V
. , . .. L 1.4-... . . ..
, fl. . _
. influV “sq”: . ; LII . r Jflubloi .. . .v
. . . .Mme L... . . 1 find“ LVHLH‘MufiVLquWW “$11....
. .. . . .7 .. . , h. wt} nvnhrfliu, ...J.uuv:...fl..(
. . . . . k... .3 . .. . ..1Lmr nailul. Wtaufltfibumuanuvlrflfnnj
a .nWWakwfid‘umimm , ..1 ..: an“ 4525‘. 4km...l$vmonbrrlm.wmmo
; . .14 .. . . ) 4 - - .I.
«a. A. rnnd J L. ,1: hundfiflfiuhurxlmWMWfiumflfirre h.
L. .11.}...1 n. 3?". . . “$3. - ..«¢..:|Js.m.s...u1\.lu.fuufiunn,lflu.r«..
- 4M3; .. 3a. , . , ‘ .

 

I
.
A-
$5.3... ..M.m.u.w&w5.

 

                

‘- ‘ I l
{N} “mm“..fi 1).. ”unzlflntflfliln} ¢
51 it. .42... Kw). . HQ. 4 .
awalernJoaiilfiuzx. . GRIN. u... a
. ‘Lpll I“ ”nad‘rltv
‘ fi.l.du1.t1::hz In": t.
“.30).. (a. .DDUMU; .UJI“:..:II§NI t

it.

’1.

      

é

        

  

       

 

.vaHdu, INN”.

v11 IOIVMQ. l a n 1

3.6.3. .29. o. afidl.‘k.fi!.fll.

.‘Vpfirhhdkwma. «litauiiluqfibm
(linin'aiic. . avdpch

   

 

       

 

. .
mafianmwmflhph 3.33 {3.115.}. 1..
.l 51:. Jififflwsbnt 1.3!}?

    

a
ca 9‘...» :1. .19 .IdtulA-J n‘a .
. . é . 3:1\J)I. r].
.31 451.1 fin).fil.i.l¢4v..v . 1'05““! 5’! o , .
.f-lnl30 ulbfil. ntVoltH..:thtcVn'l.lvlbliOa "II-tutti“ 9' ”WI Isl , 1 \4 M J
.1? 44 .‘IIHIIAIQA‘ 01-h: . .2 a 444 . .Al - 0‘4 r..i1l..§ r lluui.‘ it)?! 11b \ : ‘O’T’nlm
1:10 vgalt‘lctot t..all..!:nvl| . ‘3’!!! “NH! 1;. t
.‘1 .a‘..snpntdl.126¢vllufvlbv’ l-tlobul' I! v. on, -( q . u
I ..v v.) 241......lcvlhllnoio it n .0 0|!
. WA: .uv-ln 7000'.) .I lefioiilla, u .llll‘ \U
. . a u. .l. IaAn
o". i

I‘. v

 

._,I
i:
I:
'1
is
51
g.
:

 

yvlinizills It! 1 .
. .o'.to 1O)..§1'3 «ot‘v‘blolxal:
. 110lorv34§zl1é ‘Il‘
in)"; 10"!!! (5’!
iv: 0"

 

      

. .«IviXoIIR
13134.1»...‘utfii... .
bfinhuthvlrflflhoarli); :11 iii. ¢ ‘01:;
c. ._..c!l-kh‘..(nbn\ttoirufu-$oi Honijai..;a'.nv.flul:an%ul.
. 3..-: {1.73: 3 .35.] -I (III!
4.9.:41AHP'IUrInoIoSP Jotlbmuflff.) .1 1.3.5 \byibl.
.X‘vpvsl tltllv all:
(‘1' «1|-

 

   

     

«ll
31.1.)..11UI. 5.5% :1
I - V'ig‘zt. ,
4.! .15.».-.vvpnui 3.1.

£ if»?
{Illillllco
53:...1‘ III»
. Mm.
.

tflfirbb iii a I;

   

 

 

 

IIIIIIIIIIIIIIIIIIIIILIIIIH IIII III IIIIIII IIIIIII

31293

LIBRARY
Michigan State
University

 

 

 

 

This is to certify that the

dissertation entitled

Traditional and molecular characterization of
variability in Colletotrichum lindemuthianum

 

presented by

Ricardo Silveiro Balardin

has been accepted towards fulfillment
of the requirements for

Doctor of Philosophy degreein Crop & Soil Sciences/
Botany & Plant Pathology

///Mg//¢

Major professor

 

 

Date November 20, 1997/

 

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

PLACE IN RETURN BOX
to remove this checkout from your record.
TO AVOID FINE return on or before date due.

 

MTE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1/” Mil

TRADITIONAL AND MOLECULAR CHARACTERIZATION OF
VARIABILITY IN COLLETOTRICHUM UNDWUTHMWM

By

Ricardo Silveiro Balardin

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Crop and Soil Sciences Department
Botany and Plant Pathology Department

Dr. James D. Kelly

1997

ABSTRACT

TRADITIONAL AND MOLECULAR CHARACTERIZATION OF
VARIABILITY IN COLLETOTRICHUM LINDEAJUTHIANUM

By

Ricardo Silveiro Balardin

Colletotrichum lindemuthianum isolates were characterized into 41 races based on
virulence to twelve differential cultivars of Phaseolus vulgaris. Races 7, 65, and 73 were
widespread. No race was isolated from the Andean and the Middle American gene pools
of P. vulgaris although 39% of the races were detected multiple times. Phenetic analyses
showed no obvious geographical patterns correlated with virulence clusters. Genetic
diversity of C. lindemuthianum was shown to be the greatest in Central America.
Diversity was estimated using sequence homology and RFLP analysis of the ribosomal
subunit spacer (ITS), and RAPD analysis of total genomic DNA Polymorphism in the
rDNA spacer region was not linked to any specific genetic factor. Parsimony and
neighbor-joining analyses supported a monophyletic group formed by all except race 31.
RFLP-ITS analysis placed Andean races predominantly into group I except race 23,
which was placed within group II. The Middle American races were observed in both
groups. Molecular polymorphism among isolates of similar virulence phenotype revealed
a level of molecular variability within C. Iindemuthianum greater than the variability

characterized using virulence analysis. Thirty-four races of C. lindemuthianum were

inoculated on sixty-two cultivars of P. vulgaris. Bean genotypes clustered based on the
gene pool origin of the resistance genes present, regardless of the actual gene pool of the
host genotype. Races of C. lindemuthianum with Middle American reaction showed
broad virulence on germplasm from both gene pools, whereas races with Andean reaction
showed high virulence only on Andean germplasm. The reduced virulence of Andean
races on Middle American genotypes suggests selection of virulence factors congruent
with diversity in P. vulgaris. The majority of races of C. lindemuthianum grouped
according to specific gene pool (ie. Middle American and Andean) based on principal
component analysis, except a small group of isolates which appeared to possess factors of
virulence to both host gene pools. No apparent geographic efl‘ect was observed. Virulence
data supported variability in C. Iindemuthianum structured with diversity in Phaseolus,

whereas molecular data showed no congruence between pathogen and host populations.

DEDICATION

To my parents, my inspiration.

To my wife, Clarice, my strength and support all time long.

To my children, Ricardo and Gabriela, I hope they have dignity and force to get the best

in their lives.

ACKNOWLEDGMENTS

I would like to express my gratitude to all those individuals that helped me to
fulfill my doctoral program:

Dr. James Kelly, my academic advisor, for his guidance throughout my doctoral
program, for his support in all stages of my work, and for his fiiendship.

Drs. David Douches, Patrick Hart, and Gerard Adams, for serving on my
guidance committee, and giving helpful advise during my research studies.

Lucia, Maeli, Roberto, Jorge, Kristin, Judy, Halima, for the various help in the lab
and green house, for the discussions on many topics, and for the fiiendship.

Ministe’rio da Educacio e do Desporto, Coordenacio de Aperfeicoamento de
Pessoal de Nivel Superior, for the excellence of support in all stages of my program.

Universidade Federal de Santa Maria, Centro de Ciéncias Rurais, Departamento
de Defesa Fitossanitfiria, for supporting my candidacy to come to the Michigan State
University to pursue a PhD. degree.

Michigan State University, College of Agriculture and Natural Resources, Crop
and Soil Sciences Department, and Botany and Plant Pathology Department, for
accepting me as doctoral student, and provide the best conditions to complete my
education.

My friends, for the support in many moments of my life.

My family, for the unconditional support and love.

TABLE OF CONTENTS

LIST OF TABLES ..........................................................................
LIST OF FIGURES ........................................................................
GENERAL INTRODUCTION ...........................................................

List of references .....................................................................

CHAPTER 1

VIRULENCE AND MOLECULAR DIVERSITY IN COLLE T OTRICHUM

LINDEMUTHIANUM FROM SOUTH, CENTRAL AND NORTH AMERICA
Abstract ...............................................................................

Results .................................................................................

CHAPTER 2

RIBOSOMAL DNA POLYMORPHISM IN C OLLE TOTRICH UM

LINDEIWUTHIANUM .....................................................................
Abstract ...............................................................................

Results .................................................................................

CHAPTER 3

INTERACTION AMONG VARIABILITY IN COLLET OTRICH UM
LINDEMU T HIANUM AND DIVERSITY IN PHASEOLUS VULGARIS .........

16

24
24
26
28
39
49
53

58
58
60
62
69
80
86

89

Abstract ............................................................................... 89

Introduction ........................................................................... 91
Materials and Methods ............................................................... 93
Results ................................................................................. 99
Discussion ............................................................................. l 12
List of references ..................................................................... 118
GENERAL CONCLUSIONS ............................................................. 122
APPENDIX ................................................................................. 125
Appendix A ........................................................................... 125
Appendix B ........................................................................... 132
Appendix C ........................................................................... 133
Appendix D ........................................................................... 134
Appendix E ........................................................................... 138

Table 1.1

Table 1.2

Table 2.1

Table 2.2

Table 3.1

Table 3.2

Table A. 1

LIST OF TABLES

Race designation, number of isolates, and reaction of common bean

differential cultivars (+, susceptible and -, resistant) of 41 races of C.
lindemuthianum identified from 138 isolates collected on Phaseolus
hosts fi'om Middle American and Andean gene pools grown in the

listed countries. ..............................................................

Susceptible (+) and resistant (-) reaction of common bean differential
cultivars to races of C. lindemuthianum. Races were previously
reported in Brazil, Canada, Colombia, Guatemala, Mexico, the
Netherlands, Peru, and the United States and characterized using
RAPDs in the present study. ...............................................

Susceptible (+) and resistant (-) reaction of common bean difi‘erential
cultivars to 57 isolates of C. lindemuthianum. Isolates were collected
in Argentina, Brazil, Canada, Colombia, Costa Rica, the Dominican
Republic, Honduras, Mexico, Netherlands, Peru, and the United
States, and characterized using RFLP-ITS analysis. ..................

Sequence pair distances estimated by the Juices-Cantor one-parameter
method of C. Iindemuthianum isolates. ..................................

Identification, origin, gene pool, race, seed weight, and resistance
index of common bean genotypes (P. vulgaris L.) fi'om Brazil,
Colombia, the Dominican Republic, the Netherlands, Honduras,
Mexico and the United States inoculated with 34 races of C.
lindemuthianum from Argentina, Brazil, Colombia, Costa Rica, the
Dominican Republic, Honduras, Mexico, Peru, and United States.

Identification, origin, reaction group, and virulence index of 34 races
of C. lirrdemuthianum. ......................................................

Race designation and origin of C. Iindemuthianum isolates fi'om
samples collected on Phaseon hosts fi'om difi'erent gene pools in
Argentina, Brazil, the Dominican Republic, Honduras, Mexico and
the United States. ...........................................................

30

34

63

74

94

100

Table B. 1

Table D.1

Table E.1

Anthracnose differential series and the binary number of each

Cultivar. ....................................................................... 132
Reaction of 62 genotypes of P. vulgaris to 20 races of C.
lindemuthianum. ........................................................... 134
Reaction of 62 genotypes of P. vulgaris to 14 races of C.
lindemuthicmum. ........................................................... 138

Figure 1.1

Figure 1.2

Figure 1.3

LIST OF FIGURES

Phenogram of C. Iindemuthianum races based on virulence to
Phaseolus differential cultivars. DICE (NT SYS-pc) generated a
distance matrix data of virulence data that was used in the FITCH
program (PHYLIP) to estimate phylogeny. A. Binary identification
(Pastor-Corrales, 1991) of C. lindemuthr‘anum races. B. Origin of
isolates: Arg (Argentina), Bra (Brazil), Clb (Colombia), CR (Costa
Rica), DR (Dominican Republic), Hon (Honduras), Per (Peru), Mex
(Mexico), and US (United States); Div (races identified from
isolates collected rn different countries). C. Middle American (M)
and Andean (A) race groups according to host source. . ...

Phenogram of C. lindemuthianum races based on RAPD analysis.
Jaccard’s (NT SYS-pc) generated a distance matrix data. FITCH
program (PHYLIP) version 3.50 estimated phylogeny fi'om the
distance matrix data using the Fitch-Margoliash method. A Binary
identification of C. lindemuthianum races (Pastor-Corrales, 1991);
B. Origin of races: Bra (Brazil), Can (Canada), Clb (Colombia), CR
(Costa Rica), DR (Dominican Republic), Gua (Guatemala), Hon
(Honduras), Per (Peru), Mex (Mexico), and US (United States); C.
Middle American (M) and Andean (A) groups of races according to
host source. "' Isolate obtained fi'om the Michigan State University /
Dry Bean Breeding and Genetics Lab. ” Isolate obtained from
Centro Intemacional de Agricultura Tropical - Bean Pathology

Program. ..................................................................

Randomly amplified polymorphic DNA (RAPD) amplicons
obtained with primer 17 of Operon kit S for 24 single-spore isolates
of Colletotrr‘chum lindemuthianum: a- 2 (Peru), b- 3 (Peru), 0- 5 -
(Peru), d- 7 (Peru), e- 38 (Dominican Republic), f- 55 (Dominican
Republic), g- 15 (Colombia), h- 102 (Brazil), i— 130 (United States),
j- 23 (Brazil), k- 87 (Brazil), I- 31 (Brazil), m- 17 (Brazil), n- 65
(Brazil), 0- 81 (Brazil), p- 89(Brazil), q- 453 (Mexico), r- 73
(Mexico), s- 449 (Mexico), t- 457 (Mexico), u- 1673 (Honduras), v-
1993 (Honduras), x- 201 (Honduras), y- 2047 (Costa Rica). Races
were grouped according Andean and Middle American reaction.

43,44

45,46

47

Figure 1.4

Figure 2.1

Figure 2.2

Figure 2.3

Figure 2.4

Figure 2.5

Randomly amplified polymorphic DNA (RAPD) patterns obtained
with primer 13 of Operon kit F for 19 single-spore isolates from races
73 of Colletotrichum lindemuthianum collected in Honduras (lanes a
to d), Mexico (lanes e and f), and the United States (lanes g to s) ......

PCR amplification products of the ITSl, 5.8S rDNA, and ITSZ
region of the rDNA of selected isolates of C. lindemuthianum. Lanes
2 to 7 show non-digested ITS fragments of races 7, 8, 31, 65, 1993,
and the outgroup species C. lagenarium; Lanes 7 to 12 show the
same races digested with the restriction enzyme Msp I; Lanes 14 to
19 show the same races digested with the restriction enzyme Hae III.
Lanes 1 and 20 contain the 1.5kb DNA molecular weight marker ......

PCR amplification products of the ITSl, 5.8S rDNA, and ITSZ
region of the rDNA of C. lindemuthianum isolates digested with the
restriction enzyme Msp I, fiom races 73 (lanes a to d), 65 (lanes e and
t), 23 (lanes g and h), 130 (lanes i and j), 31 (lanes k and l),

17 (lanes m to o), 7 (lanes p to s). .......................................

Sequence alignment (5’ - 3’ direction) of the rDNA internal
transcribed spacer (ITS) sequences and the 5.88 rRNA gene of
isolates of C. lindemuthianum collected in Arg (Argentina), Bra
(Brazil), Clb (Colombia), CR (Costa Rica), DR (Dominican
Republic), Mex (Mexico), Per (Peru), and US (United States). The
sequence Clag (Colletotrichum Iagenariun) was used as outgroup in
both the phylogenetic and parsimony analysis (-) indicates identity
with Arg81 sequence; (*) indicates an introduced gap. The sequences
of the ITS I (positions 1 - 165), the 5.88 rRNA gene (positions 166 -
329), and ITS II (positions 330 - 492) are indicated. ..................

Phylogenetic tree indicating the relationships between isolates of C.
Iindemuthianum based on sequences of ITS], ITSZ and 5.8 rDNA
The tree was created using the neighbor-joining method
(NEIGHBOR program in PHYLIP) from distance values estimated
by the Jukes-Cantor one-parameter method (DNADIST program in
PHYLIP). Confidence limits of the branches, indicated above the
line, were created in a bootstrap analysis using 500 replications. Only
the bootstrap values above 60% are indicated. C. lagenarr‘um was
used as outgroup. ...........................................................

Phylogenetic tree indicating the relationships between isolates of C.
lindemuthianum based on sequences of ITS], ITSZ and 5.8 rDNA.
The tree was created using the Fitch-Margolish method (FITCH
program in PHYLIP) from distance vahres estimated by the Jukes-
Cantor one-parameter method (DNADIST program in PHYLIP). C.
lagenarium was used as an outgroup. ....................................

72

75

78

79

Figure 2.6

Figure 3.1

Figure 3.2

Figure 3.3

Figure 3.4

Phylogenetic tree indicating the relationships between isolates of

C. lindemurhianum. The tree was constructed by parsimony

analysis of ITSl, ITS2 and 5.8 rDNA sequences. Confidence

limits of the branches, indicated above the line, were created in a

bootstrap analysis using 500 replications. Only the bootstrap

values above 60% are indicated. C. lagenarr‘um was used

as outgroup. ............................................................ 81

Three-dimensional representation of the variance of severity data

from inoculation of 34 races of C. Iindemuthianum on 62

germplasm of P. vulgaris associated with the first three principal
components. The X-, Y- and Z-axis are the first, second and third

principal components, respectively. Germplasm was grouped

within Clusterk 1 to 13; B: 14 to 19; C: 20 to 36; and

D: 37 to 62 ................................................................. 103

Three-dimensional representation of the variance of severity data
from inoculation of 34 races of C. lindemuthr‘anum on 62
genotypes of P. vulgaris associated with the first three principal
components The X-, Y- and Z-axis are the first, mond and third
principal components, respectively. Origin of races: Bra (Brazil),
Clb (Colombia), CR (Costa Rica), DR (Dominican Republic),
Hon (Honduras), Mex (Mexico), Per (Peru), US (United States).
Races were grouped within clusters A: 1673Hon, 521Hon, lBra,
81Bra, 9Hon, 89Bra, 73US, 457Mex, 257Mex, 449Mex, 321Mex,
65US, 1993Hon; B: 453Bra, 17Bra, 337Bra, 357Mex, 2047CR,
31Bra, 15Clb, 23US, 87Bra; C: 8Per, 5Per, 3Per, 2Mex, 130US,
102US, 7US, 55DR, 19DR; and D: 39DR, 38DR, 47DR. 106

Phenogram of 62 genotypes of P. vulgaris based on virulence

data obtained from the inoculation of 34 races of C.

Iindemuthianum. The NEI72 coemcient (SIMGEN - NTSYS-pc)

generated a genetic similarity matrix of virulence data. SAHN

program (NTSYS-pc) estimated the genetic distances

using UPGMA ......................................................... 108,109

Phenogram of 34 C. lindemuthr‘anum races based on virulence

data obtained from the inoculation on 62 genotypes of P.

vulgaris. The NEI72 coefficient in the (SIMGEN - NTSYS-pc)

generated a genetic similarity matrix of virulence data. SAHN

program (NTSYS-pc) estimated the garetic distances using

UPGMA The binary identification of C. Ir'ndemuthianum races
(Pastor-Corrales, 1991) was followed by the origin of isolates:

Bra (Brazil), Clb (Colombia), CR (Costa Rica), DR (Dominican

Republic), Hon (Honduras), Per (Peru), Mex (Mexico), and

US (United States). .................................................... 110,111

Figure C.1 Randomly amplified polymorphic DNA (RAPD) amplicons
obtained with primer 2 of Operon kit G for 24 single-spore
isolates of Colletorrichum lindemuthianum: a- 2 (Peru), b- 3
(Peru), c- 5 - (Peru), d- 7 (Peru), e- 38 (Dominican Republic), f-
55 (Dominican Republic), g- 15 (Colombia), h- 102 (Brazil), i-
130 (United States), j- 23 (Brazil), k- 87 (Brazil), I- 31 (Brazil),
m- 17 (Brazil), n- 65 (Brazil), 0- 81 (Brazil), p- 89(Brazil), q- 453
(Mexico), r- 73 (Mexico), SF 449 (Mexico), t- 457 (Mexico), u-
1673 (Honduras), v- 1993 (Honduras), x- 201 (Honduras), y-
2047 (Costa Rica). The polymorphic amplicon indicated by an
arrow distinguished among Andean and Middle American races.
The construction of a SCAR PCR-based marker was attempted
isolating this band from a 0.005% EtBr-stained agarose gel under
UV light after electrophoresis. The isolated PCR product was
cloned into Escherichia coli competent cells pCR 2.1 vector
(Invitrogen Co., Carlsbad, CA). The fragment was sequenced
using the fluorescent dye dideoxy nucleoside triphosphate
terminator method (MSU-DNA Sequencing Facility, East
Lansing MI). Sequencing reactions were nm on a polyacrylamide
gel using the ABI 373A DNA Sequencer. Primers were
constructed considering the primer 2 of Operon kit G as starting
sequence. No polymorphism was observed using 22-, 24 and 26-
base primers. ............................................................ 133

GENERAL INTRODUCTION

Anthracnose is caused by the anamorphic fungus Colletotrichum Iindemuthianum
(Sacc. & Magnus) Lams. -Scrib. The teleomorphic phase, Glomerella lindemuthr‘ana
(Shear), has not been reported in field conditions. It is an important disease in common
bean worldwide responsible for yield losses as high as 95% (Guzman et a1, 1979).
Infected seeds serve as a primary source of inocuhrm and optimum conditions for disease
development include: early infection on susceptible cultivars associated with high
humidity and fi'equent precipitation (Zaumeyer and Thomas, 1957; Guzman et a1, 1979).
Anthracnose is an important disease in sub-tropical and temperate regions due to eficient
seed transmission (Tu, 1992), and the lack of cost-effective chemical control methods
(Pastor-Corrales and Tu, 1989). In addition, the ability of C. lindemuthianum to survive
in plant debris up to 22 months (Dillard and Cobb, 1993), and the occasional
development of sclerotia (Sutton, 1992), limits the efi‘ective control of this disease.

Anthracnose control is best achieved through an integrated pest management
approach (IPM) based on exclusion and eradication measures (Pastor-Corrales and Tu,
1989; Tu, 1988; Zaumeyer and Thomas, 1959). Exclusion of C. Iindemuthianum can be
accomplished by quarantine, establishment of tolerance levels for infected seeds, and
selection of disease the seed production areas. Eradication of C. lindemuthianum can be
achieved by seed treatment, cultural measures, sanitation procedures, firngicides

spraying, and resistant cultivars. In many countries the use of [PM for anthracnose

2

control is difi'rcult to implement. Traditionally, host resistance has been the most
appropriate control for anthracnose. However, the high variability in C. lindemuthianum
(Pastor-Corrales and Tu, 1989) has resulted in continuous breakdown of resistance in
commercial germplasm. Unfortrmately, extensive information about variability in C.
lindemuthianum has been collected at the local level with no standardized procedure. The
recommendation to use an international differential series and a rmique race designation
system was made at the First Workshop on Anthracnose of Common Bean in Latin
America (Pastor-Corrales, 1988). Since then, a standardized method has been used to

characterize the variability in C. lindemuthianum.

Traditional characterization Barrus (191 l, 1918) first described variability in C.
lindemuthianum. Races alpha and beta were identified based on reaction of 139 bean
cultivars. Isolates from different geographic origins showed different virulence
phenotypes. Resistant and susceptible cultivars accounted for 6% and 70% of all
germplasm, respectively. Burkholder (1923) identified race gamma from one isolate
virulent to the cultivar White Imperial, which was reported previously as resistant by
Barrus (1918). Andrus and Wade (1942) identified race delta from isolates collected in
North Carolina. In Chile, Mujica (1952) reported races alpha, beta and gamma. Blondet
(1963 ), Bannerot (1965), and Charrier and Bannerot (1970) reported presarce of races
alpha, beta, gamma, delta, epsilon and lambda, in France. Leakey and Simbwa-Bunnya
(1972) reported the same group of races in Uganda. Fouilloux (1975), Schnock et a1
(1979), and Hubbeling (1977) described races alpha-Brazil, iota, and ebnet virulent to the

Co-2 gene. Later, the race ebnet was designated as race kappa (Kruger et a1., 1977).

3
Hubbeling (1976) reported the race alpha-nurtant, later designated lambda. F ouilloux

(1979) identified the race lambda-mutant. Tu et a1 (1984) and Tu (1988, 1994) reported
races epsilon, delta, lambda, and alpha-Brazil in Canada. In Brazil, races alpha, Mexique
II, delta (Kimati, 1966), alpha, delta, epsilon, lambda, teta, eta, mu, zeta (Menezes and
Dianese, 1988), and alpha-Brazil (Balardin et al., 1990) were identified.

Many research groups have used local cultivars and difi‘erent codes to describe the
variability in C. Iindemuthianum. In Germany, Peurer (1931) identified A-E, G-N, and X
races, which were later reported by Schreiber ( 1932) as equivalent to races alpha, beta
and gamma, respectively. In Australia, Waterhourse (1955) and Cruikshank (1966)
identified races 1 to 8, and 1 to 3. No equivalence of the Australian races to any known
race has been reported. In Mexico, Yerkes and Teliz-Ortiz (1956) identified groups Ito
III Yerkes (195 8) reported races MA-11 to MA-13 equivalent to the alpha race. Noyola
et a1. (1984) designated races MA-21 and MA-22 equivalent to the alpha race. Garrido
(1986) related races MA-20 to MA-25 with alpha, and MA-26 to MA-30 with Mexico I.
In France, Bannerot (1965) described races PV6, D10, F86, I4, 1 and 5, which might
correspond to alpha, beta, gamma, delta, epsilon, and a combination of gamma and delta
races, respectively. In Brazil, Oliari et a1. (1973) characterized groups Brazilian 1 and II.
Pio-Ribeiro and Chaves (197 5) reported races BA-l to BA-10, and suggested equivalence
to races alpha, delta, Mexico I, and Mexico I], and groups Brazilian I and Brazilian 1].
Schwartz et al (1982) reported race C-236 identified fi'om an isolate collected in
Guatemala. Despite tentative equivalence to known races, data collected using local hosts
limited the knowledge of variability in C. lindemuthianum.

Standardized methodologies for race identification are recognized as

4

indispensable to a broader understanding of structure in the C. Iindemuthianum
population. Utilization of an international differential series was wggested (Paaor-
Corrales, 1988, 1991; Drijflrout and Davis, 1989). The differential series proposed by
Pastor-Corrales (1991) has revealed a wide variability in C. Iindemuthianum. In the
United States, races 7, 64, 65 and 73 were identified (Kelly et al., 1994;Ba1ardin and
Kelly, 1996). In Mexico, Garrido (1986) and Rodriguez-Guerra (1991) reported 32 races.
In Nicaragua, Rava et al. (1993) identified 9 races from 10 isolates. In Colombia, 33
races were characterized from a group of 17 8 isolates (Restrepo, 1994).

Since Barrus (1911), breeding programs have used C. Iindemuthianum virulence
phenotypes to screen resistant individuals. However, information related to population
structure of this pathogen has been omitted. Because only a few genes are involved in the
C. Iindemuthianum - P. vulgaris interaction, limited information is obtained from
virulence analyses In addition, virulence is under selection and may respond to
environment changes. Constant monitoring of pathogen variability is needed. Durability
of resistance depends on how efficiently new predominant virulence phenotypes are
detected within a population, and how fast resistance against the newest race is
incorporated into commercial germplasm. Therefore, collecting more extensive
information about structure of variability within C. lindemuthianum is necessary.
Analysis of population structure for making evolutionary inferences using selectively
neutral genetic markers is preferred. There is ofien strong selection for virulence
depending on the environment (Milgroom, 1995). Neutral molecular markers can provide
information on evohrtionary processes such as gene flow, drifi or recombination In

contrast, population structure inferred from selectable markers might reflect the selective

5
pressures operating on a population (Milgroom, 1995; McDonald and McDermott, 1993).

Combining virulence and molecular analyses should lead breeding programs towards

developing durable resistance to anthracnose in common bean.

Molecular characterization. Molecular genetic markers are selectively neutral,
highly informative, reproducible, and relatively easy to assay (McDonald, 1997). Markers
based on restriction fragment length polymorphisms (RFLP-based), polymerase chain
reaction (PCR-based), and isozymes are the most common molecular markers used in
fungal systematics and in population genetics (Michelmore and Hulbert, 1987; Bruns et
al., 1991;Mi1groom, 1995;1-Ii11is et al., 1996). Mapping and sequencing the ribosomal
internal transcribed DNA spacer regions (rDNA-IT‘S) and the rRNA genes (188, 5.88,
288) have been used in analyses of gene evohrtion, and studies on inter- and intraspecific
phylogenetic relationships, and population structure (Bruns et al., 1991; Hillis et al.,
1996).

Molecular analysis of entire genomes can reveal the extent of variability in one
species. RFLPs have been used to determine origin and evolution of individual isolates or
races within asexual populations exhibiting low levels of recombination (Michelmore and
Hulbert, 1987). RFLPs combined with Southem hybridization are more reproducible and
more dificult to conduct than RAPDs They are codominant and exhibit a potentially
unlimited number of alleles per locus, what makes RFLP-based markers advantageous
compared to RAPDs (McDonald, 1997). Large amounts of DNA from each individual
and laborious protocols (cloning, southern blotting, radioactive labeling of probes) are

disadvantages of RFLPs RAPD-PCR technology is an automated method, easy to

6

implement, and can be used to characterize highly variable pathogens, provided a large

bank of random primers are available (Pahrmbi, 1996).

a. RAPDs. Several advantages of RAPDs have been reported. The technique is
fast and simple, independent of gene expression, and able to be transferred between
laboratories (Williams et al., 1990, Palumbi, 1996). Another advantage of RAPDs is the
large number of different products randomly amplified in the genome. A section of every
chromosome generating markers at many loci might be amplified (Palumbi, 1996;
Williams et al, 1990). RAPDs have been used in systematics for differentiation and
grouping of isolates from different fungal species (Crowhurst et al., 1991; Aufauvre-
Brown et al., 1992; Kolmer et a1, 1995). Geographic origin and difl‘erential virulence in
fungi has been corelated to molecular diversity for many species (Megnegneau et al.,
1993; Goodwin and Annis, 1991; Nicholson and Rezanoor, 1994; Kolmer et al., 1995).
Kolmer et al. (1995) observed a high degree of molecular polymorphism among isolates
that had the same virulence phenotype in Puccinia recondita fsp. tritici. The molecular
polymorphism within virulence phenotypes was considered to be a factor that explained
the low correlation between virulence and molecular data. Guthrie et al. (1992) used
RAPDs to identify virulence phenotypes of Colletotrichum graminicola, and suggested
the influence of geographical origin on the variability of this pathogen. Mills et al. (1992)
differentiated and grouped isolates of C. gloearporioides based on RAPDs The effect of
host and geographical origin on pathogen variability was demonstrated.

Variability in C. Iindemuthianum has been characterized by RAPDs. Fabre et al.

(1995) grouped C. lindemuthianum using RAPDs, and RFLP of PCR-amplified internal

7

transcribed spacers (ITS) of rDNA No correlation between DNA polymorphism and the
geographic origin of isolates was observed. Using a similar methodology, Sicard et al
(1997) collected isolates of C. lindemuthianum from wild common bean populations in
Argentina, Ecuador and Mexico. Isolates were divided correspondingly to the P. vulgaris
gene pools observed in each of these centers of diversity. Restrepo (1994) identified and
grouped isolates collected in the Andean and Northern regions of Colombia. Vilarinhos et
al. (1996) characterized isolates from Brazil and suggested the use of RAPDs for fast
monitoring of C. lindemuthianum variability.

Recent reports have pointed out a number of problems associated with assessing
phylogenetic relationships using RAPD markers. RAPDs have several technical
limitations that make them dificult to reproduce (McDonald, 1997). Some of these
limitations can be overcome with replicate DNA preparations, southern analysis, and
conversion of RAPD amplicons into sequence characterized amplified regions -SCARS
(McDonald, 1997). Conditions of amplification are critical for the interpretation of
results, and absence of a product in a particular reaction could be caused by many
genomic differences (Palumbi, 1996). Homologous loci are very difficult to identify
limiting the use of RAPDs in interpopulational comparisons (Smith et al., 1995). In
addition, RAPDs have only two alleles (amplification or nonamplification) at each
amplicon locus. Although this is ideal for genetic mapping, it is a limitation for measures
of genetic diversity affected by the number of alleles at a locus (McDonald, 1997).
Nucleic acid sequencing has been a powerful approach for analysing intrasp ecific
diversity. Sequencing studies of the rRNA genes have proven useful in analysing

evohrtionary divergences (Hillis et al., 1996). The variable regions in the rRNA genes

make them useful for examining relationships within more closely related groups

(Mitchell et a1, 1995; Palumbi, 1996).

b. Internal Transcribed Spacers (ITS). Most fungal phylogenetic studies have
used sequences from the cluster of tandemly repeated rRNA gene. Interspersed between
the highly conserved structural regions coding for the 5.88, 188 and 288 rRNA genes, are
the variable internal transcribed spacer regions — ITS (Mitchell et al., 1995; Palumbi,
1996). PCR amplification of rDNA-ITS regions and the rRNA genes is possible using a
set of universal primers. The variability observed in rDNA-ITS region have been used for
intra- or inter-specific divergence analysis in several fungal species (Nazar et a1, 1991;
Lin and Sinclair, 1992; Kusaba and Tsuge, 1995; Bunting et al., 1996; Cooke and
Drmcan, 1997; Fouly et a1, 1997; Kropp et al., 1997).

Tisserat et al (1994) made selective primers derived fiom ITS sequences of
Ophimphaerella korrae and 0. herpotricha for rapid diagnoses of turfgrass diseases. Liu
and Sinclair (1992) identified five groups of variability in Rhizoctonia solani anastomosis
group 2 isolates using RFLP-IT‘S of the rDNA. They suggested that nnrtation at the ITS
region was the cause of evolution of these groups. Bunting et al. (1996) analyzed the
rDNA ITS-1 region of Magnaporthe spp and observed greater variability within M.
grisea than within M. poe and M. rhizophila isolates

PCR amplification of ITS region of rDNA has been used to characterize
variability in C. gloearporioides (Mills et al., 1992). Sreenivasaprasad et al. (1992)
suggested the use of rDNA-ITS 1 region to assess the molecular variability within and

between isolates of C. acutatwn, C. gloesporioides and C. fi'agariae. Isolates of C.

9

fiagariae were separated into two groups by homology analyses of rDNA-ITS sequenced
regions, while isolates fi'om C. gloesporioides formed a distinct monomorphic group.
Analysis of sequences of the ITSl region of Colletotrichum spp showed C.
gloesporioides, C. musae and C. fiuctigenum clustered along with C. acutatum
(Sreenivasaprasad et a1, 1994) Sherrifi‘ et al (1994) studied the relationships within the
genus Colletotrichum using the PCR amplification of rDNA-IT S regions Their results
showed a cluster formed by C. Iindemuthianum, C. malvarum, C. trifolii, and C.
orbicularae, whereas C. gloesporioide represented a group-species within Colletotrichum
spp. Restriction fragment length polymorphism of both rRNA genes and the ITS-rDNA
showed that C. Iindemuthianum isolates consistently formed two main groups, but some
isolates did not fall distinctly into either group (Fabre et al 1995). The limited number of
isolates used did not allow analysis of whether groupings were aligned according to host
origin or virulence. Sicard et al (1997) grouped isolates of C. lindemuthianum collected
in Mexico, Ecuador and Argentina fiom wild Phaseolus using RFLP-ITS. Adaptation of
races on cultivars of the same geographic origin was suggested because groups
corresponded to three host gene pools.

Combining virulence and molecular analyses has revealed the structure of
variability in C. lindemuthianum. In addition, the efl'ect of geographic origin of pathogen
isolates and host gene pools on variability in C. Iindemuthianum has been demonstrated.
Such coevohrtionary trends should help breeding programs decide on the specific gene
combinations to deploy according to local situations The greater pathogenic diversity of
C. Iindemuthianum in the Central American countries may require control strategies

where resistance genes should be deployed in a manner emphasizing resistance gene

10
pyramiding (Young and Kelly, 1997). In contrast, the limited variability within C.

lindemuthinaum in North America may represent reduced pathogenicity on the

commonly deployed resistance genes, resulting in an increase in the durability of resistant

cultivars.

Relationship between C. lindemuthianum and P. vulgaris variability. Anthracnose
has traditionally been endemic and more severe in Central America than in Andean South
America or in the temperate regions of North America and Europe (Pastor-Corrales et al.,
1994). Since Barrus (1918) had grouped bean cultivars according to the races alpha and
beta, resistance to anthracnose became a major objective in breeding programs
worldwide. The Andean Co-l (A) gene was the first major gene utilized to develop
anthracnose resistant cultivars in common bean (Burkholder, 1918). Prior to 1973, Co-l
gene was used as the only source of resistance in navy beans (Sanilac and Seafarer) in
Michigan and Ontario. Afier appearance of delta race in Ontario in 1978 (Tu, 1988), the
Co-2 gene became the main source of anthracnose resistance in North America.
However, identification of race 73 in Michigan and race alpha-Brazil in Ontario limited
its utilization (Kelly et al., 1994; Tu, 1994) and the pyramiding of Co-l. and Co-2 genes
was suggested as the best protection against all known races in North America (Kelly et
al., 1994; Young and Kelly, 1996).

The Co-2 (Are) gene, characterized in a black bean from Venezuela
(Mastenbroek, 1960), was the predominant source of resistance used worldwide (Kruger
et al., 1977; Fouilloux, 1979). The appearance of races virulent to the Co-2 gene (kappa,

iota, and alpha-Brazil) emphasized the need for new sources of resistance. European and

11

American cultivars were reported as resistance sources against these races (Kruger et al,
1977; Fouilloux, 1979). The resistance genes Mexique I, II and III were characterized in
France from a Mexican germplasm collection (Fouilloux, 1979). The genes Mexique II
and III conferred resistance to the races virulent to the gene Co-Z, whereas Mexique I was
susceptible to the race alpha-Brazil. Screening of 13,000 accessions for resistance to
Latin American and European isolates of C. lindemuthianum showed that Latin American
isolates were more virulent than European isolates (Schwartz et al, 1982). In addition,
only 0.25% of the germplasm was resistant to all isolates. The cultivars TU, AB 136, G
2333, G 2338, G 3991, and G 4032 were resistant to nine races of C. lindemuthianum
fi'om Brazil (Balardin et a1, 1990). G 2333 was resistant to mixtures of isolates from
Middle American and Andean South American countries based on the inoculation of
20,144 accessions in CIAT (Pastor-Corrales et al., 1995).

Resistance based on major genes has been inefl‘ective when resistance genes are
used one at a time (Duvick, 1996). Since monitoring virulence in C. lindemuthianum has
been disseminated among countries, evolving races have continually overcome resistant
germplasm. Sources of resistance widely used in many breeding programs in Europe
(Mastenbroeck, 1960; Goth and Zaumeyer, 1965; Fouilloux, 1979), North America (Tu
and Aylesworth, 1980; Tu, 1988) and South America (Menezes and Dianese, 1988;
Balardin et al., 1990) were overcome by races in difl‘erent countries The cultivars TO, PI
207262, and Mexico 222, reported resistant to European and Latin American isolates
(Kruger et a1, 1977; Fouilloux, 1979; Schwartz et al, 1982), were susceptible to races
identified in Brazil (Menezes and Dianese, 1988). In Mexico, Garrido (1986) and

Rodriguez-Guerra (1991) reported 32 different races virulent to the cultivars Cornell 49-

12
242, Mexico 222, P1207262, TO, TU, and AB 136. In Nicaragua, Rava et al. (1993)

identified isolates that overcame resistance in the cultivars Pl 207262, TO, TU, and AB
136. The most extensive variability of C. lindemuthianum in Latin American countries
might be influenced by greater genetic variability in the local hosts. As observed, major
gene resistance has not been effective. Thus, pyramiding major resistance genes may be
an appropriate breeding strategy for long-term resistance in Phaseolus (Young and Kelly,
1997). Resistance in G 2333 was demonstrated to be conditioned by the genes Co-42, Co-
5, and Co-7 (Young and Kelly, 1997; Young et al., 1997). In addition, knowledge of gene
complementarity has been suggested to improve the efficiency of pyramiding genes for
durable resistance (Duvick, 1996). A resistance gene combination must be deployed until
it is vulnerable only to virulence combinations that do not occur in the pathotype array
(Casela et al., 1996).

Morphological, biochemical and molecular markers (Gepts, 1988; Singh et al.,
1991) have been used to demonstrate the existence of Middle American and South
American Andean gene pools in P. vulgaris. prathogens were specialized on one of the
host gene pools, then transferring resistance genes between gene pools may provide a
more durable resistance (Gepts, 1988; Singh et a1, 1991). In addition, the reduced
virulence of isolates from the Andean region compared to isolates from Middle American
regions suggests that deployment of resistance genes between gene pools will improve
durable resistance (Gepts, 1988; Singh et al., 1991; Young and Kelly, 1997). Pathogen
congruence with the two P. vulgaris gene pools has been demonstrated. Uromyces
phaseoli (Stavely, 1982; Stavely, 1984; Maclean, 1995) and Phaeoisariopsis griseola

(Guzman et al., 1995) appear to be grouped according to P. vulgaris gene pools A

13

previous study of twelve C. lindemuthimum isolates found that RAPD and RFLP
markers divided the species into two groups (Fabre et al, 1995). The groups suggested
some specialization within C. lindemuthian that corresponded to the two gene pools
within P. vulgaris. Pastor-Corrales (1996) reported that C. lindemuthianum can be
divided into two groups based on virulence; one specializing on hosts from the Middle
American gene pool and the second specializing on Andean hosts. Using RFLP-ITS,
Sicard et al. (1997) grouped isolates of C. lindemuthianum collected in Mexico, Ecuador
and Argentina from wild Phaseolus. Two groups of races were observed in Mexico,
corresponding to the Middle American and Andean diversity center. Races from Ecuador
and Argentina both corresponding to the Andean gene pool, were monomorphic. The
information on variability in C. lindemuthianum and the specialization of specific races
on one of the host gene pools obtained from virulence and molecular analyses, may be
valuable in breeding for resistance. The process of pyramiding genes could be improved
through marker assisted selection (MAS) if molecular markers linked tightly to the
resistance genes were available (Kelly et al., 1994). In this case if direct selection for a
phenotype is not practical or feasible due to epistasis, MAS can accelerate the breeding
process and be used to identifiy uncharacterized resistance genes for which no race is

available (Young et al 1997).

Conclusions. Variability in C. lindemuthianum has been characterized worldwide.
Unfortunately local codes and difi‘erent differential series have biased the structrning of
broader pathogen populations. A standard differential series was proposed, but an

imbalance among cultivars from P. vulgaris gene pools might favor identification of

14
races belonging to the Middle American reaction group. The multigenic resistance in

other cultivars such as G 2333 might select races with multiple avirulence genes. Thus,
information regarding the gene-for- gene relationship in the C. lindemuthianum - P.
vulgaris pathosystem would be biased. Developing durable resistance to C.
lindemuthin must be based on reliable characterization of variability in this pathogen.
Neutral molecular markers are less influenced by changes in the environment and have
been more informative in assessing variability. RAPDs can be used to identify, group,
and relate pathogen isolates to geographic or host factors. However, problems related
with RAPDs have indicated the need for markers, which consistently reproduce the
variability within the genome of a pathogen. The analysis of the variable rDNA-ITS
regions have been used for phylogenetic analyses at inter- or intra-species level in
Colletotrichum. Since C. lindemuthianum variability has consistently been characterized,
breeding programs might be able to identify new sources of resistance for their local
needs. Resistance genes could be intelligently pyramided and segregating populations
adequately screened either directly or indirectly using MAS. Deployment of gene
combinations might be planned according to specific geographic regions resulting in a

more durable resistance to C. lindemuthianum.

General objectives. Characterization of variability in C. lindemuthianum has been
considered a crucial step to help control anthracnose using genetic resistance.
Understanding of C. lindemuthianum population structure would assist breeding
programs to select and deploy resistance genes according to specific situations. The
objectives of this work were: a characterization of diversity in C. lindemuthianum fiom

isolates collected in South, Central, and North American countries using virulence and

15

molecular (RAPD) polymorphism; b. sequence homology and RFLP analysis of rDNA-
IT‘S regions of C. lindemuthianum isolates; and, c. genetic analyses of virulence and
resistance in the P. vulgaris — C. lindemuthianum pathosystem The adequacy of
molecular markers in establishing phylogenetic relationships, the correlation of molecular
markers with race structure and geographic structure, the structure of C. lindemuthianum

populations, and the efi‘ect of host diversity on pathogen variability will be discussed.

16

LIST OF REFERENCES

Andrus, C. F ., and Wade, B. L. 1942. The factorial interpretation of anthracnose
resistance in beans. Pages 1-29: in Tech. Bull. No. 310. USDA. Washington.

Aufauvre-Brown, A., Cohen, 1., and Holden, D. W. 1992. Use of randomly amplified
polymorphic DNA markers to distinguish isolates of Aspergillus fimrigatus.
Journal of Clinical Microbiology 30:2991-2993.

Balardin, R., and Kelly, J. D. 1996. Identification of Race 65-epsilon of bean anthracnose
(Colletotrichum lindemuthianum) in Michigan. Plant Dis. 80:712.

Balardin, R. S., Pastor-Corrales, M. A., and Otoya, M. M. 1990. Variabilidade patogénica
de Colletotrichum lindemuthiwmm no Estado de Santa Catarina. Fitopatol. Bras.
15:243-245.

Bannerot, H. 1965. Results de l'infection d'rme collection de haricots par six races
physiologiques d'anthracnose. Annales de Amelioration des Plantes (Paris)
1 5:201-222.

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

Barrus, M. F. 1918. Varietal susceptibility of beans to strains of Colletotrichum
lindemuthianum (Sacc. 8L Magn.) B. & C.. Phytopathology 8:589-605.

Blondet, A. 1963. L'anthracnose du haricot: etude des races physiologiques du
Colletotrichum lindemuthianum. Ph.D. these, Faculté de Sciense, Paris, France.

Bruns, T. D., White, T. J., and Taylor, J. N. 1991. Fungal Molecular systematic. Annual
Review of Ecology and Systematic 22:525-564.

Bunting, T. E., Plumley, K A., Clarke, B. B., and I-Iillman, B. I. 1996. Identification of
Magnaporthe poae by PCR and examination of its relationship to other firngi by
analysis of their nuclear rDNA ITS-1 regions. Phytopathology 86:398-404.

Burkholder, W. H. 1923. The gamma strain of Colletotrichum lindemuthianum (Sacc. et
Magn.) B.et C.. Phytopathology 13:316-323.

Casela, C. R, Ferreira, A. S., Zeller, K. A, and Levy, M. 1995. Pathotype variation in the
sorghum anthracnose fungus: a phylogenetic perspective for resistance breeding.
Pages 257-288 in: Disease analysis through genetics and biotechnology. J. F.
Leslie and R. S. Frederiksen, eds. Iowa State University Press. Ames, IO.

l7

Charrier, A, and Bannerot, H 1970. Contribution i l'étude des races phisiologiques de
l'anthracnose du haricot. Ann. Phytopathol. 2:489-506.

Cooke, D. E. L. and Duncan, J. M. 1997. Phylogenetic analysis of Phytophthora species
baed on IT SI and ITS2 sequences of the ribosomal RNA gene repeat. Mycol. Res.
101:667-677.

Crowhurst, R. N., Hawthorne, B. T., Rikkerink, E. H. A, and Templeton, M. D. 1991.
Differentiation of Fusarium solani fsp. cucurbitae races 1 and 2 by random
amplification of polymorphic DNA Crur. Genet. 20:391-396.

Cruickshank, I. A M. 1966. Strains of Colletotrichwn lindemuthianum (Sacc. and Magn.)
in eastern Australia. J. Aust. Inst. Agric. Sci. 32:134-135.

Driflrout, E., and Davis, J. H C. 1989. Selection of a new set of homogeneously reacting
bean (Phaseolus vulgaris) difi‘erentials to differentiate races of Colletotrichum
lindemuthianum. Plant Pathology 38:391-396.

Dillard, H R., and Cobb, A C. 1993. Sruvival of Colletotrichum lindemuthianum in bean
debris in New York State. Plant Dis 77: 1233-1238.

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

Fabre, J. V., Julien, J., Parisot, D., and Dron, M. 1995. Analysis of diverse isolates of
Colletotrichum lindemuthianum infecting common bean using molecular markers.
Mycol Res. 99:429-435.

Fouilloux, G. 1975. L'anthracnose du haricot: etude des relations entre les pathotypes
anciens et nouveaux; etude de nouvelles sources de resistance “totale”. Pages 81-
92 in: Reunion Eucarpia Haricot, Versailles, France. Centre National de
Recherches Agronomiques, Paris, France.

Fouilloux, G. 1979. New races of bean anthracnose and consequences on our breeding
programs. Pages 221-235 in: Disease of tropical food crops H. Maraite and J.A
Meyer, eds Louvain-la-Neuve, Belgium.

Fouly, H. M., Wilkinson, H. T., and Chen, W. 1997. Restriction analysis of internal
trascribed spacers and the samll subunit gene of ribosomal DNA among four
Gaeumannomyces species Mycologia 89:590-597.

Garrido, E. R. 1986. Identificacion de razas fisiologicas de Colletotrichum
lindemuthianum (Sacc. & Magn.) Scrib. en Mexico y busqueda de resisténcia
gene'tica a este hongo. MSc thesis, Institucion de Ensenanza y Investigacion en
Ciencias Agricolas, Montecillos, Mexico.

l8

Gepts, P. 1988. A Middle American and an Andean common bean gene pool. Pages 375-
390 in: Genetic resources of Phaseolus beans; their maintenance, domestication,
and utilization. P. Gepts, ed. Kluwer. London.

Goodwin, P. H, and Annis, S. L. 1991. Rapid identification of genetic variation and
pathotype of Leptosphaeria maculans by random amplified polymorphic DNA
assay. Applied and Environmental Microbiology 27 :2482-2486.

Goth, R. W., and Zaumeyer, W. J. 1965. Reactions of bean varieties to four races of
anthracnose. Plant Dis Rep. 49:815-818.

Guthrie, P. A 1., Magill, C.W., Frederiksen, R. A, and Odvody, G. N. 1992. Random
amplified polymorphic DNA markers: a system for identifying and differentiating
isolates of Colletotrichwn graminicola. Phytopathology 82:832-835.

Guzman, P., Donado, M. R, and Galvez, G. E. 1979. Perdidas economicas causadas por
la antracnosis del fiijol Phaseolus vulgaris en Colombia. Turrialba 29:65-67.

Guzman, P., Gilbertson, R. L., Nodari, R., Johnson, W. C., Temple, S. R., Mandala, D.,
Mkandawire, A B. C., and Gepts, P. 1995. Characterization of variability in the
fungus Phaeisariopsis griseola suggests coevolution with the common bean
(Phaseolus vulgaris). Phytopathology 85:600-607.

Hillis, D. M., Mable, B. K, Larson, A, Davis, S. K., and Zimmer, E. A 1996. Nucleic
acids IV: sequencing and cloning. Pages 321-381 in: Molecular Systematics.
D.M. Hills, C. Moritz and BK. Mable, eds. Sinauer, Sunderland, MA

Hubbeling, N. 1976. Selection for resistance to anthracnose, particularly in respect to the
"ebnet" race of Colleotrichum lindemuthianum. Ann. Rep. Bean Improv. Coop.
19:49-50.

Hubbeling, N. 1977. The new jota race of Colletotrichum lindemuthianum. Ann. Rep.
Bean Improv. Coop. 20:58.

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

Kimati, H. 1966. Algumas racas fisiologicas de Colletotrichum lindemuthin (Sac. et
Magn.) Scrib. que ocorrem no Estado de Sio Paulo. An. Esc. Super. Agric. Luiz
de Queiroz (Sao Paulo) 23:411-437.

Kolmer, J. A, Liu, J. Q., and Sies, M 1995. Virulence and molecular polymorphism in
Puccinia recondita £sp. tritici in Canada. Phytopathology 85:276-285.

l9

Kropp, B. R, Hansen, D. R, Wolf, P. 6., Flint, K. M., and Thomson, S. V. 1997. A
study on the phylogeny of the Dyer's woad rust fungus and other species of
Puccinia from crucifers. Phytopathology 87:565-571.

Kusaba, M., and Tsuge, T. 1995. Phylogeny of A lternaria fungi known to produce host-
specific toxins on the basis of variation in internal trascribed spacers of ribosomal
DNA Curr Genet 28:491-498.

Kruger, J., Hoflinann, G. M., and Hubbeling, N. 1977. The kappa race of Colletotrichum
lindemuthianum and sources of resistance to anthracnose in Phaseolus beans.
Euphytica 26:23-25.

Kusaba, M, and Tsuge, T. 1995. Phylogeny of Alternaria fungi known to produce host-
specific toxins on the basis of variation in internal trascribed spacers of ribosomal
DNA

Leakey, C. L. A, and Simbwa-Bunnya, M. 1972. Races of Colletotrichum
lindemuthianum and implications for bean breeding in Uganda. Ann. Appl. Biol
70:25-34.

Liu, Z. L., and Sinclair, J. B. 1992. Genetic diversity of Rhizoctonia solani anastomosis
group 2. Phytopathology 82:778-787.

Maclean, D. J., Braithwaite, KS, Irwin, J.AG., Manners, J.M., and Groth, J.V. 1995.
Random Amplified Polymorphic DNA reveals relationships among diverse
genotypes in Australian and American collections of Uromyces appendiculatus.
Phytopathology 85:757-765.

Mastenbroeck, C. 1960. A breeding program for resistance to anthracnose in dry shell
haricot beans, based on a new gene. Euphytica 9: 177-184.

McDonald, B. A, and McDermott, J. M. 1993. Population genetics of plant pathogenic
fungi. BioScience 43:31 1-319.

McDonald, B. A 1997. The population genetics of fungi: tools and techniques.
Phytopathology 87:448-453.

Megnegneau, B., Debets, F., and Hoekstra, R F. 1993. Genetic variability and
relatedness in the complex group of black Aspergilli based on random
amplification of polymorphic DNA Curr Genet 23:323-329.

Menezes, J. R, and Dianese, J. C. 1988. Race characterization of Brasilian isolates of
Colletotrichum lindemuthianum and detection of resistance to anthracnose in
Phaseolus vulgaris. Phytopathology 78:650-655.

20

Michelmore, R W., and Hulbert, S. H 1987. Molecular markers for genetic analysis of
phytopathogenic fungi. Annu. Rev. Phytopathol 25:383-404.

Milgroom, M. G. 1995. Analysis of population structure in firngal plant pathogens. Pages
213-229 in: Disease analysis through genetics and Biotechnology. J. F. Leslie,
and RA Frederiksen, eds. Iowa State University Press. Ames, IO.

Mills, P. R, Sreenivasaprasad, S., and Brown, AE. 1992. Detection and differentiation of
Colletotrichum gloeosporioides isolates using PCR FEMS Microbiology Letters
98: 137- 144.

Mitchell, J. 1., Roberts, P. J., and Moss, S. 1995. Sequence or structure? A short review
on the application of nuclei acid sequence information to fimgal taxonomy.
Mycologist 9:67-75.

Mujica, R F. 1952. Razas fisiologicas y susceptibilidad varietal de los frijoles chilenos a
la antracnosis. Agri. Téc. (Santiago) 12:37-45.

Nazar, R N., Hu, X., Schmidt, J., Culham, D., and Robb, J. 1991. Potential use of PCR-
amplified ribosomal intergenic sequences in the detection and difl‘erentiation of
verticillium wilt pathogens Physiol and Mol Plant Pathol. 39: 1-1 1.

Nicholson, P., and Rezanoor, H N. 1994. The use of random amplified polymorphic
DNA to identify pathotype and detect variation in Pseudocercosporella
hertrichoides. Mycol Res. 98:13-21.

Noyola, I. T. U., Lopez, G. F., and Muruaga, J. 1984. Determinacion de razas fisiologicas
de Colletotrichum lindemuthianum (Sacc. & Magn.)Bri & Cav. Resr’rmenes del
XI Cong. Nac. de Fitopatologia Soc. Mexicana de Fitopatologia, AC. San Luis
Potosr’, Mexico.

Oliari, L., Vieira, C., and Wilkinson, R E. 1973. Physiologic races of Colletotrichum
lindemuthianum in the state of Minas Gerais, Brazil. Plant Dis. Rep. 57:870-872.

Palumbi, S. R 1996. Nucleic acids [1: the polymerase chain reaction. Pages 205-247 in:
Molecular Systematics. D. M Hills, C. Moritz, and B. K Mable, eds Sinauer,
Sunderland, MA

Pastor-Corrales, M. A (ed. ). 1988. La antracnosis del frijol comum, Phaseolus vulgaris,
en America Latina. Documento de Trabajo no. 113. CIAT. Cali, Colombia.

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

21

Pastor-Corrales, M A 1996. Traditional and molecular confirmation of the coevolution
of beans and pathogens in Latin America. Ann. Rep. Bean Improv. Coop. 39:46-
47.

Pastor-Corrales, M A, and Tu, J. C. 1989. Anthracnose. Pages 77-104 in: Bean
production problems in the tropics. H F. Schwartz and M. A Pastor-Corrales,
eds CIAT. Cali, Colombia.

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

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

Peurer, H 1931. Continued investigation on the occurrence of biological strains in
Colletotrichum lindemuthianum (Sacc. & Magn.) Bri. et Cav. Phytopathologische
Zaeitshrit 4:83-1 12.

Pio-Ribeiro, G., and Chaves, G. M. 1975. Racas fisiologicas de Colletotrichum
lindemuthianum (Sacc. et Magn.) Scrib. que ocorrem em alguns municipios de
Minas Gerais, Espirito Santo e Rio de Janeiro. Experientiae 19:95: 1 18.

Rava, C. A, Molina, J., Kaufiinann, M., and Briones, I. 1993. Determinacion de razas
fisiologicas de Colletotrichum lindemuthianum en Nicaragua. Fitopatol. Bras.
18:388-391.

Restrepo, S. 1994. DNA polymorphism and virulence variation of Colletotrichum
lindemuthianum in Colombia. MSc. thesis Universite Paris IV, Paris-Grignon,
France.

Rodriguez, R 1991. Identificacion de razas patogenicas de Colletotrichum
lindemuthin (Sacc. y Magn.) Scrib. en el estado de Durango mediante un
sistema propuesto intemacionalmente y respuesta de genotipos de frijol tolerantes
a sequia a razas del patogeno. M.Sc., Parasitologia Agricola, Universidad
Autonoma Agraria "Antonio Narro", Buenavista, Mexico.

Schnock, M G., Hofi‘mann, G. M., and Kruger, J. 1979. A new physiological strain of
Colletotrichwn lindemuthianum infecting Phaseolus vulgaris L. HortScience
10:140.

Schreiber, F. 1932. Resistenzzuchtung bei Phaseolus vulgaris. Phytopathol. Z. 4:415-
454.

22

Schwartz, H F., Pastor-Corrales, M A, and Singh, S. P. 1982. New sources of resistance
to anthracnose and angular leaf spot of beans (Phaseolus vulgaris). Euphytica
3 1:74 1-7 54.

Sicard, D., Michalakis, Y., Dron, M., and Neema, C. 1997. Genetic diversity and
pathogenic variation of Colletotrichum 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.

Sherrifi‘, C., Whelan, M. J., Arnold, G. M., Lafay, J. F., Brygoo, Y., and Bailey, J. A
1994. Ribosomal DNA sequence analysis reveals new species groupings in the
genus Colletotrichwn. Experimental Mycology 18:121-138.

Smith, J. J., Scott-Craig, J. S., Leadbetter, J. R, Bush, G. L., Roberts, D. L., and
Fulbright, D. W. 1995. Characterization of random amplified polymorphic DNA
(RAPD) products from Xanthomonas campestris: implications for the use of
RAPD products in phylogenetic analysis. Mol. Phylogenet. Evol 3: 135-145.

Sreenivasaprasad, S., Brown, A E., and Mills, P. R 1992. DNA sequence variation and
interrelationships among Colletotrichum species causing strawberry anthracnose.
Physiol. and Mol. Plant Pathol 41:265-281.

Sreenivasaprasad, S., Mills, P. R, and Brown, A E. 1994. Nucleotide sequence of the
rDNA spacer 1 enables identification of isolates of Colletotrichum as C.
acutatum. Mycol. Res. 98:186-188.

Stavely, J. R 1982. The potential for controlling bean rust by host resistance. Ann. Rep.
Bean Improv. Coop. 25:28-30.

Stavely, J. R 1984. Pathogenic specialization in Uromyces phaseoli in the United States
and rust resistance in beans. Plant Dis. 68:95-99.

Sutton, B. C. 1992. The genus Glomerella and its anamorph Colletotrichum. Pages 1-26
in: Colletotrichum-Biology, Pathology and Control. J.A Bailley and M.J. Jeger,
eds. CAB. International, Wallingford, UK

Tisserat, N. A, Hulbert, S. H, and Sauer, K M. 1994. Selective amplification of rDNA
internal transcribed spacer regions to detect Ophiosphaerella korrae and 0.
herpotricha. Phytopathology 84:478-482.

Tu, J. C. 1988. Control of bean anthracnose caused by delta and lambda races of
Colletotrichunr lindemuthianum in Canada. Plant Dis. 72:5-7.

23

Tu, J. C. 1992. Colletotrichum lindemuthianum on bean. Population dynamics of the
pathogen and breeding for resistance. Pages 203-224 in: Colletotrichum-Biology,
Pathology and Control J. A Bailley and M. J. Jeger, eds. C.AB. International,
Wallingford, UK

Tu, J. C. 1994. Occurrence and characterization of the alpha-Brazil race of bean
anthracnose (Colletotrichum lindemuthianum) in Ontario. Can. J. Plant Pathol.
16: 129- 13 1.

Tu, J. C., and Aylesworth, J. W. 1980. An efl‘ective method for screening white (pea)
bean seedlings (Phaseolus vulgaris L.) for resistance to Colletotrichum
lindemuthianum. Phytopathol Z. 99: 13 1-137.

Tu, J. C., Sheppard, J. W., and Laidlaw, D. M. 1984. Occurrence and characterization of
the epsilon race of bean anthracnose in Ontario. Plant Dis. 68:69-70.

Vilarinhos, A D., Paula Jr., T. J., de Barros, E. G., and Moreira, M. A 1995.
Characterization of races of Colletotrichum lindemuthianum by the random
amplified polymorphic DNA technique. Fitopatol Bras. 20: 194-198.

Waterhouse, W. L. 1955. Studies of bean anthracnose in Australia. Proc. Linn. Soc.
N.S.W. 80:71-83.

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

Yerkes, W. D. Jr. 195 8. Additional new races of Colletotrichwn lindemuthianum in
Mexico. Plant Dis. Rep. 42:329.

Yerkes, W. D., and Teliz-Ortiz, M. 1956. New races of Colletotrichum lindemuthianum
in Mexico. Phytopathology 46:564-567.

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

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.

Yormg, R A, Melotto, M., Nodari, R O., and Kelly, J. D. 1997. Marker assisted
dissection of oligogenic anthracnose resistance in the common bean cultivar,
G2333. Theor. Appl Genet. (in press).

Zaumeyer, W. J., and Thomas, H R 1957. A monographic study of bean diseases and
methods for their control USDA Agr. Tech. Bull. No. 868. 255p.

Chapter 1
VIRULENCE AND MOLECULAR DIVERSITY IN COLLETOTRICHUM
LHVDEMUTHIAN UM FROM SOUTH, CENTRAL AND NORTH AMERICA

ABSTRACT

One hundred thirty-eight isolates of Colletotrichum lindemuthianum from
Argentina, Brazil, the Dominican Republic, Honduras, Mexico and the United States
were characterized into 41 races based on virulence to twelve differential cultivars of
Phaseolus vulgaris. These 41 races were categorized into two groups, those found over a
wide geographic area and those restricted to a single cormtry. Races 7, 65, and 73 were
widespread. Race 73 was the most common (28%). Race 7 was found once in Argentina,
and Mexico, but at higher frequency in the United States. Race 65 was found repeatedly
in Brazil and the United States. Although 39% of the races were detected multiple times
and three races were widespread, no race was isolated from both P. vulgaris gene pools.
Phenetic analyses showed no obvious patterns correlated with virulence clusters. No
geographic patterning was evident. Molecular polymorphism revealed by random
amplified polymorphic DNA (RAPD) confirmed the extensive variability of C.
lindemuthianum. Virulence phenotypes were grouped into 15 clusters. The two largest
clusters contained isolates from all geographic regions sampled. Molecular
polymorphism was observed among isolates from races 65, and 73 within and among

countries, except among Brazilian isolates of race 65. Genetic diversity of C.

24

25

lindemuthianum was shown to be the greatest in Mexico and Honduras. Our data suggest

that C. lindemuthianum may not be highly structured to specific Phaseolus gene pools.

 

26
INTRODUCTION

Colletotrichum lindemuthianum (Sacc. & Magnus) Lams. —Scrib. causes
anthracnose of common bean (Phaseon vulgaris L. ). The disease is found worldwide
wherever common beans are grown, but is especially important in sub-tropical and
temperate region (Pastor-Corrales and Tu, 1989). Yield losses can be as high as 90% or
more in some years (Zaumeyer and Thomas, 1957). Because cost-effective chemical
control is lacking, host resistance has traditionally been used to control anthracnose,
despite the high variability in C. lindemuthianum (Andrus and Wade, 1942; Barrus, 1918;
Blondet, 1963; Burkholder, 1923; Fouilloux, 1979; Kruger et a1, 1977 ; Menezes and
Dianese, 1988; Waterhouse, 1955). A standardized binary nomenclature system based on
a set of twelve differential cultivars has been developed to characterize virulence (Pastor-
Corrales, 1991). Using this system 38 races were reported in Mexico (Garrido, 1986;
Rodrigues-Guerra, 1991); 7 races were identified in a group of 10 isolates fi'om
Nicaragua (Rava et al., 1993); 33 races were characterized from a group of 178 isolates
from Colombia (Restrepo, 1994), and 3 races were described in the United States
(Balardin and Kelly, 1996; Kelly et al, 1994) Since only a few races are extremely
widespread, it is not presently known if the virulence pattern of C. lindemuthianum is due
to repeated evolution of a race or efficient seed-bome dispersal between areas (Pastor-
Corrales and Tu, 1989).

Virulence in common bean pathogens, Uromyces appendiculatus (Maclean et al.,
1995) and Phaeoisariopsis griseola (Guzman et al., 1995), appears to be highly

structured within each species. Patterns generated using randomly anrplified polymorphic

27
(RAPD) DNA markers indicated that U. appendiculatus was highly structured

geographically (Maclean et al., 1995). RAPD patterns divided P. griseola isolates into
two distinct groups; one group was specialized on P. vulgaris hosts of Andean ancestry,
while the second group was obtained fiom common bean hosts of Middle American
origin (Gepts, 1988, Singh et al., 1991). This grouping of fungal races congruently with
host gene pools has been interpreted as evidence for coevolution (Guzman et al., 1995;
Pastor-Corrales, 1996; Singh et aL, 1991). It also suggests combining complementary
genes to exclude all pathogen lineages (Casela et al., 1995; Duvick, 1996; Levy et al,
1993; Young and Kelly, 1997) where resistance genes from Middle American hosts can
be utilized efl‘ectively in Andean hosts and vice versa. Indeed, virulence within common
bean rust (Uromyces phaseoli) has been shown to be specialized onto either Andean or
Middle American germplasm(Mac1ean et al, 1995; Stavely, 1982; Stavely, 1984),
leading some authors to advocate lineage based breeding to control this disease (Gepts,
1988) and other common bean pathogens, including C. lindemuthianum (Singh et al,
1991; Young and Kelly, 1997).

A study that utilized RAPD markers, restriction fragment length polymorphism of
both nuclear ribosomal genes and the internal transcribed spacer region of the ribosomal
subunit found that C. lindemuthianum isolates consistently formed two main groups, but
some isolates did not fill distinctly into one or the other group (Fabre et al, 1995). The
limited number of isolates used did not allow analysis of whether groupings were aligned
according to origin of host or virulence. A similar study conducted with isolates of C.
lindemuthianum collected from wild Phaseolus hosts in Mexico, Ecuador and Argentina

showed groupings of isolates corresponding to Andean and Middle American host gene

28
pools (Sicard et al., 1997). In this study we sought to determine how variability within C.

lindemuthianum is structured. Isolates were collected fi'om both Andean and Middle
American Phaseolus genotypes grown in Argentina, Brazil, Canada, the Dominican
Republic, Honduras, Mexico and the United States. Virulence analyses and RAPD

polymorphism were used separately to group isolates using cluster analyses.

MATERIALS AND METHODS

Colletotrichum lindemuthianum isolates. One hundred thirty-eight conidial
isolates were collected between 1992 and 1994 from common bean cultivars grown in
Argentina, Brazil, the Dominican Republic, Honduras, Mexico and the United States
(Table 1.1, Appendix A1. 1). The fungus was isolated from diseased leaves or pods
showing characteristic anthracnose symptoms (Pastor-Corrales and Tu, 1989). Small
pieces of infected tissue were surfice-sterilized and incubated on petri dishes containing
modified Mathur’s culture medium The culture medium was prepared with dextrose (8
g1"), MgSO4.7H20 (2.5 g1"), KHzPO4 (2.7 g1"), neopeptone (2.4 g.l'1), yeast extract
(2.0 g1"), and agar (16 gl"). Plant tissue was incubated in complete darkness at 24°C for
7 days or until formation of acervuli morphologically resembling C. lindemuthianum
(Pastor-Corrales and Tu, 1989). Spore suspensions for seedling inoculation were prepared
fiom purified single-conidial isolates by flooding plates with 5 m1 of 0.01% Tween 80 in
distilled water. Spores were dislodged by scraping the culture surface with a spatula,
filtering through cheesecloth, and the concentration was adjusted to 1.0 x 106 sporesml'1

with a hemocytometer.

29

Determination of virulence phenotypes. Races of C. lindemuthianum were
characterized based on virulence of the 138 isolates on the differential P. vulgaris series
proposed by Pastor-Corrales (1991). Seeds for each of the twelve cultivars were planted
in flats containing Baccto planting mix (Michigan Peat Co., Houston TX), and grown
rmder greenhouse conditions (16-h day length at 25°C), for 7 to 10 days until seedlings
had reached the primary leaf stage. Six to 10 seedlings were spray-inoculated with
standardized spore suspensions of each isolate of C. lindemuthianum. Suspensions were
applied until runofi‘ on the stem and to both surfaces of the unifoliolate leaves. After
inoculation, plants were maintained in high humidity (>95%) for 48 h at 22 to 25°C.
Plants were allowed to dry and were then transferred to greenhouse benches for 5 days
before disease symptom evaluation. Seven days after inoculation, seedlings were rated
for disease reaction based on a 1 to 9 severity scale (Balardin et al, 1990). Disease
reactions were recorded as resistant (grade 1 to 3) for those plants with no visible disease
symptoms or only a few, very small lesions mostly on primary leaf veins. Plants with
numerous enlarged lesions or with sunken cankers on the lower sides of leaves or
hypocotyls were recorded as susceptible (grade 4 to 9). lnoculations were repeated at
least twice for each isolate. The identified races were assigned a value using the binary
nomenclature system (Pastor-Corrales, 1991). Each differential cultivar has an assigned
number (2") where n corresponded to the place occupied by the cultivar within the
difierentfil series (Appendix A1.2). The designation of a race number was obtained by
summing the 2" vahres of all cultivars exhibiting susceptible reaction to the isolate being

inoculated.

30

 

.822va

 

....ANEVSZ
...?EE:
£3215 - - - - - + - - + - - + m.
.323:
vegan - - - - - + - - - - - + no
2395 - - - - - - + + - + + + mm
ciao - - - - - - - + - - + + a
.3353 - - - - - - - + - - - + 2
5235 .. - - - - - - - + .. .. + a
*3<va
.23 52 .3452 - - - - - - - - - + + + s
3932 - - - - - - - - - - + - N
432:5 - - - - - - - - - - - + _
334.319 36 new 16 T6 «.5 New
9838. a M a _ m e L m a b ... < .8:—

 

Puoflow ooflaumah g6: GHQ uhl>m==nv ima—uouosa

 

sown—See won: 25 3 56% £25

25w 53224 v5 quote—5‘ 0822 Sea mac: “388.4% .8 380:3 333 ”2 Sea 33:53 geicassnc: O we moose :4
me 352%: .- can 033388 .i ESE—3 330.5% 53 5888 ,«e genes“: 28 Joan—8m me 8:85: £38363 35— - A ._ 033—.

31

 

...:zvuom
: ..-vsz
:28:
22332
:23:

2:282

29?:

.8232
$2on

.32on

22:5

.3552
:2qu

‘22on

.32on2
:23:

.3225:
:2on
«£235

:2va

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

++++++++++

+++++

++++++++

+++-

+

++++

+

nae
n3
nun
a3.
new
raw
mmv
a:

hmm
ran
an
cum
has
emu
EN
EN
n3

an

5

9.88v 2 2.3

.885 RES— 3 880.38 ..
805803 «e: 099.com “me: .
goon 08% 8383‘ 2: 82m 83:30 8
632268. 3358 .3 vow—238 8:.
3338 me $8388 2a meta. 35802 2525a we: 8. 682. $88.88“. an? ion 25w 2:. $5850 05 8 338m 23 38 2:
883.8 cox—8.8 2: Ea A855 0:622 - 2 88 8383 - 3 .een 25m we: 05 .3 385823 08 8 832.8 .3025 @0885
m: as .8821»: .8585 8m .385 50888 an .835 am Agassi m2 ”menace .88 029880 .
A82 .8 a 88> 62: i=8 2a 886 .568 as 88» 85.8.2 8 .565. .58 . 22
9 a as .32 - on as o. .3: - P: a .38 - ob _ .32 - 888 E = .3 - NNN 88.25 .am - 82.3 m .§ - 3:25
m .3 - ~83 8:88 a .c. - >852 80% o .3 - >268 Be 85 5836 m .2 - QEBOEV < ”:2: 81:8 52.:
03.3 .038; 038258 805 .3 33e=eu 848833833: .D me 808 0856820 8 some 82530 3898.56 885 8.88.5 ...

32

A33 mac—«boonsmamv 80% 23268082. 3.83 3 mam—:83 83833802 35— .

 

«A3225: - + + + + + - - + - - + nag
£35.53 .. + + + + u - u + - - + ana—
aczveem - + + - + + - - + + - + :5
228.5 - + + - + - - - + + - + Fm:
Layne: - + + - + .. - - + - - + new—
Szveem - + + - - + .. - - - .. + 33
:23: - + + - - + - - - - - - 88
2283.— - + + - - - - - + - - + 9.3
«£2382 - + - + + + - - - - - - Nb:
32102 - + - + + - - + - + + + 3.3
£2382 - + - + - + - - - - - - 3.2
...C.o..vx02 - + .. .. + - - - + + . + $3

 

33
Determination of molecular phenotypes. Sixty single-conidial isolates of C.

lindemuthianum were characterized using RAPD marker data. These included 28 races
selected as representative of the full range of virulence phenotypes (Table 1.1), plus 32
isolates previously reported in Brazil, Canada, Colombia, the Dominican Republic,
Guatemala, Mexico, the Netherlands, Peru, and the United States (Andrus and Wade,
1942; Barrus, 1918; Burkholder, 1923; Driflrout and Davis, 1989; Menezes and Dianese,
1988, Pastor-Corrales et al., 1994; Pastor-Corrales et al., 1995; Restrepo, 1994; Schwartz
et al., 1982; Tu et aL, 1984) but not found in this survey (Table 1.2).

DNA extraction of C. lindemuthianum races followed the protocol modified fi‘om
Edwards et al. (1991). Mycelium of each race, grown in petri dishes containing 20 ml of
liquid potato-dextrose medium (20% potato) and 240 rd ampicillin solution (25 mg. ml'1
in 70% EtOH), was used for DNA extraction. Each petri dish was inoculated with 4 plugs
of one single-conidial race, incubated for 7 days at 22 to 25°C in darkness without
shaking. Mycelium was harvested, surface-dried, lyophilized for 36 h, and stored at -
20°C. Dried mycelium was ground to a fine powder using a precooled sterile mortar and
pestle. Microcentrifuge tubes (1.5 ml) were filled with 260 mg of grormd mycelium
which was dispersed in 400 hr ofhot (65°C) 2x CTAB extraction bufi‘er (2% CTAB, 100
mM Tris Base, 10 mM EDTA, 0.7 M NaCl, pH 7.0). Four hundred ul of
phenol:chloroform°isoamyl alcohol (24:1 v/v) was added, the mixture was agitated on a
shaker for 15 min, and centrifuged at 3000 rpm for 5 min. The aqueous top layer was

transferred to a new microcentrifuge tube (1.5 ml) containing 400 pl of isopropanol,

34

 

 

 

 

83 £25 Ba .885 .3 - - + . . . . - . 2 as: 8
33 £25 88 825.5 saw—z - .. - + + + + + + .. - no
82 .8055 assesses: ..EN "966 - - - + + - + + + < 8m an
- .8528 - - - + - - + + - < 33 an

32 .8285 was mono—52
.82 £25 8s 2585 .288 - - - - + + + + + 2 502.5 ...:
$3 .083 e8 882 SE 00: - - - - + - + + + < 82.3.45 .8
- Ems. - - - - + - - + + < 582 S

82 .8 e 5
82.358 .32 - - - - + - - - + 2 3.80 2
32 6852 «£0 88 .8 - - - - - + + + + < £0 ...2
33 228: .58 ms .6 - - - - - - + + + < 3.8 a.
33 5853. 188.6 - - - - - - + - + < ass ...m
32 6853. .58 m .6 - - - - - - - + + < .82 .30.. m
32 6883. was 8 .6 - - - - - - - + - < .58 N
ooaocomom c35cm— : ~ I 0 9m m Q so on < 85 A«38:5 .32

a 83.30 38889

$85 8808 2: 8 was uni:

32088—0 83 835 385 05 88 gm ween—Baez 2: .8802 £883.30 8382.5 .3280 £85 8. 3832 £38.68
053 $22 832633882: U .«e 808 3 8.528. 3880MB. 52. 8888 me 8832 3 8388. Ea At 03:33.5 - .N._ 03:.

35

w§o> boo. i=3. ...... wast 3:8 62.56% 5.2. .8883 2.. 89a £3530 ... van—66230 macaw 356.8..— .vacm mmmu

w. .. ...... ..AVN... ...... ...... V. .AN. n- B. . .A-RN ob. A...N.- NoNSN .... .. .Az- NNN 8.52. 6 ..AN... $83.. .. ..A.. 3.6.3.
... .A..- NZ...- ..2...6. a 51352 as... 6 .AN- .83. ..3. .....N. 59.0.2. m .A-. 022.22. < .28. 3.2.6.82...
032’ Eng 03.0030.— .fioa ha v030=Oh §3~=E§- ..U mo moon.— oNflana-aao ca 68: @3333 €38.5me anus £68800 a
505a... 0.......2 u 2 2.8.5. I < .8. 2.8 ...... .
A822 8...... m: ...... .Amefisfiz. ..uz .A8...o2.
.32 .A...ao§o. .....o 62...»... 52......5. .... .8... 32.6. .6 .3228. ..6 3.56. ...6 .222... .... .8258 ..
A8... .....3. ...... 5.5....
Au. :3- ...... .A... 32.. on. A... 2......»- -N... .A... ..a - S A”. so- .. .2. .36...- 2 .A... 2......- 2 .A... 3.....- MN .9. 2......- N.
.336. .680 .3 cognwmmoc 2.333.... 305— .23. «0.260.665. 80.9? 233258.... .92.... o... no v8.3 80.... no nomaunwmmon— .

 

n8. .... .0 8.85.32...
3... .... .0 3.2.8.82...
3... £25 ...... 3255
ama— 6355 ...... $3.62
9.... £25 ...... 256.5
.2... £25 ...... .825
3a. .2255

33. .... u ...H

Na... .... ... N532

wwo. .8235 ...... $5.62
mg. .m .2 .355

«ma. .3285 ...... $5.52

awa— .omo....5 ...... $3.62

udowmqo
u a 40
xQNc.

..on BAD—0
«van

amo—

.32
.82

.32 .966
.2. .966

.2. .966

+++

+

+

+-

+

+

+

...T

+

1T

1T

1T

+

+-

+

E

22222

E<

#0
.82
.82

....m
.32
.82

m3 .50
.30

m9 .....m
Em
502 .....m

*hch
inhvg
quou
mnv
*vwn
«Na—

«cmA
A:

{Nam
65”
Am

 

A228. .N.. n...:

36

.2 ._ 25mg 2 03.: a Ba: 353 2w 3 8&2: a £33 3% aaaowazasz 3 Biz: 838:
Ammoo< .n was 3353— va - @2on .%E I 8.:on gauged :85 3.332 58902 a
32 .8 - is £258 (gm—am - $5 0 8.5 8 3330.. 2. 5282 880 .
A3382 1...: - .55 £36.84 was.” 2. c2828»? 32cm?
33 dd - $25 35 3mg .5 £3 85$ 2a 5.85 53m 5 $5535 3% 3222...
Amozuooéua 55 - fiasco a6 4335 «3:8ng 8 3285.5 .550 u :5 _.
”Coneaomou 23 “8:8 €880: 358 828— h
32— 25» 535‘ 2: 3 wane—on 332:0 .
“.8 2.. «.8 .38 . 5.8V 4 2: .38 - on a: v. .G
.8 - P: a $69 - ob _ .98 - SN 88:: w .360 - $8.. $58 a 578 - >262 Ba £5 320% m ”A32 4. .o

37
mixed well, and incubated at room temperature for 5 min. The emulsion was centrifirged

at 3500 rpm for 5 min, the supematant was discarded, and tubes were inverted for 5 min
to allow complete evaporation of the isopropanol. The pellet was dissolved in 100 pl TE
(10 mM tris-HCl, pH 8.0, 0.5 mM EDTA), 10 ugml'l RNAse A were added, and the
emulsion was incubated for 10 min at room temperature. Adding 50 pl cold 100% EtOH
precipitated DNA. After 10 min of incubation at room temperature, the mixture was
centrifuged at 3500 rpm for 5 min, and the pellet was washed twice with EtOH (100%).
Extracted DNA was dissolved in 100 pl TE and stored at -20°C. Concentration of DNA
samples was standardized to 10 ugh!" by DNA fluorometry (TKO 100, Hoefer
Scientific, San Francisco CA).

The PCR procedure reported by Williams et al. (1990) was followed with minor
modifications. Approximately 1.6 ul of genomic DNA template (10 ng.ul") and 1.6 ul of
single decamer primer (10 ng.p1") (Operon Technologies Inc., Alameda CA) was used in
a 18.22 ul amplification reaction. The reaction contained also 0.15 ul of Stofl‘el Fragment
Polymerase (10 units/ill) (Perkin Elmer Cetus, Norwalk, CN), 1x Stofl‘el bufi‘er (10x), 5
mM MgCl; (25 mM), and 2 mM of dNTPs (200 mM) (Perkin Elmer Cetus), overlaid
with 25 pl of sterile mineral oil prior to amplification. To ensure that amplification
product was not primer artifacts (Williams et al., 1990), genomic DNA was omitted fiom
a control reaction included for each primer examined.

Amplification was performed in a DNA thermal cycler (Perkin Elmer 480 Cetus)
programmed for one initial denaturation cycle (93°C/3 min), 40 step-cycles (94°C/45 sec;
38°C/45 sec; 72°C/1 min), one time-delay cycle (72°C/10 min), and a universal soak

cycle (8°C). Amplified RAPD products were electrophoresed at 70 V for 3.5 h on 1.2%

38
agarose gel using 0.5x THE (45 mM tris-borate, 1 mM EDTA. pH 7.0) nmning bufi‘er.

RAPD products were observed on 0.005% EtBr-stained agarose gel under UV light alter
electrophoresis. Ten pl lambda DNA, restricted with Hind III and EcoR Iwere included
as molecular weight marker.

A total of 3 ll primers were initially screened to select polymorphic primers for
races 3, 9, 17, 65, 73, 89, 102, 130, 201, 384, 449, 453 of C. lindemuthianum. These races
were chosen because of their pathogenicity to the first nine-anthracnose difi‘erential
cultivars. Eight primers were selected (GPA-9, GGGTAACGCC; OPP-13,
GGCTGCAGAA; OPG-Z GGCACTGAGG; OPG~3, GAGCCCTCCA; OPS-l7,
TGGGGACCAC; OPT-14, AATGCCGCAG; OPV-7, GAAGCCAGCC; and OPV-lo,
GGACCTGCTG) which consistently generated major polymorphic amplicons in all
isolates. Selected primers were then tested on all 60 isolates of C. lindemuthianum.
Molecular polymorphism among isolates with identical virulence phenotypes was
investigated. Isolates from races 3, 17, 23, 31, 55, 65, 73, 81, 102, 130 and 453 from the
same and difl‘erent countries, and isolates fi'om races 65 and 73 fi'om the same country
only, were tested with the eight selected primers. Two replicates of the RAPD assay were

run with different template DNA obtained from different DNA isolations.

Data analysis. The virulence and molecular data were analyzed separately using
cluster analyses. A single isolate of each unique combination of virulence and molecular
phenotypes was included in the analyses.

Separate data matrices were generated for the virulence and molecular data by

scoring resistance as 0 and susceptibility as l, and absence or presence of RAPD

39
amplicon as 0 or 1, respectively. RAPD amplicons were considered polymorphic if

shared by fewer than 57 of the 60 isolates of C. lindemuthianum. Similarity matrices for
both virulence and molecular data were derived with the SIMQUAL program in the
Numerical Taxonomy and Multivariate Analysis System for personal computer (NT SYS-
pc) version 1.70 (Exeter Software, Setauket NY). The DICE and Jaccard’s coefl'rcients
were used to compute distances in virulence and RAPDs data, respectively. The ratio
between presence of characters among two individuals and all possible combination of
unmatched characters among the same individuals is considered by both coeficients,
although DICE weights matches twice that of mismatches. The FITCH program in the
Phylogeny Inference Package (PHYLIP) version 3.50 (Department of Genetics,
University of Washington, Seattle WA) was used to estimate clustering fi'om the distance
matrix data using the Fitch-Margoliash method. Phenograms for both virulence and

molecular data were produced using the DRAWGRAM program (PHYLIP).

RESULTS

Virulence phenotypes. Forty-one races were identified among the 138 C.
lindemuthianum isolates tested for virulence (Table 1.1). Races fell into two categories,
those found over a wide geographic area (i. e., difl‘erent continents) and those restricted to
a single country. Only three races (7, 65 and 73) were widely distributed. Race 73 was
the most common, comprising 28% of the total sample. This race was found in four
separate countries and was isolated repeatedly in North, South and Central America. Race
7 was found at low frequency but extremely widespread, being detected once in South

and Central America, but it was isolated repeatedly from only the United States. The third

40

race, 65, was moderately common (14% of all isolates), but was isolatedly repeated from
only Brazil and the United States. The remaining races were detected in only a single
country, with about a third of these localized races being isolated multiple times within a
country. These races may be locally common, but geographically restricted. Variability
was highest in Central America (76%; ie., number of detected races /number of isolates
from a specific region) and decreased either north (North America 7%) or south (South
America 17%).

Although 39% (16 of 41) of the races were detected multiple times and three races
were widespread, no race was isolated fi'om plants from both P. vulgaris gene pools. For
example, races 65 and 73 were common and widespread, but they were always isolated
from hosts of the Middle American gene pool In contrast, race 7 was isolated from three
different countries but always from host from the Andean gene pool.

Races that were isolated from Middle American hosts were virulent to all
cultivars present in the difl‘erential series except cultivar L (Table 1.1). These races can
be categorized into two groups. One group was vimlent to specific resistance genes (Co-
2, Co-3, Co-4, Co-5, and Co-6). The majority of races fi'om this group attacked
differentials A and G. The differentials with the Co-2 and Co-4 genes were mostly
susceptible to races fi'om Honduras and Mexico, respectively. Races 1545, 1600, 1601,
1673, 1929, 1993 were virulent to both the Co-5 and Co-6 genes. Races from Honduras
and Mexico were the only races virulent to the Co-6 gene. Races virulent to genotypes
belonging to both gene pools formed the second group. Race 357 was virulent to the
Andean cultivars C and F and the Middle American cultivars G and I. Races 453, 469,

1165, 1431, 1677 and 1741 attacked the Andean cultivar C. Races 453 and 469 were

41

vimlent to the Middle American cultivars G, H and I. Races 1165, 1431, 1677 and 1741
caused disease on the Middle American cultivar K.

The group of races isolated from Andean hosts was virulent to resistance genes
from the Andean and Middle American gene pools. Race 2, 7, and 19 were virulent to the
Co-I gene in cultivar B, whereas race 7 was also virulent to the resistance genes in the C
cultivar. Race 55 was virulent to all three Andean cultivars B, C and F. These Andean
races were also virulent to the Middle American cultivars, A and E.

The phenetic analysis included 54 virulence phenotypes of C. lindemuthianum; 41
races identified in this survey (Table 1.1), and 13 additional races previously reported
from Brazil, Canada, Colombia, the Dominican Republic, Guatemala, Mexico, the
Netherlands, Peru and the United States (Table 1.2). The races clustered into three
groups, with one of these groups containing a single race, 2047 (Figure 1.1). There were
no obvious factors that correlated with the virulence clusters. The two multiple isolate
chrsters contained isolates from hosts of Andean and Middle American ancestry, and both
clusters contained isolates fiom North, South and Central America. However, isolates
from South America were slightly more common in cluster 2. Races that are known to be
widespread geographically were also found in clusters 1 and 2.

There were no obvious patterns within clusters 1 and 2, with races obtained fi'om
difi‘erent host gene pools ofien clustered tightly (e.g., 1033 and 23; 320 and 102; 31 and
2). No geographic patterning was evident within a cluster since isolates from each

geographic region were spread throughout each cluster.

42

Molecular variation Sixty isolates of C. lindemuthianum were separated based on
amplification product patterns from 312 combinations of primer-DNA templates. Primers
CPA-9, 0PF-13, OPG~2, OPG-3, OPS-17, OPT-14, OPV-7, and OPV-10 generated 11,
10, 10, 9, 7, 17, 11, and 18 distinct and reproducible polymorphic amplicons,
respectively. Ten RAPD amplicons were monomorphic. Fifteen separate clusters were
distinguished using the RAPD data (Figure 1.2). The number of isolates within a cluster
ranged fi'om 19 for cluster 2 to one for seven separate clusters There was no congruence
between the RAPD and virulence phenograms. Isolates that~ were identical for virulence
(ie., the same race) were most oflen dissimilar for RAPD markers. For example, the four
isolates that were classified as race 73 were distributed across three diflerent clusters
within the RAPD phenogram. RAPD-patterns obtained with primer CPA-09 and OPGOZ
are shown (Figure 1.3, Figure Al. 1). The analyses of isolates fi'om race 73 showed
polymorphism within and among countries (Figure 1.4). Isolates from race 73
characterized in Honduras and Mexico showed two distinct RAPD patterns among
isolates fi'om each country, whereas only two (lanes q and s) of the 13 isolates from the
United States were polymorphic (Figure 1.4). Isolates from race 65 characterized in the
United States showed different RAPD-pattem, whereas isolates from Brazil were
monomorphic. Of the ten races that displayed intra-race variability for RAPD phenotype,
only isolates from races 55 and 453 clustered within the same group of the RAPD
phenogram The clustering in the RAPD phenogram was not associated with geography
or the host gene pool from which the isolate was obtained. The two largest clusters, 1 and
2, contained isolates from all geographic regions sampled, and were collected on hosts

from both P. vulgaris gene pools. Even clusters that contained few isolates were

43

Figure 1.1. Phenogram of C. lindemuthianum races based on virulence to Phaseolus
differential cultivars. DICE (NT SYS-pc) generated a distance matrix data of virulence
data that was used in the FITCH program (PHYLIP) to estimate phylogeny. A. Binary
identification (Pastor-Corrales, 1991) ofC. lindemuthin races. B. Origin of isolates:
Arg (Argentina), Bra (Brazil), Clb (Colombia), CR (Costa Rica), DR (Dominican
Republic), Hon (Honduras), Per (Peru), Mex (Mexico), and US (United States); Div
(races identified from isolates collected in different countries). C. Middle American (M)
and Andean (A) race groups according to host source.

CllCC

31am

mm

Figure 1.1

 

44

 

 

 

 

l

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

39890‘3§“38§§§§§m”

1m

 

r-Ll‘ém. re.

 

 

 

 

 

‘ ...-5'0qu
Ease s

DIVM

 

S¥§§§9S§§ES§§§ig§g

as

45

Figure 1.2. Phenogram of C. Iindemuthiamm races based on RAPD analysis. Jaccard’s
(NT SYS-pc) generated a distance matrix data. FITCH program (PHYLIP) version 3.5c
estimated phylogeny from the distance matrix data using the Fitch-Margoliash method.

A Binary identification of C. lindemuthianum races (Pastor-Corrales, 1991); B. Origin of
races: Bra (Brazil), Can (Canada), Clb (Colombia), CR (Costa Rica), DR (Dominican
Republic), Gua (Guatemala), Hon (Honduras), Per (Peru), Mex (Mexico), and US
(United States); C. Middle American (M) and Andean (A) groups of races according to
host source. * Isolate obtained from the Michigan State University / Dry Bean Breeding
and Genetics Lab. ** Isolates obtained from Centro Intemacional de Agricultura Tropical
— Bean Pathology Program

46

CMJJAMMAAAAMMMAMAAAAJMM MMMMAMMMAAMMMAAMM...AMMMMMMMMMAMAAMMMMJ

am..mmmmmmmmhm.mmmmw.mwwmmmhwmmwmmmmhwmmh.mwmmmmwmmmmmmmmmmm.

Aamusmnsgvannmnmswummmnammmmmmmmmnmnenmmra nmmmmmmmM7mzmmmamm

EE .L

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ltd’s
3 5c
hot
iginof
an
lg to
rapid

Figure 1.2

47

Andean Middle American

 

abcdel'ghijklmnoporstuvx

    

-- '- ----—-~.-~—- - .H""
- -- - - - -- _—-.
-—- -

T
l — _--
1 ’--—-—-—- ——————————— --—-——..

Figure 1.3. Randomly amplified polymorphic DNA (RAPD) amplicons obtained with
primer 17 of Operon kit S for 24 single-spore isolates of Colletotrichum lindemuthianum:
a- 2 (Peru), b- 3 (Peru), 0- 5 - (Peru), d- 7 (Peru), e- 38 (Dominican Republic), f- 55
(Dominican Republic), g- 15 (Colombia), h- 102 (Brazil), i- 130 (United States), j- 23
(Brazil), k- 87 (Brazil), 1- 31 (Brazil), m- 17 (Brazil), n- 65 (Brazil), 0- 81 (Brazil), p-
89(Brazil), q- 453 (Mexico), r- 73 (Mexico), 9- 449 (Mexico), t- 457 (Mexico), u- 1673
(Honduras), v- 1993 (Honduras), x- 201 (Honduras), y- 2047 (Costa Rica). Races were
grouped according Andean and Middle American reaction.

48

abcdafghijklmnopqrs

 

Figure 1.4. Randomly amplified polymorphic DNA (RAPD) patterns obtained with
primer 13 of Operon kit F for 19 single-spore isolates from races 73 of Colletotrichum
lindemuthianum collected in Honduras (lanes a to (1), Mexico (lanes e and f), and the
United States (lanes g to s).

'I‘

49
geographically variable. For example clusters 4, 6, 8 and 13 contained isolates fi'om both

Central and South America. Clusters 4 and 8 also contained isolates obtained from hosts
of Andean and Middle American ancestry. The only geographic tendency was for isolates

from North America to be found predominantly in clusters 1 and 2.

DISCUSSION

Molecular markers are used extensively to characterize plant pathogens
(Crowhurst et al., 1991; Guthrie et al, 1992; Michelmore and Hulbert, 1987; Mills et al.,
1992). When combined with data on virulence, these markers can often elucidate the
population genetic structure and evolutionary relationships of plant pathogens
(McDonald and McDermott, 1993; Milgroom, 1995; Sreenivasaprasad et al, 1992). This
information can suggest novel strategies for the control of pathogens For example, rice
blast populations, Pyricularia oryzae, are often composed of several distinct lineages,
with each lineage being virulent on a sub-set of available host cultivars (Casela et al.,
1995; Levy et a1, 1993). Because of this structure, breeding programs have concentrated
on obtaining resistance against specific pathogen lineages and not virulence phenotypes
(Casela et a1, 1995; Levy et al., 1993).

The division of common beans, P. vulgaris, into Andean and Middle American
gene pools (Gepts, 1988; Singh et al., 1991), suggests the possibility of a similar lineage
based control strategy. prathogens are specialized on one of the host gene pools, then
transferring resistance genes between gene pools may provide a more durable resistance
(Gepts, 1988; Singh et al, 1991). Common bean rust (Uromyces phaseoli; Maclean et al.,

1995; Stavely, 1982; Stavely 1984) and angular leaf spot (Phaeoisariopsis griseola;

50
Guzman et al., 1995) appear to be specialized in congruence with the two P. vulgaris

gene pools. A previous study of twelve C. lindemuthianum isolates found that RAPD and
restriction fragment length polymorphism (RFLP) markers divided the species into two
groups (Fabre et al., 1995). While not absolute, the groupings suggested some
specialization within C. lindemuthianum that corresponded to the two gene pools within
P. vulgaris. Pastor-Corrales (Pastor-Corrales, 1996) reported that C. lindemuthianum can
be divided into two groups based on virulence; one specializing on hosts from the Middle
American gene pool and the second specializing on Andean hosts.

Our data for virulence phenotypes and RAPD markers suggest that C.
lindemuthianum is not structured in congruence with host gene pools Virulence
phenotypes clustered predominantly into two large groups, but each group contained
races isolated from both of the host gene pools. The virulence data can also be examined
for the ability of races, isolated from hosts from one gene pool, to infect differential
cultivars fiom the other gene pool. Differential cultivars B, C and F, which represent the
Andean gene pool, were susceptible to 3%, 17% and 3%, respectively, of the 37 races
isolated from hos of Middle American origin (Table 1.1). Further, some of the isolates,
previously collected from Middle American hosts, were found to infect these Andean
cultivars (Table 1.2). Three of the four C. lindemuthianum races obtained fi'om Andean
hosts could infect at least one differential cultivar fi'om the Middle American gene pool
(Table 1.1). We note, however, that these results are based on greenhouse inoculations
Despite the ability of some races to overcome resistance genes from both the Middle
American and Andean host gene pools, each race was actually collected from hosts of

only one gene pool.

5 1
A previous study reported that widespread races (i. e., races found in five or more

Latin American countries) of C. lindemuthianum infected a smaller number of differential
cultivars (Pastor-Corrales, 1996). Our data tend to agree with this finding; the three
widespread races, 7, 65 and 73, could attack two or three differentials, while the average
number of difl‘erentials infected by a race was 3.73. Although our sampling was not
designed to explicitly measure diversity within an area, our data suggest that C.
lindemuthin is more variable in Central America, than in either South or North
America. This finding is consistent with the earlier work of Pastor-Corrales (Pastor-
Corrales, 1996) who found that the C. lindemuthianum population in Central America
was more diverse than isolates from Andean areas. Also, genetic diversity of C.
lindemuthianum populations was higher in Mexico compared with Ecuador and
Argentina (Sicard et al, 1997).

RAPD markers separated our isolates into 15 clusters. The clustering within the
RAPD phenogram was not congruent with either geographic location or host gene pool;
nor was there any congruence with the virulence phenogram. Thus, the RAPD primers
utilized here cannot be used for grouping C. lindemuthianum according to the host gene
pool. It is not clear to what degree our RAPD phenogram represents phylogentic
relationships within C. lindemuthianum. Recent reports have pointed out a number of
problems associated with assessing phylogenetic relationships using RAPD markers
(Pahrmbi, 1996). Our groupings differ from an earlier report based on twelve C.
lindemuthin isolates (F abre et al., 1995), which found two groups using both RAPD
markers and two different sets of RFLP markers from the ribosomal genes. Sicard et al

(1997) grouped C. lindemuthianum isolates using RFLP of the amplified ribosomal

52
internal transcribed spacer region (rDNA-ITS). The authors demonstrated that the three

groups corresponded to the host gene pools. These results suggest an adaptation of strains
on cultivars of the same geographic origin.

Taken together, our data indicate that C. lindemuthianum may not be highly
structured to specific P. vulgaris gene pools. Therefore, control strategies that transfer
resistance genes fi'om one P. vulgaris gene pool to the other may not confer durable
resistance to C. lindemuthianum. We urge caution in the utilization of this strategy until
races are further evaluated, under field conditions, for their ability to infect hosts from
both gene pools, and a more comprehensive analysis of the evolutionary relationships

within C. lindemuthianum is completed.

53
LIST OF REFERENCES

Andrus, C. F., and Wade, BL. 1942. The factorial interpretation of anthracnose
resistance in beans. Pages 1-29 in: Tech. Bull. No. 310. USDA. Washington.

Balardin, R, and Kelly, JD. 1996. Identification of Race 65-epsilon of bean anthracnose
(Colletotrichum lindemuthianum) in Michigan. Plant Dis. 80:712.

Balardin, R. S., Pastor-Corrales, MA, and Otoya, M.M. 1990. Variabilidade patogénica
de Colletotrichum lindemuthianum no Estado de Santa Catarina. Fitopatol. Bras.
15:243-245.

Balardin, R. S. and Kelly, J. D. 1997. Re-characterization of Colletotrichum
lindemuthianum races. Ann. Rep. Bean Improv. Coop. 40: 126-127.

Barrus, M. F. 1918. Varietal susceptibility of beans to strains of Colletotrichum
lindemuthin (Sacc. & Magn.) B. & C.. Phytopathology 82589-605.

Blondet, A. 1963. L'anthracnose du haricot: etudé des races physiologiques du
Colletotrichum lindemuthianum. Ph. D. these, Faculté de Sciense, Paris, France.

Burkholder, W. H 1923. The gamma strain of Colletotrichum lindemuthianum (Sacc. et
Magn.) B.et C.. Phytopathology 13:316-323.

Casela, C. R, Ferreira, AS., Zeller, KA, Levy, M. 1995. Pathotype variation in the
sorghum anthracnose firngus: a phylogenetic perspective for resistance breeding.
Pages 257-288 in: Disease analysis through genetics and biotechnology. J. F.
Leslie and R S. Frederiksen, eds. Iowa State University Press Ames, IO.

Crowhurst, R. N., Hawthorne, B.T., Rikkerink, E.HA., Templeton, MD. 1991.
Differentiation of Fusarium solani £sp. cucurbitae races 1 and 2 by random
amplification of polymorphic DNA Curr. Genet. 20:391-396.

Driflrout, E., and Davis, 1.1-1C. 1989. Selection of a new set of homogeneously reacting
bean (Phaseolus vulgaris) difi‘erentials to difl‘erentiate races of Colletotrichum
lindemuthianum. Plant Pathology 38:391-396.

Duvick, D.N. 1996. Plant Breeding, an evolutionary concept. Crop Sci. 36:539-548.
Edwards, K C., Johnstone, C., and Thompson, C. 1991. A simple and rapid method for

the preparation of plant genomic DNA for PCR analysis. Nucleic Acids Res.
19:1349.

54

Fabre, J. V., Julien, J., Parisot, D., and Dron, M. 1995. Analysis of diverse isolates of
Colletotrichum lindemuthin infecting common bean using molecular markers.
Mycol. Res. 99:429-435.

Fouilloux, G. 1979. New races of bean anthracnose and consequences on our breeding
programs. Pages 221-235 in: Disease of tropical food crops. H Maraite and J.A
Meyer, eds. Louvain-la-Neuve, Belgium.

Garrido, E. R 1986. Identificacion de razas fisiologicas de Colletotrichum
lindemuthianum (Sacc. & Magn.) Scrib. en Mexico y busqueda de resisténcia
genética a este hongo. MSc thesis, Institucion de Ensenanza y Investigacion en
Ciencias Agricolas, Montecillos, Mexico.

Gepts, P. 1988. A Middle American and an Andean common bean gene pool Pages 375-
390 in: Genetic resources of Phaseon beans; their maintenance, domestication,
and utilization. P. Gepts, ed. Kluwer. London

Guthrie, P. A I., Magill, C.W., Frederiksen, RA, and Odvody, G.N. 1992. Random
amplified polymorphic DNA markers: a system for identifying and difl‘erentiating
isolates of Colletotrichum graminicola. Phytopathology 82:832-835.

Guzman, P., Gilbertson, RL., Nodari, R, Johnson, W.C., Temple, S.R, Mandala, D.,
Mkandawire, ABC, and Gepts, P. 1995. Characterization of variability in the
fungus Phaeoisariopsis griseola suggests coevolution with the common bean
(Phaseolus vulgaris). Phytopathology 852600-607.

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

Kruger, J., Hoffinann, GM, and Hubbeling, N. 1977. The kappa race of Colletotrichum
lindemuthianum and sources of resistance to anthracnose in Phaseolus beans.
Euphytica 26:23-25.

Levy, M., Correa-Victoria, J., Zeigler, RS., Xu, 8., and Hamer, J. 1993. Genetic diversity
of the rice blast fungus in a disease nursery in Colombia. Phytopathology
83: 1427-1433.

Maclean, D. J., Braithwaite, KS., Irwin, J.AG., Manners, J.M., and Groth, J.V. 1995.
Random Amplified Polymorphic DNA reveals relationships among diverse
genotypes in Australian and American collections of Uromyces appendiculatus.
Phytopathology 85:757-765.

McDonald, B. A, and McDermott, J. M. 1993. Population genetics of plant pathogenic
firngi. BioScience 43:311-319.

55

Menezes, J. R, and Dianese, JG. 1988. Race characterization of Brasilian isolates of
Colletotrichum lindemuthianum and detection of resistance to anthracnose in
Phaseolus vulgaris. Phytopathology 7 8:650-655.

Michelmore, R W., and Hulbert, SH 1987. Molecular markers for genetic analysis of
phytopathogenic firngi. Annu. Rev. Phytopathol 25:383-404.

Milgroom, M. G. 1995. Analysis of population structure in fimgal plant pathogens. Pages
213-229 in: Disease analysis through genetics and Biotechnology. J. F. Leslie,
and RA Frederiksen, eds. Iowa State University Press. Ames, IO.

Mills, P. R, Sreenivasaprasad, S., and Brown, AE. 1992. Detection and difl‘erentiation of
Colletotrichum gloeosporioides isolates using PCR FEMS Microbiology Letters
98: 137- 144.

Palumbi, SR 1996. Nucleic acids 11: the polymerase chain reaction. Pages 205-247 in:
Molecular Systematics. D.M. Hills, C. Moritz and BK. Mable, eds. Sinauer,
Sunderland, MA

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

Pastor-Corrales, M. A 1996. Traditional and molecular confirmation of the coevolution
of beans and pathogens in Latin America. Ann. Rep. Bean Improv. Coop. 39:46-
47.

Pastor-Corrales, M. A, Erazo, O.A., Estrada, E. 1., and Singh, S. P. 1994. Inheritance of
anthracnose resistance in common bean accession G 2333. Plant Dis. 78:959-962.

Pastor-Corrales, M. A, Otoya, M.M., and Molina, A 1995. Resistance to Colletotrichum
lindemuthianum isolates fiom Middle America and Andean South America in
difl‘erent common bean races. Plant Dis. 79:63-67.

Pastor-Corrales, M. A, and Tu, JG. 1989. Anthracnose. Pages 77-104 in: Bean
production problems in the tropics H. F. Schwartz, and Pastor-Corrales, M.A,
eds. CIAT. Cali, Colombia.

Rava, C. A, Molina, J., Kaufiinann, M., and Briones, I. 1993. Determinacion de razas
fisiologicas de Colletotrichum lindemuthianum en Nicaragua. Fitopatol. Bras.
18:388-391.

Restrepo, S. 1994. DNA polymorphism and virulence variation of Colletotrichum
lindemuthianum in Colombia. M. Sc. thesis Universite Paris IV, Paris-Grignon,
France.

56

Rodriguez-Guerra, R 1991. Identificacion de razas patogenicas de Colletotrichum
lindemuthianum (Sacc. y Magn.) Scrib. en el estado de Durango mediante un
sistema propuesto intemacionalmente y respuesta de genotipos de fiijol tolerantes
a sequia a razas del patogeno. M.Sc., Parasitologia Agricola, Universidad
Autonoma Agraria "Antonio Narro", Buenavista, Mexico.

Schwartz, H. F., Pastor-Corrales, M.A, and Singh, S. 1982. New sources of resistance to
anthracnose and angular leaf spot of beans (Phaseolus vulgaris). Euphytica
3 1:741-754.

Sicard, D., Michalakis, Y., Dron, M., and Neema, C. 1997. Genetic diversity and
pathogenic variation of Colletotrichum 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.

Sreenivasaprasad, S., Brown, AE., and Mills, RR 1992. DNA sequence variation and
interrelationships among Colletotrichum species causing strawberry anthracnose.
Physiol. and Mol. Plant Pathol 41:265-281.

Stavely, JR 1982. The potential for controlling bean rust by host resistance. Ann. Rep.
Bean Improv. Coop. 25:28-30.

Stavely, J .R 1984. Pathogenic specialization in Uromyces phaseoli in the United States
and rust resistance in beans. Plant Dis. 68:95-99.

Tu, J. C., Sheppard, J.W., and Laidlaw, D.M. 1984. Occurrence and characterization of
the epsilon race of bean anthracnose in Ontario. Plant Dis. 68:69-70.

Waterhouse, W. L. 1955. Studies of bean anthracnose in Australia. Proc. Linn. Soc.
N.S.W. 80:71-83.

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

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

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

57

Yormg, RA, Melotto, M., Nodari, R0, and Kelly, JD. 1997. Marker assisted dissection
of oligogenic anthracnose resistance in the common bean cultivar, G2333. Theor.
Appl. Genet. (in press).

Zaumeyer, W.J. and Thomas, HR. 1957. A monographic study of bean diseases and
methods for their control. USDA

Chapter 2

RIBOSOMAL DNA POLYMORPHISM IN COLLETOTRICHUM
LHVDEMUTHIANUM

ABSTRACT

Intra-specific divergence among isolates of Colletotrichum lindemuthianum
collected in Argentina, Brazil, Colombia, the Dominican Republic, Honduras, Mexico,
Peru and the United States was determined using RFLP analysis and sequencing of the
rDNA region of the two internal transcribed spacers (ITS l and ITS 2) and the 5.88
rRNA gene. A reproducible 0.58kb fragment was PCR amplified in the 57 isolates
including C. lagenarium used as an outgroup species. Races were grouped into two
clusters by RFLP-ITS analysis. Group I was formed predominantly by Andean races
except race 23, which was placed within group II. The Middle American races were
observed in both groups. A bootstrap value of 100% in the parsimony and 88% in the
neighbor-joining analyses supported a monophyletic group formed by all isolates except
race 31. Genetic distances among races of C. lindemuthianum ranged fi'om 0.2% to 2.9%.
Sequence homology analysis did not show a pattern parallel to a specific host reaction
group or associated with an obvious geographic distribution. Likewise, phenetic and
parsimony analysis did not show polymorphism in the rDNA region linked to any
specific factor. Molecular polymorphism among isolates of races 7, 17, 31, and 73

collected in difl‘erent cormtries was demonstrated by RFLP-ITS analysis. Sequence

58

59
homology of ITS regions of isolates of race 73 from Mexico and the United States

showed the greatest genetic distance value among all isolates These findings support a
level of molecular variability within C. lindemuthianum greater than the variability

previously characterized by virulence analysis.

60
INTRODUCTION

Virulence and molecular techniques have been used to characterize variability in
C. lindemuthianum (Fabre et al., 1995; Sicard et al., 1997, Balardin et al., 1997). Genetic
diversity inferred from virulence data is restricted to the virulence genes and does not
reflect the whole genetic diversity present in the C. lindemuthianum genome (Balardin et
al, 1997). The genes involved in host-specificity represent a very small fiaction of the
genes in the pathogen and may be subjected to strong selective pressure by the host
(Lermg et al., 1993). Genetic markers that are selectively neutral, reproducible, and
relatively easy to assay have been suggested for population genetics analysis (Milgroom,
1995; McDonald, 1997). In addition, the analysis of high numbers of independent loci
with no effective selection pressure might give more accurate information about genetic
relationships among individuals within a population (Leung et a1, 1993). Because a large
number of amplicons can be screened in a relatively short period of time, RAPDs are
especially useful in difl‘erentiating clonal lineages for fungi that reproduce asexually
(McDonald, 1997 ). However, sample purity, lack of reproducibility of PCR reactions
among labs, and data interpretation are limitations that might bias the potential role of
RAPD markers in phylogenetic analysis (Smith et al., 1995).

Nucleotide sequences offer extremely high resolution of intraspecific diversity
and generate data that can be used to estimate sequence divergence (Dowling et al.,
1996). The nuclear rDNA sequences coding for the small subunit (188) and large unit
(288) RNAs show little evolutionary change and sequence data can be used to measure
evolutionary divergence among organisms within a population (Hillis et al., 1996).

Analyses of closely related groups based on the variable regions within the rRNA genes

61
have been used to resolve taxonomy ambiguities (W aalvvijk et a1. 1996; Arora et al.,

1996). PCR amplification of ITS region of rDNA was used to characterize variability in
C. gloeosporioides (Mills et al., 1992). Molecular variability within and between isolates
of C. acutaiwn, C. gloeosporioides and C. fiagariae was assessed using rDNA-ITS 1
region (Sreenivasaprasad et al, 1992). Isolates of C. fiagariae were separated into two
groups by homology analyses of ITS sequenced regions, whereas isolates fi'om C.
gloesporioides were monomorphic. Similarity among isolates of C. gloesporioides, C.
musae, C. fi-uctigenum, and C. acutatum was observed based on the sequence of ITSl
region of Colletotrichum spp (Sreenivasaprasad et al, 1994) Sherrifl‘ et al. (1994) studied
the relationships within the genus Colletotrichum using the PCR amplification of rDNA
Their results clustered C. lindemuthianum, C. malvarum, C. irifolii, and C. orbicularae,
whereas C. gloesporioide represented a group-species within Colletotrichum spp.

Restriction fiagment length polymorphism of both nuclear ribosomal genes and
the internal transcribed spacer region of the ribosomal genes showed that C.
lindemuthianum isolates consistently formed two main groups, but some isolates did not
fall distinctly into one or the other group (Fabre et al 1995). The limited number of
isolates used in this study did not allow determination of whether the groups were aligned
according to host origin or virulence. Using RFLP-ITS, Sicard et al. (1997) grouped
isolates of C. lindemuthianum collected fi'om wild Phaseolus in Mexico, Ecuador and
Argentina. Adaptation of races on cultivars of the same geographic origin was suggested
because groups corresponded to three host gene pools. In this work we sought to resolve
the genetic variation between 45 isolates of C. lindemuthianum collected in South,

Central and North American countries. Variability within virulence phenotypes was

62
observed and the efl‘ect of host gene pool on C. lindemuthianum diversity is discussed.

MATERIALS AND METHODS

Colletotrichum lindemuthianum isolates, DNA extraction and amplification
protocol. Filly- seven isolates from Argentina, Brazil, Colombia, Costa Rica, the
Dominican Republic, Honduras, Mexico, Peru and the United States were characterized
based on RFLP analysis or sequencing of the rDNA region comprising the two internal
transcribed spacers (ITS 1 and ITS 2) and the 5.88 rRNA gene (Table 2.1).

DNA was extracted from races of C. lindemuthianum according to protocol described in
Balardin et al. (1997). PCR amplification of the rDNA region between 3’ end of the 18S
gene and the 288 gene was conducted using the primers PN3: 5’-
CCGT‘TGGTGAACCAGCGGAGGGATC- 3’, and PN10: 5’-
TCCGCT'I‘AT'I‘GATATGCT'I‘AAG- 3’ (Fabre et al. 1995). Approximately 3.0 pl of
genomic DNA template (10 ng.p1'l) and 1.5 pl of each primer (10 ng.pl")
(Biotechnologies Inc., Woodlands, TX) was used in a 30p] amplification reaction. The
reaction contained 0.15 pl of Stofi‘el Fragment Polymerase (10 units/ pl) (Perkin Elmer
Cetus, Norwalk, CN), 1x Stofi‘el bufl‘er, 5 mM MgC12, and 2 mM of dNTPs (Perkin
Elmer Cetus), overlaid with 25 pl of sterile mineral oil prior to amplification.
Amplification was performed in a DNA thermal cycler (Perkin Elmer 480 Cetus) with the
following profile: one initial denaturation cycle (94°C/4 min), 30 step- cycles (94°C/ 1
min; 55°C/45 sec; 72°C/2 min), one time-delay cycle (72°C/7 min), and a universal soak

cycle (8°C).

63

 

 

 

_ - . - - . - - + + + < .26 SN 6 sneeze m.
a - - - - - . - + - + 2 ._ .3 8: 6 :58: a
a - - . - - - - + - - < .5.— e 6 and a
_ - - - - - - - - - + < ._ .2 m: 6 33m 8%:
a - .. - - - - - - - + < Eu.— 2. 6 and
a - - - - - - - - - + < ._ .2. as: 6 See:
a - - - - - - - - - + < .66 s: 6 £53
a - - - - - - - - - + < L .N we... 6 Edema. a.
_ - - - - - - - - + + < es.— s 6 Bed w
_ - - - - - - - - - + < .3 m 6 Edd n
_ - - - - - - - - - - < es.— 5 6 ads N
.— - - - - - - - - - + 2 ...: em 6 62m _

.— 6. a _ a e ... m a <
2.5.5. ..5 .328— Feeeu .85.
a 93.530 15.6.6.5:

 

mazes when? was geese ea «3% use: 2.. ..a

icon dean—6:62 .8u82 @2365: .3333— :aomflaen— 05 .32 330 £38.60 .3260 468m academic. £ 360:3 89$
858— .EBBESESSS U .«e moan—8m hm 3 3.333 326.6% .62— .5883 we gunned 3 .3662 can A+V o-awaoomzm A .u 02:.

64

 

++++++++

++++++++

++++++++

+++++++++-

++++++

+++I

++++++++

222<<<22<<<<22222222

. 33 E66
.3 m2 6
._ .e m: 6

._ «N .32 6
._ ... 8: 6
.3 am 6
.2: m: 6
.3. .266
.EN .966
Rn S .6
$6 v .6
.ES : 6
.sm 2 .6
.am .966
.32

.2 .966

._ .N an 6
.32

.82

.2 am 6

:55

3.8»...

.35 use:
:3on

mans—6&5

62m

.85 8&5

.55

:35

£23.. 5.388
232—3— 5638.5
cacao... $238.5
$355.2

:55

.33 8a.:

62m

.325 8.388
.35 8%:
3260

Ban

5

Mb

we

mm

hv

an

an

MN

@—

b—

 

92506 5 as..."

65

 

EHEEEEEEEEEE—‘HEHHEH—

+++++

++++++++

+++

+

+

+++++++++++.

+

2<<<2222222222222222

L.~N no: .5
0N2 .6: .5
L .m an: .5
L6: .6: .5
L6: :6: .5
LA: 52 .5
L .2 .6: .5
L .2 x02 .5
o awn #755
L .3 ~82 .5
L .3” x02 .5
L .3... x02 .5
L 6 x02 .5
L d— 52 .5
L 6m 682 .5
L .N .6: .5
93m:

.DmE

«awn R755
omen “H755

3.526:
6.826:
3202
8.526:
2626:
8202
man—26:
8202
:65
320—2
3202
8202
8202
8202
3202
3526:
933m 632:
3:56:62
:22:
~23:

8:
new:
«be—
me u a
mac:
man
“an
FM?
nmv
a:

bmm
nun
hm"
on"
SN

am—
No:
an

66

..Baw£ m: 23.. are... ...—.2 3 38.5. a 2.. _ 32.26 .
.Nm: .2...» <26 mm.“ .5: 33% .836 28.. ...—.2 .
SEN .. 22
9 .— 2... .58 - on as 6. as m - at H .33 - ob. .32 - SNSN E = A3 - N2 8826 o .5. - :83: L AB - 86va
m .3 - ~32. 6.88 a .c. - >85: >56 6 .3 - 666 3. £5 5&6sz m .2 - 052226 < ”:2: ......Eoéué
2:; .933 guacamou .66 .3 326:2 E§6$L=Eu2$ .5 .6 v.38 366.36 8 com: @6226 326696 .62. 68860 n
.3055. .822 I 2 .885. n < L8.— 250 .8: .
.35. .8 .65 £38 49285 - ~22 o 3.6; 3 9.33.5 3 .8382 8.86 .
.Amoauooéufl a. ...6 £828 .66 .2553? 3%: 5 .286 €865.25 - :6 .
3.3. d d .835 Bi: .5 6&3 2m .883 mom a... 86 .556: 23m 3222 .
.358 3:8— .—
.ASE .23 2. 6331338 - ...? .av 32. - cm .3 «an...» - NS .83 6.262% n a
.3 a: n a .9 so I a .3 86% u 3 .3 «Ba: 1 a .3 gas. - 3. .@ £8 - ...N .03 386:6... .. 2 .3 2...: - S
”floun— 335 .3 632363 33236 .605— .39: «02.60.6655 6898 0666.686: >663 05 _6 26.3 v.33 .6 6:398: .

 

HF-‘HI—‘fl

+++++

+++++

+

+

+

+++++

+

+

+

+

+++++

+

+

+++++

2

2222

La 6 .6

L .8 8: 6
L .M: 8: 6
L .o 8: 6
L._ 8: 6

32 360
3626:
mgcac:
2:326:

.6326:

ban
mam:
ana—
:3:
are—

67
Restriction endonuclease digestion of PCR-I T S. The ITS-PCR products were

digested with l U of restriction endonuclease (Hoe 111, Msp I - GibcoBRL Grand Island,
NY; EcoR I, Hind Ill and BarnH I - Boehringer Mannheim, Germany) according to the
manufacturer’s specifications. The digestion reaction contained 10 ul of the amplified
product, 2 ul of the digestion enzyme bufl‘er and 8 ul of dngO. The digestion products
were electrophoresed at 70 V for 3.5 h on 2% agarose gel using 0.5x THE (45 mM tris-
borate, 1 mM EDTA, pH 7.0) nmning bufl‘er. RFLP products were observed on 0.005%
EtBr-stained agarose gel under UV light afier electrophoresis. Ten pl of the 1.5 kb DNA

molecular weight marker were included. The experiments were performed twice.

Automated DNA Sequencing. In order to sequence ITS l and 2 regions in both
directions, PN3 and PNIO were added to individual reaction mixtures afler PCR-
amplification. For each DNA strand a ZO-ul reaction mixture was prepared with 10 pl of
the PCR product, 1 ul of each primer, and 9 ul of sterile H20. Sequences were obtained
using the fluorescent dye dideoxy nucleoside triphosphate terminator method (MSU-
DNA Sequencing Facility, East Lansing, MI). Sequencing reactions were run on a

polyacrylamide gel using the ABI 373A DNA Sequencer.

Data analysis. Nucleotide sequence data were analyzed using different
phylogenetic methods. Sequences of nucleotides were aligned using a subroutine in the
SEQUENCI-IER program based on the CLUSTAL V Method with weighted residues.
The nucleotide sequences (232986 - ITSI, Sreenivasaprasad et a1, 1996; no. 218975 -

ITSZ, Sherrifl‘ et aL, 1994), available in the EMBL database, were used for the sequence

68
alignment. In order to find the most likely tree topology, phylogenetic trees were

generated by both distance and parsimony methods.

The distances among isolates of C. lindemuthianum were estimated by the Jukes-
Cantor one-parameter method (Jukes and Cantor, 1969) in the DNADIST program,
Phylogeny Inference Package (PHYLIP) version 3.5c (Department of Genetics,
University of Washington, Seattle WA). The Juices-Cantor method (J-C) assumes an
equal rate of substitution between all pairs of bases as the distance measures the
variability. Phylogenetic analysis was carried out by the Neighbor-Joining (N-I) method
(Saitou and Nei, 1987) in the NEIGHBOR program (PHYLIP), and Fitch-Margoliash
method in the FITCH program (PHYLIP), using the distance matrix computed by J-C
method. The N-J method follows the minimum-evolution principle, which consists of
choosing the tree with the smallest sum of branch lengths. The Fitch-Margoliash method
was used to estimate clustering fi'om the distance matrix computed by J-C method.

The bootstrap was performed by SEQBOO'I‘ program (PHYLIP) to determine the
confidence limits for the internal branches in each phenogram An unrooted parsimony
analysis was carried out using the DNAPARS program (PHYLIP), which analyzed each
bootstrap replicate. Using the CONSENSUS program (PHYLIP) a majority rule
consensus tree was generated. The consensus tree consists of monophyletic groups that
occur as oflen as possible in the data, and can be considered an overall estimate of the
phylogeny (Felsenstein, 1985).

The patterns of DNA bands originating from RFLP patterns within ITS fiagments
were compiled in a matrix considering presence (1) and absence (0) of bands. Only the

polymorphic patterns inferred from endonuclease digestions were included in the matrix

69
The relationships among isolates of C. lindemuthianum were estimated using the DOLLO

program (PHYLIP). Phenograms were generated by DRAWGRAM program (PHYLIP).

RESULTS

Amplification and RFLP of I T S], 5. 8S, and 1 T S2 regions. PCR amplification of
the region between 188 and 288 genes (ITSI, 5.88 rRNA gene and ITSZ) with PN3 and
PNIO primers produced a single double-stranded reproducible 0.58 kb fi'agment in all 57
isolates of C. lindemuthianum (Table 2.1) including C. lagenarium (Figure 2.1). The
endonucleases Hae III, Msp I, EcoR I, Hind III and Band! I were used to digest the
amplified rDNA region of C. lindemuthianum isolates. No polymorphisms were observed
with EcoR I, Hind III and M I although these mdonucleases showed restriction sites
within ITS 1 and ITS 2 regions. The isolates of C. lindemuthianum formed two groups
according to restriction of the ITS-DNA region with the endonucleases Msp I and Hae III
(Table 2.1). Group I was formed by isolates exhibiting both the Middle American and
Andean reaction, whereas the group II was formed predominantly by isolates with
Middle American reaction, except for the presence of the Andean race 23. The
geographic origin was not a factor influencing the clustering of isolates of C.
lindemuthinaum, except the Andean isolates from the Dominican Republic, which were

placed within group I (Table 2.1).

70

Undigested Msp I Hae III

058 kb
—>

 

Figure 2.1. PCR amplification products of the ITS], 5.88 rDNA, and ITSZ region of the
rDNA of selected isolates of C. lindemuthianum. Lanes 2 to 7 show non-digested ITS
fragments of races 7, 8, 31, 65, 1993, and the outgroup species C. lagenarium; Lanes 7 to
12 show the same races digested with the restriction enzyme Msp I; Lanes 14 to 19 show
the same races digested with the restriction enzyme Hae II]. Lanes 1 and 20 contain the
1.5kb DNA molecular weight marker.

7 1
Intra-race variability based on RFLP and sequencing of the 1 TS 1, 5. 8S, and I T S2

regions. RFLP analyses of several isolates of races 7, 17, 23, 31, 65, 73, and 130
collected in difl‘erent cormtries was carried out to investigate intra-race polymorphism
within the ITS region. Of the seven races analyzed, races 23, 65 and 130 did not display
intra-race variability (Figure 2.2). Isolates of race 7 characterized in Argentina,
Colombia, and Mexico showed distinct RFLP patterns compared to the isolate fiom the
United States. Isolates of race 17 characterized in Canada were different from the isolates
collected in Brazil and the United States. Distinct RFLP patterns for isolates of race 31
collected in Brazil and Netherlands was observed. Isolates of race 73 collected in Brazil,
Honduras and the United States showed different RFLP banding pattern to the isolate
from Mexico. The genetic distances computed by the J-C method based on analysis of
sequences of C. lindemuthianum showed that race 73 from Mexico had the greatest
genetic distance from isolates of races CR 2047, US 23 and US 73 (Table 2.2). The
genetic dissimilarity among isolates of race 73 was supported by results from RFLP-ITS.
Genetic similarity among races was observed despite the host reaction group or
geographic origin of races. For instance, the shortest genetic distance was observed
among the Middle American races 8, 81, 89, 457 and 2047 and the Andean races 5, 15,
and 23 (Table 2.2). Races 8 and 5 were collected in Peru and the other races were

collected in Argentina, Brazil, Colombia, Costa Rica, Mexico and the United States.

Phylogeny of C. lindemuthianum isolates. The phylogeny of C. lindemuthianum
isolates was inferred from the sequence comprising ITS l and ITS 2 regions and the 5.88

rRNA gene. The ITS 1 and ITS 2 sequences were aligned to the accessions 232986 and

72

73 65 23130 31 17 7
I 'abcd'et'gh'lj'kl'mno'pqrs‘

 

 

Figure 2.2. PCR amplification products of the ITSl, 5.8S rDNA, and ITSZ region of the
rDNA of C. lindemuthianum isolates digested with the restriction enzyme Msp I, from
races 73 (lanes a to d), 65 (lanes e and f), 23 (lanes g and h), 130 (lanesi and j), 31 (lanes
k and 1), 17 (lanes m to o), 7 (lanes p to s).

73
Z18975 in the EMBL database (Figure 2.3). The multiple sequence alignment of ITS 1

showed differences among sequences of isolates due to single nucleotide deletions,
substitutions or insertions in 25% of the 162 positions. In the ITS 2 region, only single
nucleotide insertions were observed in 8% of the 167 positions (Figure 2.3).

The phylogenetic tree derived from the Bootstrap and NJ analyses of the C.
lindemuthianum isolates comprised a major branch and an outlier race (Figure 2.4). The
major branch with a bootstrap value of 88% comprised all races except race 31. Races 23
and 457 formed a cluster with a bootstrap value of 82%, and race 89 with race 2047
formed a cluster supported by a bootstrap value of 75%. The branches combining all
other races showed bootstrap values of less than 60%. The phylogenetic tree constructed
using the Fitch-Margoliash method is comparable to the tree constructed based on N-J
method (Figure 2.5). The outlier race 31 exhibited major differences from the cluster
formed by all other isolates.

Parsimony analysis showed similar results to both the NJ and Fitch-Margoliash
methods. On the basis of a bootstrap value of 100% all isolates formed a monophyletic
group. A subgroup formed by races 457 and 23, also demonstrated by phylogenetic and
parsimony analysis was supported by a bootstrap value of 82% and 70%, respectively.
Race 31, an outlier in the other phenograms generated by the distance methods, was
placed outside the group formed by all other races and was supported by a bootstrap

value of 67%.

74

.3 .m :33? bugs: Samoa Hum—OSLO 05 .3 339:8 was .552: 538.5908 8:80.».8—3. ..
A33 .mofibooéeuamv 839$ .935 05 8 $533: But—82 mound
Assam 8.28 m: .35.:
so €821»: defies gasses ea .32 288 do .Eeasooo eo .Afiee em Agassi we. ease into
.2. m3 -2 .3 m3 .3 .3 m3 -2 .w .5.— -S .n 8m -2 .m 5m .o— .2. 52 .o .52. 52 .w .wm MG -h .hwoN
mo .0 .2 .5 -n deuce 95.» as £5883 gasesaoo ... ._ m 2m .... .3 an -N .3 m2 -_ “238335835 0 .... v.82 .

ooo._ 3
:od ooo._ w—
m—od wood ooo.— 2
:od :od god ooo._ Nu
oNod wood m—od wood ooo.— ~—
m—od :od god wood wood ooo._ o—

ouo .o wNod oNod oNod NNod ouod ooo._

:od wood mood :od god :od wuod oooA

oNod Qod :od :od Sod :od wNod wood ooo._

wood wood wood Cod Sod wood ouod wood Bod coo.—

m—od Cod Sod mood wood wood Nuod Sod Cod :od coo.—

Nmod m—od «Nod :od Sod h — od mmo .o oNod oNod Nuod oNod ooo._

A.Eod wood A...—od wood wood wood Naod Sod Cod Bod wood :od 89—

wood wood wood :od Sod wood wNod wood Cod flood wood oNod Sod coo.—

Zod wood :od wood wood mood «nod wood wood wood wood n—od wood wood nooog
m— w— an Nu 5 on a a b w n w n N .—

F‘NMVW‘OI‘QO‘

moan—o3 $335333 .0 we venue:— Bwoaflaoéuo Heaaoéoxsn 05 .3 33850 305% Ma 3:253 .N.N 03am.

_’. 75

ITSl

ARG81 CTTGTTAAAA TTGGGGGGTT TACGGCAGGA GTGCCCTGCG GGTCCCGGTG CGA
BR89 .................. G. ..............................

BR31 ..................................................

Clag ..... A.... .................... G ...................
CLBlS ..................................................
CR2047 .................. G. ..............................

DR38 ..................................................
MX45? .................. G. ..............................

Mx73 .................................................. ....
PERS .................................................. ....
PER? ..................................................

PERS ..................................................

0823 ..................................................

0865 .................................................. ....
US73 .................. G. .............................. ....

ARG81 GGTGGTATGT TACTA*CGCA AAGGAGGCTC CGCG*AGGGT CCGCCA*CTG TCTT
BR89 .................................................. ....
BR31 ..................................................
Clag .............................. ....G ...............
CLBIS ............... A.... ..............................
CR204? ..................................................
DR38 ..................................................
MX45? ............... A.... .......................... A...
Mx73 .............................. ....GG..TC .GC.AC....
PERS ..................................................
PER? ..................................................
PERS ..................................................
0823 ..................................................
USES .................................................. ....
U873 ................................... G.... .......... ....

ARG81 T*G*AGGGCC CA*CGTCAGC CGTGG*AAGC CCCAA*CGCC AAGCGG*TGC TTGA
BR89 .................................................. ....
BR31 ..................................................
Clag ..................................................
CLBlS ..................................................
CR204? ..................................................
DR38 .T ................................................
Mx457 .T ........ ..A ...................... A.... ..........
MX73 ................................... A.... ..........
PERS ..................................................
PER? .................................................. ....
PERB .................................................. ....
U823 .T.G ............................... A.... ...... G...
0865 ..................................................
US73 ................................... A.... ..........

->
ITSB

ARG81 GTGCGTTCAA AGATTCGATG ATTCACTGAA TTCTGCAATT CACATTACTT ATCG
BR89 .................................................. ....
BR31 ..................................................
Clag ..................................................
CLBlS ..................................................
CR204? ..................................................
DR38 ..................................................
MX457 ..................................................
Mx73 ..................................................
PERS .................................................. ....
PER? .................................................. ....
PER8 .................................................. ....
U823 .................................................. ....
USGS .................................................. ....
US73 .................................................. ....

ARGSl CATTTCGCTG CGTTCTTC*A TC*GATGCCA GAACCAAGAG ATCCGTTGTT AAAA
BR89 .................................................. ....
BR31 .................................................. ....
Clag .................................................. ....
CLBlS ..................................................
CR2047 ..................................................
DR38 ..................................................
MX4S? .................. C. ..C ...........................
MX73 ..................................................
PER? ..................................................
PER8 ..................................................
0823 .................... ..C ...........................
U865 ..................................................
US73 ..................................................

_>

ITSZ
ARG81 GTTTTGATTA TTTGCTTGT* GCCACTC*AG AAGAGACGTC GTTAAAATAG AGTT
BR89 ....... C.. ........................................ ....
BR31 ..................................................
Clag ..................................................
CLBlS ............................................. G....
CR204? ....... C.. .......................... A... ..........
DR38 ..................................... T.. ...*...
MX4S7 ....... C.. ......... T ....... C.. .......... ...T... .
Mx73 ............................ C. ....................
PERS ..................................................
PER? ................................................ *.
PER8 ..................................................
U823 ....... C.. ................. C.. .......... ...*...A..
0865 ....... C.. ...................................... *.
U873 ....... C.. ........................................

7?

ARGBl GGGTTTTCCT *CCGGCGGGC GCCCC*GCGA GCGGGGCC*G GGGGG*AGG* CGGA
BR89 ......................................................
BR31 ....... *.C T .............. C ........... A ................
Clag ....... T.C T ......... T....C ............ C ...... G...G ....
CLBlS ..................................... A ................
CR2047 ......................................................
DR38 .............................. A ....... A ...............
MX4S7 ......................................................
Mx73 ..................................... A ................
PERS ..................................... A ................
PER? .......... T .......................... A
PER8 ..................................... A
0823 .......... T ........ * ....................... C ...... *...
U865 ......................... C ............................
US73 ...... C..C ............................................

0000000000000000

ARGBl CCTCCC*GCC CGCCG*AAGC AACGGTTTGG *TATGTTC*A CAAAGGGTT* ATAGAGG
BR89 .........................................................

BR31 ...... C ............ * .......... TA* ..... C ................ AC
Clag ............... G ........... A.. TA ...... C .......... T ..... AC
CLBlS .........................................................
CR2047 .........................................................
DR38 .............................. T* ................. T * ..... C
MX4S7 .............................. T* ........................ C
Mx73 ........................................................ C
PERS .........................................................
PER? .............. *G..* ........... G ...................... A.AC
PER8 ............................. G .......................... C
U823 .............................. T* ................. T * ..... C
U865 .............. CG..A .................................... AC
US73 .........................................................

Figure 2.3. Sequence alignment (5’ - 3’ direction) of the rDNA internal transcribed
spacer (ITS) sequences and the 5.88 rRNA gene of isolates of C. lindemuthianum
collected in Arg (Argentina), Bra (Brazil), Clb (Colombia), CR (Costa Rica), DR
(Dominican Republic), Mex (Mexico), Per (Peru), and US (United States). The sequence
Clag (Colletotrichum lagenariun) was used as outgroup in both the phylogenetic and
parsimony analysis. (-) indicates identity with Arg81 sequence; (*) indicates an
introduced gap. The sequences of the ITS 1 (positions 1 - 165), the 5.8S rRNA gene
(positions 166 - 329), and ITS 11 (positions 330 - 492) are indicated.

78

 

75 l Bra 89
~ CR 2047

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

US 73
Mex 73
r
_ Clb 15
Per 8
Per 5
Arg 81
DR 38
88
82 US 23
Mex 45?
I Per 7

 

US 65

 

Bra 31

 

 

C. Iagenarium

Figure 2.4. Phylogenetic tree indicating the relationships between isolates of C.
lindemuthiarmm based on sequences of ITSI, ITS2 and 5.8 rRNA. The tree was created
using the neighbor-joining method (NEIGHBOR program in PHYLIP) from distance
values estimated by the Jukes-Cantor one-parameter method (DNADIST program in
PHYLIP). Confidence limits of the branches, indicated above the line, were created in a
bootstrap analysis using 500 replications. Only the bootstrap values above 60% are
indicated. C. lagenarium was used as outgroup.

79

 

US 65

US 23

  
 
   
 

 

Mex 457

 

US 73

CR 2047

 

 

 

Bra 89
L— DR 38
Arg 81
Clb 15
Per 8
_ Per 7
Mex 73
Per 5
- Bra 31

 

C. Iagenarium

Figure 2.5. Phylogenetic tree indicating the relationships between isolates of C.
lindemuthianum based on sequences of ITS l, ITSZ and 5.8 rDNA. The tree was created
using the Fitch-Margolish method (FITCH program in PHYLIP) from distance values
estimated by the Jukes-Cantor one-parameter method (DNADIST program in PHYLIP).
C. lagenarium was used as outgroup.

80
Genetic distances among races of C. lindemuthianum ranged from 0.2% to 2.9%

indicating variable levels of dissimilarity. Sequences did not show pattern parallel to a
specific reaction group (Table 2.2). For instance, the Andean race 38 showed homology
to the other Andean races ranged fiom 1.1% to 1.3%, whereas the homology to the
Middle American races ranged from 0.8% to 2.6%. The Middle American race 73 from
the United States showed homology to the Andean races ranging from 1.3% to 2.0%,
whereas the homology to the other Middle American races ranged from 0.6% to 2.9%.
Isolates of race 73 showed the greatest genetic distance (2.9%) among races of C.
lindemuthianum. Likewise, the intra-race dissimilarity was supported by RFLP-ITS
(Table 2.1) and phylogenetic analyses (Figure 2.4, 2.5, 2.6). Our data suggest that
polymorphism in rDNA is not linked to specific P. vulgaris gene pool In addition,
population structure of C. lindemuthianum is not structured along with the geographic

origin of races.

DISCUSSION

IT S-rDNA has been used to estimate intra- or inter-specific divergence analysis in
several fungi species (Nazar et al., 1991; Lin and Sinclair, 1992;1(usaba and Tsuge,
1995; Bunting et al, 1996; Cooke and Duncan, 1997; Fouly et al, 1997;1(ropp et al,
1997). The high resolution of sequence analysis of the variable regions within the rDNA
has been used to solve taxonomic misclassifications in Colletotrichum species (Sherrifl‘ et
al, 1994; Sreenivasaprasad et al, 1994; Sreenivasaprasad et al, 1996). ITS 1 region

within the rDNA region appears to contain most of the divergence among Colletotrichum

67

100

 

 

 

Clb 15

Per 8
Mex 73
Per 5

US 65

7
C US 23

US 73

 

 

CR 2047

Bra 89

 

Arg 81

DR 38

 

Per 7

Bra 31

P— C. lagenarium

Figure 2.6. Phylogenetic tree indicating the relationships between isolates of C.
lindemuthianum. The tree was constructed by parsimony analysis of ITS], ITS2 and 5.8
rDNA sequences. Confidence limits of the branches, indicated above the line, were
created in a bootstrap analysis using 500 replications. Only the bootstrap values above
60% are indicated. C. lagenarium was used as an outgroup.

82
species and other fungal species (Sreenivasaprasad et al., 1996; Bunting et al., 1996;

Cooke and Duncan, 1997). However, Sherrifi‘ et al. (1994) observed closer relationship of
C. lindemuthianum to C. malvarum, C. orbiculare, and C. trr'folii on the basis of the ITS 2
region and the 28S rDNA sequences. The high resolution of sequences within ITS 2
region allowed identification of more than one species among isolates from C.
gloesporior’des. Analysis of both ITS l and ITS 2 regions might be more informative if
pathogens exhibit high homology within ribosomal genes (Nazar et al, 1991) or in the
case of closely related isolates within a species (Faris-Mokaiesh et al., 1996). Most of the
variability observed in the sequence analyses of 14 isolates of C. lindemuthin was
restricted to the ITS 1 region which exhibited 25% of variability, whereas 8% of
variability was present in the ITS 2 region (Figure 2.3). However, the analysis of the
region comprising the ITS l, 5.8S rRNA and ITS 2 in the C. lindemuthianum isolates
allowed better resolution of some groups than the separate analysis of each ITS region.

Comparison of restriction patterns derived from ITS regions has proved to be a
useful molecular approach to classify fungal species (Fouly et al., 1997). RFLP analysis
of ITS regions in Colletotrichum demonstrated considerable heterogeneity within C.
acutatum isolated fiom difi‘erent hosts (Sreenivasaprasad et al., 1992). Difi‘erences in the
RFLP-ITS banding patterns of isolates of C. nympaheae from Europe compared to
isolates infecting the same host in North America supported the naming of C.
mrpharicola as a new species (Johnson et al, 1997).

Some degree of specialization within C. lindemuthianum that corresponded to the
two host gene pools within P. vulgaris (i. e. Middle American and Andean gene pools)

was demonstrated by virulence analysis and molecular markers, such as RAPDs and

83
RFLP-ITS (Fabre et al., 1995; Pastor-Corrales, 1996; Sicard et al., 1997). However,

previous work did not show separation of races congruent with specific host gene pool
based on virulence and RAPD analyses. Isolates virulent to each gene pool showed high
levels of similarity and they were present in all clusters (Balardin et al., 1997). Our data
for RFLP-ITS analysis showed no cluster formed exclusively by one reaction group.
Andean races clustered predominantly into group 1, although the Andean race 23 was
observed within the group II (Table 2.1). Middle American races were observed in both
groups.

Polymorphism in rDNA-RFLP linked to the geographic origin of isolates has
been suggested among Colletotrichum spp (Sreenivasaprasad et al., 1992; Johnson et al.,
1997). However, sequencing analysis of rDNA of Puccinia spp showed no obvious effect
of geographic origin on clustering of isolates fiom Europe and the United States (Kropp,
et al., 1997). Our data based on RFLP-ITS and sequencing-ITS did not cluster all isolates
from the same country. Bootstrap analysis supported the cluster formed by race 89 from
Brazil and race 2047 from Costa Rica. Likewise, the cluster formed by race 23 from the
United States and race 45? from Mexico was supported by a bootstrap value of 82%
(Figure 2.4). In C. lindemuthiamm, polymorphisms in rDNA appear not to be linked to
the geographic origin of races.

Molecular polymorphism within similar fungal virulence phenotypes has been
reported in some fungal species Kolmer et al (1995) observed a high degree of
molecular polymorphism among isolates that had the same virulence phenotype, in
Puccinia recondita £sp. tritici. The molecular polymorphism within virulence

phenotypes was considered to be a factor that erquained the low correlation between

84
virulence and molecular data. Balardin et a1 (1997) observed RAPD polymorphism

within and among several isolates often different races from different countries. Our data
based on RFLP-ITS or sequencing-ITS analyses supported these findings. Discrete
molecular variations may not be detected by virulence analysis. The distance matrix
value, which reflects the divergence in the ITS region, was the highest between isolates
of race 73 from Mexico and the United States (2.9%). The intra-race polymorphism
observed using either molecular marker suggests a high level of molecular variability
within C. lindemuthianum. Further analyses might indicate if such intra-race variability is
significant for the pathogen population structure.

A monophyletic group of C. lindemuthianum races was observed in both the
parsimony and phylogenetic analyses. On the basis of bootstrap values, similarity was
observed among all races. The only exception was race 31, outlier in the phenograms
generated by the distance methods, and placed outside the group formed by all other
races. This race showed the lowest genetic divergence from C. lagenarium (outgroup
species). In addition, high homology was observed between race 31 and the Andean races
5 and 7 fi'om Peru and 15 from Colombia. Interestingly, race 31 was collected fi'om
Middle American hosts in Brazil, but showed low homology to the Middle American
races 73 fi'om the United States, 89 fi'om Brazil, 457 fi'om Mexico and 2047 from Costa
Rica. Race 31 is one of the few races with equivalent high levels of virulence on both
Andean and Middle American resistance genes.

Variable level of homology among races and polymorphism in the rDNA region
not associated with geographic origin is a clear evidence of the high level of molecular

variability within C. lindemuthianum. In addition, high molecular homology between

85
Andean and Middle American races was indicative of no parallel evolution among C.

lindemuthianum and P. vulgaris. These findings disagree with previous work, which
showed separation of races of C. lindemuthianum congruent to the P. vulgaris gene pools
(Fabre et al., 1995; Sicard et a1, 1997). The intra-race variability emphasized the
limitations of virulence analysis. Taking together, our data showed a level of variability
within C. lindemuthianum higher than that previously characterized by virulence analysis
(Balardin et al., 1997). Long-term resistance might be more efl‘ective if resistance genes
were combined based on their effectiveness against the broader variability of this

pathogen defined by both virulence and molecular analyses.

86
LIST OF REFERENCES

Arora, D. K, Hirsch, P. R., and Kerry, B. R 1996. PCR-based molecular discrimination
of Verticillium chimnydosporium isolates. Mycol. Res. 100:801-809.

Balardin, R. S. and Kelly, J. D. 1997. Re-characterization of Colletotrichum
lindemuthianum races. Ann. Rep. Bean Improv. Coop. 40:126-127.

Balardin, R.S., Jarosz, A. M., and Kelly, JD. 1997. Virulence and molecular diversity in
Colletotrichum lindemuthianum from South, Central and North America.
Phytopathology 87:(in press).

Brmting, T. E., Plumley, K. A., Clarke, B. B., and Hillman, B. I. 1996. Identification of
Magnaporthe poae by PCR and examination of its relationship to other fungi by
analysis of their nuclear rDNA ITS-l regions. Phytopathology 86:398-404.

Cooke, D. E. L. and Duncan, J. M. 1997. Phylogenetic analysis of Phytophthora species
baed on ITSI and ITS2 sequences of the ribosomal RNA gene repeat. Mycol. Res.
101:667-677.

Dowling, T. E., Moritz, C., Palmer, J. D., and Rieseberg, L. H. 1996. Nuclei acids 1]]:
analysis of fragments and restriction sites. Pages 249-320 in: Molecular
Systematics. D.M. Hills, C. Moritz and BK Mable, eds. Sinauer, Srmderland,
MA

Fabre, J. V., Julien, J., Parisot, D., and Dron, M. 1995. Analysis of diverse isolates of
Colletotrichum lindemuthianum infecting common bean using molecular markers.
Mycol. Res. 99:429—435.

Faris-Mokaiesh, S., Boccara, M., Denis, J. B., Denien, A, and Spire, D. 1996.
Differentiation of the “Ascochyta complex” frmgi of pea by biochemical and
molecular markers. Curr. Genet. 29:182-190.

Felsenstein, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap.
Evohrtion 39:783-791.

Fouly, H. M., Wilkinson, H. T., and Chen, W. 1997. Restriction analysis of internal
transcribed spacers and the small subunit gene of ribosomal DNA among four
Gaeumamromyces species. Mycologia 89:590-597.

Hillis, D. M., Mable, B. K, Larson, A., Davis, S. K, and Zimmer, E. A. 1996. Nucleic
acids IV: sequencing and cloning. Pages 321-381 in: Molecular Systematics.
D.NL Hills, C. Moritz, and BK Mable, eds. Sinauer, Sunderland, MA

87

Johnson, D. A., Canis, L. M., and Rogers, J. D. 1997. Morphological and molecular
characterization of Colletotrichum nymphaeae and C. nupharicola sp. nov. on
water-lilies (Nymphaea and Nuphar). Mycol. Res. 101:641-649.

Jukes, T. H., and Cantor, C. R. 1969. Evohrtion of protein molecules Pages 21-132 In:
Mammalian protein metabolism. H. N. Munro, ed. Academic Press, New York

Kolmer, J. A, Liu, J. Q., and Sies, M. 1995. Virulence and molecular polymorphism in
Puccinia recondita £sp. tritici in Canada. Phytopathology 85:276-285.

Kropp, B. R., Hansen, D. R., Wolf, P. G., Flint, K M, and Thomson, S. V. 1997. A
study on the phylogeny of the Dyer‘s woad rust fungus and other species of
Puccinia from crucifers. Phytopathology 87:565-571.

Kusaba, M., and Tsuge, T. 1995. Phylogeny of Altemaria firngi known to produce host-

specific toxins on the basis of variation in internal trascribed spacers of ribosomal
DNA. Curr Genet 28:491-498.

Leung, H, Nelson, R. J., and Leach, J. E. 1993. Population structure of plant pathogenic
fungi and bacteria. Advances in Plant Pathology 10:157-205.

Liu, Z. L. and Sinclair, J. B. 1992. Genetic diversity of Rhizoctonia solam‘ anastomosis
group 2. Phytopathology 82:778-787.

McDonald, B. A 1997. The population genetics of fungi: tools and techniques.
Phytopathology 87:448-453.

Milgroom, M. G. 1995. Analysis of population structure in fungal plant pathogens. Pages
213-229 in: Disease analysis through genetics and biotechnology. J. F. Leslie, and
RA. Frederiksen, eds Iowa State University Press. Ames, IO.

Mills, P. R., Sreenivasaprasad, S., and Brown, AE. 1992. Detection and difl‘erentiation of
Colletotrichum gloeosporioides isolates using PCR. FEMS Microbiology Letters
98: 137- 144.

Nazar, R. N., Hu, K, Schmidt, J., Culham, D., and Robb, J. 1991. Potential use of PCR-
amplified ribosomal intergenic sequences in the detection and difl‘erentiation of
verticillium wilt pathogens. Physiol. and Mol. Plant Pathol. 39:1-11.

Palumbi, S. R. 1996. Nucleic acids 11: the polymerase chain reaction. Pages 205-247 in:
Molecular Systematics. D. M. Hills, C. Moritz, and B. K Mable, eds. Sinauer,
Stmderland, MA.

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

88

Pastor-Corrales, M. A. 1996. Traditional and molecular confirmation of the coevolution
of beans and pathogens in Latin America. Ann. Rep. Bean Improv. Coop. 39:46-
47.

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

Sherrifl; C., Whelan, M.J., Arnold, G. M., Lafay, J. F., Brygoo, Y., and Bailey, J. A
1994. Ribosomal DNA sequence analysis reveals new species groupings in the
genus Colletotrichum. Experimental Mycology 18: 12 1-138.

Smith, J. J., Scott-Craig, J. S., Leadbetter, J. R., Bush, G. L., Roberts, D. L., and
Fulbright, D. W. 1995. Characterization of random amplified polymorphic DNA
(RAPD) products from Xanthomonas campestris: implications for the use of
RAPD products in phylogenetic analysis. Mol. Phyl. Evol 3: 135-145.

Sreenivasaprasad, S., Brown, A E., and Mills, P. R. 1992. DNA sequence variation and
interrelationships among Colletotrichum species causing strawberry anthracnose.
Physiol and Mol Plant Pathol 412265-281.

Sreenivasaprasad , 8., Mills, P. R., and Brown, A E. 1994. Nucleotide sequence of the
rDNA queer 1 enables identification of isolates of Colletotrichum as C.
acutatum. Mycol. Res. 98:186-188.

Sreenivasaprasad , S., Mills, P. R., Meehan, B. M., and Brown, A E. 1996. Phylogeny
ans systematics of 18 Colletotrichum species based on ribosomal DNA spacer
sequences. Genome 39:499-512.

Waalwijk, C., de Koning, J. R. A, Baayen, R P. 1996. Discordant groupings of
F usarium spp from sections Elegans, Liseola and Dlamim'a based on ribosomal
ITSI and ITSZ sequences. Mycologia 88:361-368.

Chapter 3
INTERACTION AMONG VARIABILITY IN COLLETOTRICHUM
LHVDEMUTHIANUM AND DIVERSITY IN PHASEOLUS VULGARIS

ABSTRACT

Thirty-four races of Colletotrichum lindemuthianum from Argentina, Brazil,
Colombia, Costa Rica, the Dominican Republic, Honduras, Mexico, Peru, and the United
States were inoculated on sixty-two cultivars of Phaseolus vulgaris from Brazil, the
Dominican Republic, Honduras, Mexico, and the United States. Bean genotypes clustered
based on the gene pool origin of the resistance genes present, regardless of the actual
gene pool of the host genotype. Further sub-groups of cultivars based on overall level of
resistance within each gene pool, were observed. Races of C. lindemuthianum with
Middle American reaction showed broad virulence on germplasm from both gene pools,
whereas races with Andean reaction showed high virulence only on Andean germplasm.
The reduced virulence of Andean races on Middle American genotypes suggests
selection of virulence factors congruent with diversity in P. vulgaris. In addition, races of
C. lindemuthianum were grouped according to specific gene pool (i e. Middle American
and Andean reaction groups) based on principal component analysis However, the
overlapping of specific races with races from difl‘erent reaction groups might indicate that
this group of isolates possesses factors of virulence to both host gene pools Similar

results from phenetic analyses showed races grouped according to specific gene pool

89

90
Most races with Andean reaction were observed in our cluster B, except races 15 and 23,

which clustered with Middle American races. Only races 38, 39, and 47 from the
Dominican Republic showed high similarity in both multivariate analyses, and clustered
based on geographic effect. Data based on virulence supports variability in C.

lindemuthianum structured with diversity in Phaseolus.

9 1
INTRODUCTION

Anthracnose, caused by Colletotrichum Iindemuthirmum (Sacc. & Magnus) Lams.
—Scrib. is an endemic and more severe disease in Central America than in Andean South
America or in the temperate regions of North America and Europe (Pastor-Corrales et a1,
1994). Yield losses from anthracnose can be as high as 95% (Guzman et al., 1979).
Control of this disease is dimcult due to the efficient seed transmission of the pathogen
(Tu, 1992), lack of cost-efl‘ective chemical controls (Pastor-Corrales and Tu, 1989),
ability of C. lindemuthianum to survive in plant debris up to 22 months (Dillard and
Cobb, 1993), and eventual development of sclerotia (Sutton, 1992). Although the best
approach to controlling this disease is through an integrated pest management regime
(IPM), host resistance seems to be more appropriate in countries where implementation
of [PM measures is not feasible. However, the magnitude of variability in the C.
lindemuthicmum reported worldwide (Andrus and Wade, 1942; Barrus, 1918; Blondet,
1963; Burkholder, 1923; Fouilloux, 1979; Menezes and Dianese, 1988; Garrido, 1986;
Balardin et al, 1990; Waterhouse, 1975), has been the major limitation for developing
durable resistance in commercial germplasm.

The gene-for-gene system regulates the interaction among P. vulgaris and C.
lindemuthianum. Evolving races of C. lindemuthianum continue to overcome specific
resistance sources Prior to the appearance of delta race of C. lindemuthianum, the Co-l
gene was the only resistance source present in navy beans in North America (Tu, 1988).
The Co-2 gene, which has conferred resistance to races 17, 130, 102, 23, 65 and 55, was

the predominant resistance source used in Europe and North America (Mastenbroek,

92
1960;1(ruger et al., 1977; Fouilloux, 1979). However, identification of races 31, 63, and

89 from isolates collected in Brazil and Europe (Kruger et al., 1977; Fouilloux, 1979),
necessitated the identification and characterization of new resistance sources 'Ihree
resistance genes, Co-4, Co-5, and Co-3, fiom a Mexican germplasm collection were
characterized The CM and Co-5 genes conferred resistance to the kappa, iota and alpha-
Brazil races, whereas the Co-3 gene was susceptible to the race alpha-Brazil (Fouilloux,
1979)

Long term resistance based on major genes has been inefi‘ective when resistance
genes are deployed one at a time (Duvick, 1996). For instance, the bean genotypes TO
(Co-4), PI 207262, and Mexico 222 (Co-3), reported to be resistant to European and
North American races (Kruger et al, 1977; Fouilloux, 1979; Schwartz et al., 1982), were
susceptible to Latin American races (Menezes and Dianese, 1988; Rava et al, 1993;
Rodriguez, 1991; Restrepo, 1994; Pastor-Corrales et al, 1995). Isolates from Honduras
overcame resistance in the cultivars TU (Co-5) and AB 136 (Co-6)(Ba1ardin et al,
1997), which have been used as parents in various breeding programs in Latin America.
The development of complementary resistance genes to counter most pathogen variability
(Duvick, 1996; Casela et a1, 1996; Levy et al, 1993), and pyramiding resistance genes
from gene pools other than fiom which the breeding germplasm has been selected was
suggested as an eficient approach to developing durable anthracnose resistance (Young
and Kelly, 1997).

Variability in Uromyces phaseoli (Stavely, 1982; Stavely, 1984; Maclean et al.,
1995), Phaeorlsariopsrs griseola (Guzman et al., 1995), and C. lindemuthianum (F abre et

al, 1995; Pastor-Corrales, 1996; Sicard et al, 1997) congruent with P. vulgaris

93
germplasm diversity in the Middle American and South American Andean gene pools

was suggested as an evolutionary event occurring among host and pathogen populations.
As a result, reciprocal selection of resistance genes in P. vulgaris and virulence genes in
C. lindemuthianum might have occurred (Gepts, 1988). In this study our objective was to
determine the reaction of germplasm from Brazil, Honduras, Mexico and the United
States to a representative group of races of C. lindemuthiarmm from South, Central and
North America. Virulence tests and multivariate analyses were used to demonstrate the
effect of P. vulgaris gene pool on the population structure of the anthracnose pathogen
and the reciprocal effect of the pathogen on the same host cultivars. Breeding strategies

to improve durable resistance to C. lindemuthianum based on these findings is discussed.

MATERIALS AND METHODS

Phaseolus vulgaris germplasm and Colletotrichum lindemuthianum races. Sixty-
two genotypes of P. vulgaris were inoculated with thirty-four races of C.
lindemuthianum. Genotypes were divided into two groups: a) 50 cultivars fiom Brazil,
the Dominican Republic, Honduras, Mexico, and the United States, and b) the 12
anthracnose-difl‘erential cultivars(Pastor-Corra1es, 1991) which were used as control and
inoculated to verify the identity of a particular race. The 34 races of C. lindemuthianum
were identified fi'om isolates collected in Argentina, Brazil, Colombia, Costa Rica, the
Dominican Republic Honduras, Mexico, Peru, and the United States. Identification,

origin and characteristics of the genotypes is shown in Table 3.1. The gene pool of

94

Table 3.1. Identification, origin, gene pool, race, seed weight, and resistance index of
common bean genotypes (P. vulgaris L. ). Genotypes were fi'om Brazil, Colombia, the
Dominican Republic, the Netherlands, Honduras, Mexico and the United States and were
inoculated with 34 races of C. lindemuthianum from Argentina, Brazil, Colombia, Costa
Rica, the Dominican Republic, Honduras, Mexico, Peru, and United States.

 

 

Genotypes' OriginB GI" Racei SW‘ R1'
1 - MDRK US A Nueva Granada 34.9 44
2 - Perry Marrow US A Nueva Granada 44.2 59
3 - Charlevoix US A Nueva Granada 53.0 59
4 - Seafarer US M Meso America 17.2 47
5 - Montcalm US A Nueva Granada 54.0 67
6 - Red Hawk US A Nueva Granada 56.0 56
7 - Widusa Nth M Meso America 15.6 70
8 - Pinto Villa Mex M Durango 37.2 67
9 - Bayo Victoria Mex M Durango 44.0 67
10 - Isles US A Nueva Granada 62.0 76
ll - Kaboon Nth A Nueva Granada 36.1 79
12 - Pompadour Checa 50 DR A Nueva Granada 44.7 67
13 - Ruddy US A Nueva Granada 52.0 56
14 - Cacahuate 72 Mex A Nueva Granada 43.5 82
15 - Chihuahua 21 Mex A Nueva Granada 35.5 6
l6 - Bayomex Mex A Nueva Granada 22.5 6
17 - Isabella US A Nueva Granada 51.0 0
18 - Cardinal US A Nueva Granada 62.0 3
19 - Taylor Horticultural US A Nueva Granada 54.0 3
20 - G 2333 Mex M Meso America 18.8 100
21 - SEL 1308 US M Meso America 24.0 97
22 - Catrachita Hon M Meso America 31.2 94
23 - BAT 93 US M Meso America 17.0 85
24 - SEL 1360 US M Meso America 24.0 97
25 - TO Mex M Meso America 28.4 73
26 - AB 136 Mex M Meso America 23.8 91

 

Table 3.1 (cont’d)

95

 

27 - PI 207262

28 - FM M38

29 - Yeguare

30 - Macanudo

31 - Negro 150

32 - Zacatecas 15
33 - TU

34 - Blackhawk

35 - Newport

36 - Cornell 49242
37 - Mexico 222

38 - Amarillo de Calpan
39 - Rio Tibagi

40 - Amarillo 169
41 - Flor de Mayo
42 - Puebla 36

43 - C-20

44 - Michoacan 8-A
45 - Durango 32

46 - Gilguerillo

47 - Carioca

48 - Criollo Negro
49 - MD 2324

50 - Tio Canela 75
51 - Schooner

52 - Bayo Berrendo
53 - MD 3037

54 - Azufrado

55 - T-39

56 - Negro Durango
57 - Dorado

Mex
Mex
Hon
Bra

Mex
Mex
Mex
US

US

US

Mex
Mex
Bra

Mex
Mex
Mex
US

Mex
Mex
Mex
Bra

Mex
Hon
Hon
US

Mex
Hon
Mex
US

Mex
Hon

22332232323233!ZZZZZZSZZZZZZZZZ

Meso America
Jalisco

Meso America
Meso America
Jalisco
Durango
Meso America
Meso America
Meso America
Meso America
Meso America
Jalisco

Meso America
Jalisco

Jalisco

Jalisco

Meso America
Jalisco
Durango
Jalisco

Meso America
Meso America
Meso America
Meso America
Meso America
Meso America
Meso America
Durango
Meso America
Durango
Meso America

20.2
28.2
19.0
23.3
29.0
14.5
21.2
19.0
25.0
17.7
25.1
27.5
21.0
21.7
22.8
24.0
18.4
27.5
20.0
24.5
18.5
26.8
20.5
17.5
19.0
17.5
22.8
28.0
20.0
32.5
23.0

79
82
79
65
62
67
as
67
59
65
59
47
51
47
23
35
41
65
42
3s
3s
35
38
15
23
17
20
41
23
17
26

 

96

 

58 - Desarrural IR Hon M Meso America 17.0 3
59 - Danli 46 Hon M Meso America 22.8 17
60 - Zamorano Hon M Meso America 20.0 6
61 - FI‘ 83-120 Bra M Meso America 22.2 26
62 - Michelite US M Meso America 16.4 15

 

‘ Seed provided by: Fundacio Estadual de Pesquisa Agropecuaria (FEPAGRO),
Brazil/RS; Escuela Agricola Panamericana Zamorano, Honduras; National Research
Institute for Forestry and Agriculture (INIFAP), Mexico; Centro Intemacional de
Agricultura Tropical (CIAT), Colombia; Bean Breeding and Genetics Program (Michigan
State University), United States.

b Origin of cultivars: Bra (Brazil), Clb (Colombia), DR (Dominican Republic), Hon
(Honduras), Mex (Mexico), Nth (Netherlands), US (United States).

° Andean (A) and Middle American (M) gene pools of P. vulgaris (Singh et al., 1991).

" Races ofP. vulgaris (Singh et al, 1991).

‘ Seed weight (g/ 100 seeds).

f Resistance Index: (total no. of resistance reactions / 34), 34 being the total number of
races used as inoculum in this trial

97
genotypes was inferred based on previous knowledge of host cultivars. Identification,

reaction group and origin of all races is shown in Table 3.2.

Determination of a genotype reaction Inoculum of C. lindemuthianum was
increased in modified Mathur’s medium The culture medium was prepared with dextrose
(8 g.l'1), MgSOa7H20 (2.5 g.l"), KHzPO4 (2.7 gl'l), neopeptone (2.4 g1"), yeast extract
(2.0 g1"), and agar (l6 g1"). Spore suspensions for seedling inoculation were prepared
from purified single-conidial isolates by flooding plates with 5 ml of 0.01% Tween 80 in
distilled water. Afler scraping the culture surface with a spatula, the dislodged spores
were filtered through cheesecloth. The spore concentration was adjusted to 1.0 x 106
sporesml'l with a hemocytometer.

Seeds for each of the sixty-two P vulgaris genotypes were planted in flats
containing Baccto planting mix (Michigan Peat Co., Houston TX), and grown under
greenhouse conditions (16-h day length at 25°C), for 7 to 10 days until seedlings had
reached the full expanded primary leaf stage. Six seedlings were spray-inoculated with
the standardized spore suspension. The suspension of inoculum was applied until runoff
on the stem and to both surfaces of the rmifoliolate leaves Afier inoculation, plants were
maintained in high humidity (>95%) for 48 h at 22 to 25°C. Plants were allowed to dry
and were then transferred to greenhouse benches for 5 days. Disease reaction was rated
seven days after inoculation based on a 1 to 9 severity scale (Balardin et al, 1990). Plants
with no visible disease symptoms or only a few, very small lesions mostly on primary
leaf veins were recorded as resistant (scale 1 to 3). Plants with numerous small or

enlarged lesions, or with sunken cankers on both the lower sides of leaves and the

98
seedling stern, were recorded as susceptible (scale 3.1 to 9). Intermediate reactions (3.1-

6.9) were considered as susceptible for the purpose of this work.

Data analysis. Virulence index of each race of C. lindemuthianum was computed
using the number of cultivars with susceptible reaction over the 62 inoculated genotypes
(15 from the Andean gene pool, 47 from the Middle American gene pool). The resistance
index of each genotype was computed using the number of cultivars, which exhibited
resistance reaction divided by 34. The total number of inoculated races was 34 (Table
3.1).

Principal component analysis was used to derive the variance from severity data
associated with the first three principal components. A correlation matrix of severity data
obtained fi'om inoculation of 34 races of C. lindemuthianum on 62 genotypes was
computed by the INTERVAL program in the Numerical Taxonomy and Multivariate
Analysis System for personal computer (NT SYS-pc) version 1.70 (Exeter Sofiware,
Setauket NY). The eigenvalue and eigenvector matrices were derived fi'om the
correlation matrix by the EIGENVECTOR program (NT SYS-pc). 'Ihe PROJECTION
program (NT SYS-pc) projected t severity data fi'om the severity data matrix onto Ic=3
axes to express the coordinate of each severity datum relative to the axes. The MXPLOT
and MOD-3D programs (NT SYS-pc) were used to generate the three-dimensional
graphics.

The Phenetic analyses of virulence data was based on a data matrix generated by
scoring resistant reaction as 0, and both intermediate and susceptible reactions as l.

Similarity matrices for virulence data were derived with the similarity for genetic data in

99
the SIMGEN program (NT SYS-pc). Genetic differences between individuals within both

P. vulgaris and C. lindemuthianum populations was calculated using the NEI72
coefficient. Polymorphisms were related to both the accumulated number of gene
difl‘erences per locus and the proportion of common genes among individuals within
populations (N ei, 1972). Cluster analysis was performed using the unweighted pair-group
method (UPGMA) in the SAHN program (NT SYS-pc). Phenograms for virulence data

were produced using the TREE DISPLAY program (NT SYS-pc).

RESULTS

Andean races of C. lindemuthianum exhibited more than 70% virulence on
Andean P vulgaris germplasm and less than 30% on Middle American germplasm In
contrast, Middle American races of the pathogen had similar virulence indexes on both
Andean (55%) and Middle American (60%) germplasm (Table 3.2). Middle American
races 31, 87, 357 and 2047 showed a virulence index above 80% on Andean germplasm,
whereas only races 1993 and 2047 exhibited a similar virulence index on Middle
American germplasm.

The principal component analysis of 62 genotypes of P. vulgaris inoculated with
34 races of C. lindemuthianum grouped the germplasm on the basis of gene pool and each
gene pool was subgrouped according to its overall resistance levels to C. lindemuthinaum
(Figure 3.1). Four distinct clusters were observed. Cluster A was the most diffuse and
included genotypes from both gene pools, although the majority of genotypes were from
the Andean gene pool. Cluster B consisted of genotypes exclusively fi'om Andean origin,

whereas clusters C and D were comprised of genotypes from the Middle American gene

 

100

 

 

 

 

 

 

 

 

Ne :: : o c an a: m a :2 36.: a
R :: : c o 2. N: N v s: :55 :
:N :: : MN : 6: a S o: < 8:5 82.5 cm:
:N :: : 2: N N: w 2. :: < 63% 82:5 2:
2. NN N 2: m :N o: 2: m: < £2.23: 5:288 mm
N... NN N 2: N on N: 2: n: < 622:8: 22282: t.
on :: : 2: N 2. ON 2: m: < 2:22;: 5:62:82: 2..
3. o c 2: N N N: 2.. v: < 622.8: 28388 an
no NN N S N 8 ”N 8 N: < 6.8% 82:5 NN
mm NN N 2: : N N: 2: a < 623:3: 52222: a:
8 NN N S N No 6N S N: < 2:26:60 m:
a. :: : S N 3 6: 8 N: < 83m 825 N
N :: : mm : :N o: R w < a»: m
R :: : mm : N N: 2. :: < 86.: n
N c 8 MN : N: N 8 6 < 8:52 N
2. .\. ...z .x. .62 .\. ...z .x. ...z
.49: .9 99: .< .9: sure 22.85.53
:13 .H 2.55.5.5: gang—agony 3am

 

noun— 3:335

 

2326335282.: O we meow: X we 22:: 3:233 as: .9:er 8330: Emma detaomuaovm .N.n 03am.

 

101

ha
mm
cm
cc
8
we
:.
Z.
8
mm
on
mm
2.
2.”
Nn
no
3
an
an

on
2.
en
em
on

3333

mm
NN
mm
mm
mm
mm
NN
mm

NN
NN

NNMNMMMMNMVQ’VQ‘WNWF”

o3

OOOOO

OONOOONOOOOF‘F—‘OOOOOM

ea
ea
an
3.
he
3.
NA.
50
cw
hm
hm
hm
on
on
mm
we
hm
ow
mm

3

«N
3
2..
mm
vm
Nm
wN
um
AN
AN
mm
5
mm
3..
mm
mm
mm

2:
co
cv
ov
— n
_ m
50
cm
no
we
mm
5.
ma
2
5.
mm
5

mm
ow

2222222222222222222

SE 330
3.26:8:
manage:
mange:

8:52
3:52
:35
cows:
:55
3:on
9:on
:55

:55
2:892.
635 2%:
633m 325
fie:

mafia
3.26:8:

2.28: N.m 6::

five"
mam—
abe—
gum
59.
a3.
mmv
5mm
ham
mun
bmn
an
5
5
av
me
3

b“
a

102

.mombonow we \ 89¢on o_£.§8 .«o .2. .83 ”x2...— oo=o_.E> .38. .

.. .m 2%.:

23...... 3.85%... 2 .... . .322... 3.8%.»... ................m 2% .... .3... u 2.3.2.... 3.8%»... .2. 5.9a... 2.2.... % Sn .

.. .m 2%.: 2228 3.8%.»... < n. .2523 3.8.0.»... 2.2%....m <% .... .3... u 2.22... 3.8%.»... 2. 52.3% Sn ..
¢>e x3...— 3=o=5> 295.3 3.85% .

.. ... 2%.: .252...» 2 $18.52...» 22...»...m 2% .... .3... u 52......» ...: §Ea< 2.....2 % Sc .

A. .» 2%.: .2525» < n. 325......» 2......»88 <% .... .3... u a........%» .3 52.5% So ..

.50. ..2.... 8.223 8.23.2. o

2»...— ....» at: 05 8 manage» 35mm»... moan—o». 52.55 0:.th 2... 5.2.5. .953 stoned ..

.3... 2...... ...... 2.5.5. 0. 38 - m2. .8. :2. I an. .8 «an...» I N... .83 352......

. a» ...... .5. - b» .3. a. .. ... .... 8...... . n» ...: ......a... - n» A... ......a. n N» .6. 3.2. u m» .. V... 9.8.. 8%.... u ... ...... 2...... - S
.338— ..35 .3 335.3% 233.6..— »oo.: .38. woibco..8»»$ 88%» 232038.... 52.5 05 .3 .5»...— »oo... .«o 82.3.89.

103

 

 

0 Andean gene pool
I Middle American gene pool

Figure 3.1. Three-dimensional representation of the variance of severity data from
inoculation of 34 races of C. lindemuthianum on 62 germplasm of P. vulgaris associated
with the first three principal components. The X-, Y- and Z-axis are the first, second and
third principal components, respectively. Germplasm was grouped within Cluster A: l to
13; B: 14 to 19; C: 20 to 36; and D: 37 to 62. Identification of cultivars is according to
Table 3.1.

104
pool. The first three principal component axes accounted for approximately 30% of the

total variation within the 34 x 34 data matrix.

In chister A, Andean genotypes included three members of the difl‘erential series,
MDRK, Perry Marrow and Kaboon, along with three contemporary cultivars, Montcalm,
Pompadour Checa and Charlevoix, and two new dark red kidney cultivars, Isles and
Redhawk, bred for resistance to anthracnose. The Middle American cultivars in cluster A
were located close to the more resistant Middle American members of cluster C. These
include the Michigan navy bean cultivar Seafarer, the European garden bean cultivar
Widusa (a member of the differential series), and the two Mexican cultivars Pinto Villa
and Bayo Victoria. The average resistance index of germplasm within cluster A was 55%
(Table 3.1). The Andean cultivar Ruddy, an outlier in this cluster, was located between
cluster A and cluster D. Only Andean genotypes were observed in cluster B. These
included the highly susceptible cranberry bean cultivar, Taylor Horticultural and Cardinal
fiom the United States, Cacahuate 72, Chihuahua 21 and Bayomex fi'om Mexico, and the
kidney bean cultivar Isabella from the United States. Germplasm in this cluster exhibited
a 5.5% level of resistance (Table 3.1). Cluster C was formed exclusively by
Mesoamerican genotypes that were either members of the differential series or
contemporary cultivars bred for resistance to anthracnose. Germplasm within cluster C
exhibited 78% level of resistance (Table 3.1). Some genotypes showed resistance levels
over 80%, such as G 2333 (100%), breeding lines SEL 1308 and SEL 1360 (97%),
Catrachita (94%) and AB 136 (91%). Cultivars G 2333, TO, TU, PI 207262, and Cornell
49242, members of the difl‘erential series, fell within this cluster. Cluster D was formed

by Mesoamerican genotypes, which were either the most susceptible member of the

105
differential series, older landraces, traditional or modern cultivars bred for characteristics

other than anthracnose. In chrster D, genotypes showed a mean level of resistance of 30%
(Table 3.1). Within this group, some traditional cultivars such as Danli 46, Desarrural IR,
Dorado, and Zamorano from Honduras, Carioca and Rio Tibagi fiom Brazil, and the
cultivars Schooner, T-39 and 020 from the United States, were present. The most
susceptible difl‘erential cultivars Michelite and Mexico 222, were also observed within
cluster D.

The principal component analysis of the 34 races of C. lindemuthianum is shown
in Figure 3.2. The first three principal component axes accounted for approximately 20%
of the total variation within the 62 x 62 data matrix. Four distinct clusters are observed.
Clusters A and D were comprised of races exhibiting the Middle American and Andean
reaction, respectively. Cluster B showed a predominance of races with Middle American
reaction along with the Andean races 23 and 15. In contraa, cluster C showed a
predominance of races with Andean reaction along with the Middle American race 8.
Races within the clusters A, B, and C came from Argentina, Brazil, Honduras, Mexico,
Peru, and the United States (Table 3.2), whereas races in cluster D came exclusively from
the Dominican Republic. The average virulence indexes of races within clusters A, B, C,
and D was 55%, 70%, 35%, and 51%, respectively (Table 3.2).

In cluster A, no race was virulent to any Andean resistance source present in the
differential series. However, these races showed 41% of virulence on Andean germplasm
from the Dominican Republic, Mexico, and the United States. Races within clusters B
and C were virulent to resistance sources from both gene pools. Races from these groups

showed a wide range in vimlence (53% to 100%) on all genotypes. In cluster D, races 38,

106

an
3
H
In
G
8
N

130
“7 511 3 102%
ll ' 5
“9 2047 7
ll 19 56

 

 

 

N
L":
3
E
B 3
——I

 

 

 

0 Andean reaction group
I Middle American reaction group

Figure 3.2 Three-dimensional representation of the variance of severity data from
inoculation of 34 races of C. lindemuthianum on 62 genotypes of P. vulgaris associated
with the first three principal components. The X—, Y- and Z-axis are the first, second and
third principal components, respectively. Origin of races: Bra (Brazil), Clb (Colombia),
CR (Costa Rica), DR (Dominican Republic), Hon (Honduras), Mex (Mexico), Per (Peru),
US (United States). Races were grouped within clusters A: 1673Hon, 521Hon, lBra,
81Bra, 9Hon, 89Bra, 73US, 457Mex, 257Mex, 449Mex, 321Mex, 65US, 1993Hon; B:
453Bra, 17Bra, 337Bra, 357Mex, 2047CR, 31Bra, lSClb, 23US, 87Bra; C: 8Per, 5Per,
3Per, 2Mex, 130US, 102US, 7US, 55DR, 19DR; and D: 39DR, 38DR, 47DR.

107
39, and 47 were virulent to all Andean sources present in the difl‘erential series. Races

from Honduras showed only a Middle American reaction, whereas races from the
Dominican Republic showed only an Andean reaction. Races from Brazil, Mexico, Peru,
and the United States were virulent on germplasm from both gene pools.

Phenetic analyses showed three clusters for the host germplasm (Figure 3.3). In
cluster I Middle American genotypes predominate, but two sub-groups of Andean
genotypes were also observed. The differential cultivars Michelite, Cornell 49242,
Mexico 222, and T0 were placed in this group. Genotypes within cluster I corresponded
to those with the lowest level of resistance observed in clusters B and D (Figure 3.1).
Exceptions were the Andean cultivars Montcalm, Redhawk, Charlevoix, and the Middle
American cultivar Seafarer (Figure 3.1), which corresponded to the highest resistance
level within cluster A. Likewise, the cultivars Cornell 49242, Blackhawk, Zacatecas,
Negro 150 and Flor de Mayo M38 which showed high levels of resiaance grouped in the
same region of cluster 1 (Table 3.1). Andean germplasm predominated in cluster 11.
Interestingly, the highly resistant Middle American breeding lines SEL 1308 and SEL
1360 showed similarity to MDRK. In addition cultivars Bayo Victoria, Pinto Villa and
Widusa were placed in cluster 11. Although, these cultivars exhibit characteristics of
Middle American germplasm, the resistance they possess is derived from Andean
sources.

Two major clusters of races corresponding to the Andean and Middle American
reaction groups were observed in Figure 3.4. Clusters I and II were 68% dissimilar. In
cluster I, Middle American races predominated along with the Andean races 15 and 23.

Races with Andean reaction, exclusively formed cluster II. Phenetic analyses showed

108

Figure 3.3 Phenogram of 62 genotypes of P. vulgaris based on virulence data obtained
from the inoculation of 34 races of C. lindemuthianum. The NEI72 coeflicient (SIMGEN
- NTSYS-pc) generated a genetic similarity matrix of virulence data. SAHN program
(NT SYS-pc) estimated the genetic distances using UPGMA

1._2

0.9

109

NEI SIMILARITY INDEX

0.6

0.3 0.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 3.3

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Michelite

110

Figure 3.4. Phenogram of 34 C. lindemuthianum races based on virulence data obtained
fiom the inoculation on 62 genotypes of P. vulgaris. The NEI72 coeficient in the
(SIMGEN - NTSYS-pc) generated a genetic similarity matrix of virulence data. SAHN
program (NT SYS-pc) estimated the genetic distances using UPGMA. The binary
identification of C. lindemuthianum races (Pastor-Corrales, 1991) was followed by the
origin of isolates: Bra (Brazil), Clb (Colombia), CR (Costa Rica), DR (Dominican
Republic), Hon (Honduras), Per (Peru), Mex (Mexico), and US (United States).

0.4

0.3

111

NEI SIMILARITY INDEX

0.2 0.1 0.0

l 1 l

 

 

 

 

 

—:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

‘-

 

 

 

 

 

Figure 3.4

 

 

 

 

 

 

 

 

 

 

 

 

 

l Bra
l7 Bra
89 Bra
9 Hour
73 US
337 Bra
453 Bra
357 Mex
257 Mex
321 Max
65 US
31 Bra
87 Bra
23 US
15 Clb
1993 Hon
2047 CR

l 12
congruence with some sub—clusters in the principal component analyses. Races 38, 39 and

47 showed a tight cluster in both analyses Races 31, 87, 23 and 15, which formed a tight
sub-cluster in cluster B (Figure 3.2) showed 88% similarity in the phenetic analyses. In
contrast, race 8 that clustered along with Andean races in cluster C (Figure 3.2) clustered
among Middle American races in cluster I (Figure 3.4). Multivariate analyses showed

congruence of the 34 races with the P. vulgaris gene pools.

DISCUSSION

Phaseolus vulgaris germplasm carrying different combinations of resistance
genes from sources in Brazil, the Dominican Republic, Honduras, Mexico and the United
States demonstrated a reciprocal influence of variability in C. lindemuthiarmm with the
diversity in P. vulgaris. The genotypes were grouped by multivariate analyses on the
basis of gene pools (Andean and Middle American) and within each gene pool, genotypes
were grouped according to their overall resistance levels to C. lindemuthianum. Cluster A
was the most interesting since it contained genotypes from both gme pools Among the
Andean genotypes were members of the difi‘erential series such as MDRK, Perry Marrow
and Kaboon, and the cultivar Montcalm, known to carry the same Co-I gene as MDRK,
and the cultivar Charlevoix, bred for resistance to the Andean race 130 (Andersen et al,
1963). In addition, Isles (Kelly et al., 1994) and Redhawk (Kelly et al., 1997) bred for
increased levels of resistance to anthracnose and known to possess the Co-I and Co-2
genes, were present. No information on the other Andean cultivar, Pompadour Checa 50

fi'om the Dominican Republic was available. However, its presence within cluster A

1 13
would suggest that it possesses similar high levels of resistance as the other germplasm in

cluster A in contrast to the susceptible Andean germplasm in cluster B. The four Middle
American cultivars in cluster A included Seafarer, known to possess the Andean Co-l
gene, Widusa previously classified as being of Andean origin (Sicard et al, 1997), the
Mexican cultivars Pinto Villa and Bayo Victoria, reported to carry resistance genes from
the Andean parent Canario 101 (Acosta-Gallegos et a1 1995). It would appear that cluster
A is comprised of Andean resistance genes regardless of whether they are present in
Andean cultivars or have been introgressed into Mesoamerican cultivars. Chrster B
comprises highly susceptible Andean genotypes such as Taylor Horticultural and
Cardinal from the United States and Cacahuate (cranberry) from Mexico and the United
States kidney bean Isabella. Isabella does not possess the Co-I gene present in other
kidney bean cultivars such as Montcalm, MDRK and Charlevoix (Young and Kelly,
1996). Genetic differences, confirmed from other studies, are supported by the principal
component analysis in our study. Middle American genotypes with known sources of
resistance, and cultivars bred for resistance, to C. lindemuthianum predominated in
cluster C. In contrast, the Mesoamerican genotypes in cluster D are older landraces,
traditional cultivars, or modern cultivars bred for characteristics other than anthracnose.
For instance, a number of cultivars from Honduras such as Tio Canela 75 and MD 3037
were bred primarily for resistance to Bean Golden Mosaic Virus (Rosas et al., 1997) as
opposed to the cultivar Catrachita (in chrster C) bred for resistance to C. lindemuthianum
(Young and Kelly, 1996). Two members of the difl‘erential series fell in cluster D,
Michelite the rmiversal susceptible cultivar, and Mexico 222 (source of Co-3 gene)

recognized for its limited resistance to C. lindemuthianum (Menezes and Dianese, 1988;

1 14
Rava et al., 1993; Rodriguez-Guerra, 1991; Restrepo, 1994; Pastor-Corrales et a1, 1995).

Other cultivars clustering in cluster D include Carioca and Rio Tibagi, the most widely
grown cultivars in Brazil (Voysest et al., 1994), a number of Mexican cultivars not
recognized as possessing resistance to anthracnose, the susceptible landrace cultivar
Zamorano fi'om Honduras and the cultivars Schooner and C-20 from the United States
bred for high yield (Kelly et al., 1982).

Phenetic analysis showed both cultivars and races grouped congruently to the P.
vulgaris gene pools and groups of genotypes with similar overall resistance were
observed (Figures. 3.3, 3.4). No grouping of germplasm based on geography alone was
observed. Since few sources have been used in breeding for resistance to C.
lindemuthinaum, groups of genotypes possessing similar resistance genes were observed.
For instance, cultivars Charlevoix, Seafarer, Montcalm and Redhawk known to possess
the Co-I resistance gene, grouped in cluster I. In contrast, cultivars Isabella, Chihuahua
21, Bayomex, Taylor Horticultural and Cardinal with no known resistance genes grouped
in a difl‘erent cluster.

Germplasm with resistance to C. lindemuthiarmm may have increased the genetic
diversity among genotypes because genotypes with no resistance exhibited higher
similarity than did those genotypes with resistance to the pathogen (Figure 3.4). Large
genetic distances were observed among clusters. Clusters II and III showed high overall
resistance. In cluster 11 resistance from Andean sources predominated, despite the gene
pool of the genotypes. For instance, the resistance to C. lindemuthicmum present in the

cultivars Seafarer, Pinto Villa and Widusa is from Andean sources despite the fact these

1 15
cultivars are from the Middle American gene pool Cluster 111 showed the largest genetic

dissimilarity in relation to the other clusters.

Congruence among other bean pathogens and host gene pools was reported
(Stavely, 1982;Stave1y, 1984; Guzman et al., 1995; Maclean et al., 1995; Pastor-
Corrales, 1996; Sicard et al, 1997). In all cases, pathogens were divided into two groups
corresponding to the Middle American or Andean gene pools Phenetic analysis indicated
that virulence factors might have been selected according to the predominant host gene
pool within each country (Figure 3.4). Selection of specific virulence factors in Honduras
and the Dominican Republic resulted in the predominance of Middle American and
Andean races in each cormtry, respectively (Table 3.2). For instance, races from
Honduras overcame no Andean resistance genes present in the differential series, whereas
races 38, 39, 47, and 55 from the Dominican Republic overcame all Andean resistance
sources present in the difi‘erential series. Germplasm grown in Mexico and the United
States belongs to both gene pools. Races fi'om these countries overcame resistance
sources from both the Andean and Middle American gene pools (Table 3.2). These races
were observed in all but cluster 1V (Figure 3.4). In contrast, germplasm grown in Brazil
belongs mostly to the Middle American gene pool (Voysest et al., 1994). Nevertheless,
races carrying virulence factors to both host gene pools have been reported consistently
(Meneses and Dianese, 1988;Ba1ardin et al., 1990). For instance, races 31, 55 and 87
overcame Andean resistance sources present in MDRK and Perry Marrow, whereas race
55 also overcame the highly resistant Andean cultivar Kaboon. The principal component
analysis grouped the races fi'om Brazil within cluster A (race 1), cluster B (races 31 and

87), and cluster C (races 55, 102 and 130) confirming the pathogen diversity (Figure 3.2).

l 16
The Andean races, 15 and 23, consistently clustered among Middle American

races in cluster B (Figure 3.2) and cluster I (Figure 3.4). In contrast, race 8, from the
Andean region and virulent to the Middle American Co-2 gene, clustered along with
Andean races in cluster C (Figure 3.2) and with the Middle American races in cluster I
(Figure 3.4). These results suggest the presence of both Middle American and Andean
virulence factors within these races. Similarly, race 31 and 87 virulent to the Andean
resistance sources Co-I and Perry Marrow, have been consistently reported in Brazil on
Middle American hosts (Menezes and Dianese, 1988; Balardin et al., 1990).
Nevertheless, these races were grouped along with Andean races 15 and 23 in cluster B
(Figure 3.2) and showed high similarity with the same races in cluster I (Figure 3.4).
Therefore, categorization of races within a specific reaction group based on origin of
isolates and virulence could be biased if broader virulence would result from adaptation
of some C. lindemuthianum races to both host gene pools. Our virulence data would
suggest that the 34 races appear to effectively identify the presence of resistance genes of
distinct genetic origin. The clusters, generated as a result of the multivariate analysis,

would suggest congruency between the pathogen variability and the host diversity.

Implications on breeding strategy. The large number of virulence factors within
theC. lindemuthianum population overcame resistance in all genotypes but the cultivar G
2333. The combination of resistance genes Co-42, Co-5, and Co-7 appeared to create
greater genetic variability in G 2333 (Young et al., 1997), which remained resistant to all
races in our work. Previous work based on virulence analyses of variability in C.

lindemuthin showed no clustering of races according to the host gene pools of the

l 17
differential cultivars (Balardin et al, 1997). However, the larger set of genotypes in this

study appears to be most distinguishing for virulence factors within races of C.
lindemuthianum. Difl‘erent genes for virulence and resistance in both Middle America
and Andean groups might have been selected as a result of host divergence (Gepts, 1988).
Differences among groups of races specialized in a specific host gene pool seem to be
derived from the adaptive process of races to hosts. Pyramiding genes based on a
knowledge of the complementary effect of large numbers of genes has been suggested as
an approach to increase the longevity of resistance under field conditions (Duvick, 1996).
For instance, the incorporation of Andean resistance genes in bean breeding populations
in Honduras could result in more durable resistance within the country. Similarly,
incorporation of Middle American resistance genes in germplasm in the Dominican
Republic also would result in more durable resistance. In contrast, the presence of
genotypes from both the Andean and Middle American gene pools in a country might
have led to the selection of broader virulence in C. lindemuthinaum than in those
countries where only one or the other gene pools existed. In these countries the use of
resistance genes, based on their complementary action to the races currently present,

would be necessary to exclude most of the pathogen variability.

l 18
LIST OF REFERENCES

Acosta-Gallegos, I. A, Ochoa-Marquez, R., Arrieta-Montiel, M. P., Ibarra-Pérez, F .,
Ravelero A. P., and Valdez, L S. 1995. Registration of ‘Pinto Villa’ common bean.
Crop Sci. 35:1211.

Andersen, A L., Adams, M. W., and Whitford, G. 1963. Charlevoix anthracnose-
resistant dark red kidney bean. Mich. Agric. Exp. Stn Res. Rep. 6: 1-3.

Andrus, C. F., and Wade, BL. 1942. The factorial interpretation of anthracnose
resistance in beans. Pages 1-29 in: Tech. Bull. No. 310. USDA. Washington.

Balardin, R. S. and Kelly, J. D. 1997. Re-characterization of Colletotrichwn
lindemuthianum races. Ann Rep. Bean Improv. Coop. 40:126-127.

Balardin, RS, Jarosz, A M., and Kelly, JD. 1997. Virulence and molecular diversity in
Colletotrichum lindemuthianum from South, Central and North America.
Phytopathology 87:(in press).

Balardin, R. S., Pastor-Corrales, M.A, and Otoya, M.M. 1990. Variabilidade patogénica
de Colletotrichum lindemuthianum no Estado de Santa Catarina. Fitopatol. Bras.
15:243-245.

Barrus, M. F. 1918. Varietal susceptibility of beans to strains of Colletotrichum
lindemuthianum (Sacc. & Magn.) B. & C.. Phytopathology 8:589-605.

Blondet, A 1963. L'anthracnose du haricot: etude des races physiologiques du
Colletotrichum lindemuthianum. PILD. these, Faculté de Sciense, Paris, France.

Burkholder, W. H. 1923. The gamma strain of Colletotrichum lindemuthianum (Sacc. et
Magn.) B.et C.. Phytopathology 13:316-323.

Casela, C. R., Ferreira, AS., Zeller, K.A., Levy, M. 1995. Pathotype variation in the
sorghum anthracnose fungus: a phylogenetic perspective for resistance breeding.
Pages 257-288 in: Disease analysis through genetics and biotechnology. J. F. Leslie
and R. S. Frederiksen, eds. Iowa State University Press Ames, IO.

Dillard, HR. and Cobb, AC. 1993. Survival of Colletotrichum lindemuthianum in bean
debris in New York State. Plant Dis. 77:1233-1238.

Duvick, D.N. 1996. Plant Breeding, an evolutionary concept. Crop Sci. 36:539-548.
Fabre, J. V., Julien, J., Parisot, D., and Dron, M. 1995. Analysis of diverse isolates of

Colletotrichum lindemuthiamm infecting common bean using molecular markers.
Mycol. Res. 99:429-435.

119

Fouilloux, G. 1979. New races of bean anthracnose and consequences on our breeding
programs. Pages 221-235 in: Disease of tropical food crops. H. Maraite and J.A
Meyer, eds. Louvain-la-Neuve, Belgium.

Garrido, E. R 1986. Identificacion de razas fisiologicas de Colletotrichum
lindemuthianum (Sacc. & Magn.) Scrib. en Mexico y busqueda de resisténcia
genética a este hongo. MSc thesis, Institucion de Ensenanza y Investigacion en
Ciencias Agricolas, Montecillos, Mexico.

Gepts, P. 1988. A Middle American and an Andean common bean gene pool Pages 375-
390 in: Genetic resources of Phaseolus beans; their maintenance, domestication,
and utilization. P. Gepts, ed Kluwer. London.

Guzman, P., Donado, M. R, and Galvez, G. E. 1979. Perdidas economicas causadas por
la antracnosis del fiijol Phaseolus vulgaris en Colombia. Turrialba 29:65-67.

Guzman, P., Gilbertson, RL., Nodari, R, Johnson, W.C., Temple, S.R, Mandala, D.,
Mkandawire, ABC, and Gepts, P. 1995. Characterization of variability in the
fungus Phaeisariopsis griseola suggests coevolution with the common bean
(Phaseolus vulgaris). Phytopathology 85:600-607.

Kelly, J. D., Adams, M. W., Saettler, A W., Hosfield, G. L., and Ghaderi, A 1984.
Registration of C-20 navy bean. Crop Sci. 24:822.

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

Kelly, J. D., Hosfield, G. L., Varner, G. V., Uebersax, M. A, Long, E. A, and Taylor, J.
1994. Registration of 'Isles' dark red kideny bean. Crop Sci. 34: 1407-1408.

Kelly, J. D., Hosfield, G. L., Varner, G. V., Uebersax, M. A, Long, E. A, and Taylor, J.
1997. Registration of 'Redhawk' dark red kidney bean. Crop Sci 37 (in press).

Kruger, J., Hofinann, G.M., and Hubbeling, N. 1977. The kappa race of Colletotrichum
lindemuthianum and sources of resistance to anthracnose in Phaseolus beans.
Euphytica 26:23-25.

Levy, M., Correa-Victoria, J., Zeigler, RS., Xu, 8., and Hamer, J. 1993. Genetic diversity
of the rice blast fungus in a disease nursery in Colombia. Phytopathology
83: 1427-1433.

Maclean, D. J., Braithwaite, K.S., Irwin, J.AG., Manners, J.M., and Groth, J.V. 1995.
Random Amplified Polymorphic DNA reveals relationships among diverse
genotypes in Australian and American collections of Uromyces appendiculams.
Phytopathology 85:757-765.

120

Mastenbroeck, C. 1960. A breeding program for resistance to anthracnose in dry shell
haricot beans, based on a new gene. Euphytica 9: 177-184.

Menezes, J. R, and Dianese, JG 1988. Race characterization of Brasilian isolates of
Colletotrichum lindemuthianum and detection of resistance to anthracnose in
Phaseolus vulgaris. Phytopathology 78:650-655.

Nei, M. 1972. Genetic distance between populations. The American Naturalist 106283-
292.

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

Pastor-Corrales, M. A 1996. Traditional and molecular confirmation of the coevohrtion

of beans and pathogens in Latin America. Ann. Rep. Bean Improv. Coop. 39:46-
47.

Pastor-Corrales, M. A, Erazo, O.A, Estrada, E. 1., and Singh, S. P. 1994. Inheritance of
anthracnose resistance in common bean accession G 2333. Plant Dis. 78:959-962.

Pastor-Corrales, M. A, Otoya, M.M., and Molina, A 1995. Resistance to Colletotrichum
lindemuthianum isolates from Middle America and Andean South America in
difi‘erent common bean races. Plant Dis. 79263-67.

Pastor-Corrales, M. A, and Tu, J.C. 1989. Anthracnose. Pages 77-104 in: Bean
production problems in the tropics. H. F. Schwartz, and Pastor-Corrales, M.A,
eds. CIAT. Cali, Colombia.

Rava, C. A, Molina, J., Kauflinann, M., and Briones, I. 1993. Determinacion de razas
fisiologicas de Colletotrichum lindemuthianum en Nicaragua. Fitopatol. Bras.
18:3 88-39 1.

Restrepo, S. 1994. DNA polymorphism and virulence variation of Colletotrichum
lindemuthianum in Colombia. M. Sc. thesis. Universite Paris IV, Paris-Grignon,
France.

Rodriguez, R 1991. Identificacion de razas patogenicas de Colletotrichum
lindemuthianum (Sacc. y Magn.) Scrib. en el estado de Durango mediante un
sistema propuesto intemacionalmente y respuesta de genotipos de fiijol tolerantes
a sequia a razas del patogeno. M. Sc., Parasitologia Agricola, Universidad
Autonoma Agraria "Antonio Narro", Buenavista, Mexico.

Rosas, J. C., Varela, O. 1., and Beaver, J. S. 1997. Registration of 'Tio Canela-75' small
red bean (Race Mesoamerica). Crop Sci. 37: 1391.

121

Schwartz, H. F., Pastor-Corrales, M.A, and Singh, S. 1982. New sources of resistance to
anthracnose and angular leaf spot of beans (Phaseolus vulgaris). Euphytica
3 1 :7 4 1-7 54.

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

Singh, S. P., Gepts, P., and Debouck, D. 1991. Races of common bean (Phaseolus
vulgaris, Fabaceae). Econ. Bot. 452379-396.

Stavely, J.R 1982. The potential for controlling bean rust by host resistance. Ann. Rep.
Bean Improv. Coop. 25:28-30.

Stavely, J .R 1984. Pathogenic specialization in Uromyces phaseoli in the United States
and rust resistance in beans. Plant Dis. 68:95-99.

Sutton, BC. 1992. The genus Glomerella and its anamorph Colletotrichum. Pages 1-26
in: Colletotrichum-Biology, Pathology and Control. J.A Bailley and M.J. Jeger,
eds. C.AB. International, Wallingford, UK.

Tu, J.C. 1988. Control of bean anthracnose caused by delta and lambda races of
Colletotrichum lindemuthianum in Canada. Plant Dis. 72:5-7.

Tu, J.C. 1992. Colletotrichum lindemuthianum on bean. Population dynamics of the
pathogen and breeding for resistance. Pages 203-224 in: Colletotrichum-Biology,
Pathology and Control. J.A Bailley and M.J. Jeger, eds. C.AB. International,
Wallingford, UK.

Voysest, 0., Valencia, M. C., and Amexquita, M. C. 1994. Genetic divesity among Latin
American Andean and Mesoamerican common bean cultivars. Crop. Sci.
34:1100-1110.

Waterhouse, W. L. 1955. Studies of bean anthracnose in Australia. Proc. Linn. Soc.
N. S.W. 80:71-83.

Young, R A, and Kelly, JD. 1996. Characterization of the genetic resistance to
Colletotrichum lindemuthianum in common bean difl‘erential cultivars. Plant Dis.
80:650-654.

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

Young, RA, Melotto, M., Nodari, R0, and Kelly, JD. 1997. Marker assisted dissection
of oligogenic anthracnose resistance in the common bean cultivar, G2333. Theor.
Appl. Genet. (in press)

GENERAL CONCLUSIONS

1. Virulence and molecular analysis showed high genetic variability within
Colletotrichum lindemuthianum. The greatest variability was observed in Central

America, decreasing towards North and South America.

2. The characterization of races based on the difl‘erential series appear to be limited
because of unequal number of genotypes fi'om each P. vulgaris gene pool. In addition,
presence of genotypes with resistance governed by two or more resistance genes might
bias the interpretation of the gene-for-gene interaction in the C. lindemuthianum — P.
vulgaris pathosystem Taking together these findings support the lack of congruence

between populations of C. lindemuthianum and the host gene pools.

3. Grouping races based on virulence to resistance sources from a specific host gene pool
did not truly represent its virulence. The Andean races 15 and 23, virulent on Andean
sources, clustered consistently among Middle American races In contrast, the Middle
American race 31 virulent to hosts from both the Andean and Middle American gene
pool grouped along with Andean races and was collected in Brazil only from Middle
American hosts. Similarly, the geographic origin of races was not a consistent factor
categorizing races. Race 8, virulent to the Middle American Co-2 gene, was collected in

an Andean country and showed higher levels of similarity to Andean than Middle

122

123
American races. Combining virulence and molecular data appears to give more consiaent

support to race categorization into a qrecific reaction group (i. e. Middle American or
Andean) than assigning an origin for individual races based on the host genotype fiom

which the race was initially collected.

4. Parallel evolution among variability in C. lindemuthianum and diversity in Phaseolus
was demonstrated by virulence analysis based on a large set of host genotypes. Pathogen
population was structured according to the bean gene pools Likewise, the gene pool of
resistance genes was the factor, which grouped bean genotypes regardless of the gene

pool of the genotype, however, grouping of races did not follow a geographic pattern.

5. Molecular analysis based on RAPDs, ITS-RFLP, and sequencing ITS showed no clear
congruence among variability in C. lindemuthianum and diversity in Phaseolus. The
polymorphism in the regions of the genome analyzed by these molecular markers did not

follow the evolution in Phaseolus.

6. Monitoring variability using virulence and molecular methods appears to be the best
approach to resolve the population structure of C. lindemuthianum. Molecular
polymorphism within similar races illustrates the limitations in the race identification
based only on virulence to the anthracnose difl‘erential cultivars. In addition, poor
sampling of isolates of similar virulence phenotype may underestimate the molecular

variability within this pathogen. It would appear that variability based on virulence

124
analysis is limited and that the more diverse molecular variability in C. lindemuthianum

is a result of sampling a great portion of the pathogen genome.

7. The population structure of C. lindemuthianum based on virulence analysis suggests
pyramiding resistance genes as an efficient breeding strategy for improving durable
resistance in Phaseolus. However, variability detected by molecular markers suggested
that known resistance genes are insufficient to compensate for the broader variability in
C. lindemuthianum. This large variability appears to be resulted fiom such efficient
assexual mechanism of variability that only complex gene combinations based on
knowledge of gene complementarity would provide gernrplasm with lont-term resistance

to C. lindemuthianum.

APPENDIX A

APPENDIX A

Table A1 Race designation and origin of C. lindemuthianum isolates from samples
collected on Phaseolus hosts fi'om different gene pools in Argentina, Brazil, the
Dominican Republic, Honduras, Mexico and the United States

 

 

Racea Isolateb Origin° Host d GPe
1 Bra 5.1 Brazil FT 83120 (black) MA
1 Bra 10.1 Brazil FT 83120 (black) MA
1 Bra 15.1 Brazil FT 83120 (black) MA
2 Mex 15.1 Mexico Cacahuate (cranberry) A

7 Arg 2.1 Argentina Phaseolus aborigenous f A

7 Mex 46.1 Mexico Bayomex (yellow) A

7 US 12.1 United States Isabella (kidney) A

7 US 40.1 United States Cranberry A

7 US 55.1 United States CELRK (kidney) A

7 US 56.1 United States Chinook (kidney) A

7 US 57.1 United States Cranberry A

7 US 58.1 United States Chinabean (red) A

9 Hon 3.1 Honduras California Small White MA
9 Hon 9.1 Honduras Rojito (red) MA
9 Hon 19.1 Choluteca Unknown -

9 Hon 24.1 Honduras Desarrural (red) MA
17 Bra 1.1 Brazil Carioca (carioca) MA
17 Bra 8.1 Brazil Carioca (carioca) MA

 

125

126

Table A1.1 (cont’d)
17 Bra 12.1 Brazil Turrialba-4 (black)

17 Bra 14.1 Brazil Turrialba-4 (black)

17 Bra 16.1 Brazil Carioca (pinto)

19 DR 2.1 Dominican Republic Pompadour Checa

19 Mex 40.1 Mexico Negro Mexico (black)

5 5 DR 1.1 Dominican Republic Pompadour Checa
65 Bra 3.1 Brazil Rio Tibagi (black)
65 Bra 7.1 Brazil Fl‘ 83120 (black)
65 Bra 11.1 Brazil Macanudo (black)
65 Bra 13.1 Brazil Rio Tibagi (black)
65 Bra 17.1 Brazil Rio Tibagi (black)
65 Bra 19.1 Brazil Rio Tibagi (black)
65 US 1.1 United States Mayflower (navy)
65 US 2.1 United States T-39 (black)

65 US 10.1 United States navy (white)

65 US 13.1 United States T-39 (black)

65 US 23.1 United States T-39 (black)

65 US 28.1 United States Midnight (black)
65 US 30.1 United States navy

65 US 31.1 United States Norstar (navy)

65 US 33.1 United States T-39 (black)

65 US 38.1 United States T-39 (black)

§§§§§§§§§§§§§§E§>§>§EE

 

 

65

65

65

73

73

73

73

73

73

73

73

73

73

73

73

73

73

73

73

73

73

73

73

US 43.1

US 46.1

US 50.1

Bra 2.1

Bra 6.1

Bra 9.1

Hon 4.1

Hon 5.1

Hon 25.1

Hon 26.1

Mex 12.1

Mex 24.1

US 4.1

US 5.1

US 6.1

US 7.1

US 8.1

US 11.1

US 14.1

US 15.1

US 16.1

US 17.1

US 18.1

United States
United States
United States
Brazil

Brazil

Brazil
Honduras
Honduras
Honduras

El Paraiso
Mexico
Mexico
United States
United States
United States
United States
United States
United States
United States
United States
United States
United States

United States

T-39 (black)
T-39 (black)

T-39 (black)

Rio Tibagi (black)
AN 910342 (black)
FT 83120 (black)
Mexico 309 (black)

Frijol negro (black)

Oriente (red)

Oriente (red)

Phaseolus coccineus

T 3144-2 (pinto)

Aztec (pinto)

92T-8056 (small red)
92T-3052 (Great Northern)
B90202 (black)

Blackhawk (black)

Aztec (pinto)

Sierra (pinto)

Alpine (Great Northern)

Red Mexican

Olathe (pinto)

Cahone (pinto)

EEEEEEEEEE

EEEEEEEEEEEE

 

128

 

73

73

73

73

73

73

73

73

73

73

73

73

73

73

73

73

73

73

81

89

89

193

201

U819]
U82Ll
U826]
U829]
U832]
U834]
'U835J
U836]
'US37J
US4L1
U842]
U844]
U852]
U860]
U862]
U863]
‘US661
‘US681
rhglJ

Bnl8J

Bra 20.1

Mex 49.1

Hon 2.1

United States
United States
United States
United States
United States
United States
United States
United States
United States
United States
United States
United States
United States
United States
United States
United States
United States
United States
Argentina
Brazil

Brazil
Mexico

Hondudras

92T-3055 (pinto)
92T-3051 (pinto)
Avanti (navy)
navy

Schooner (navy)
Huron (navy)

Red Mexican
T3054-1 (pinto)
Red Mexican
Avanti (navy)
Mayflower (navy)
Vista (navy)
Mayflower (navy)
Blackhawk (black)
Blackhawk (black)
Mayflower (navy)
Blackhawk (black)
Vista (navy)

black

Fl‘ 85198 (black)
BRA 410 (black)
Bayo Madero

Unknown

EEEEEEEEEEEEEEEEEEEEEE

 

129

 

209

256

257

320

321

321

321

337

357

357

357

448

448

448

448

448

449

449

449

449

449

449

449

Mex 45.1

Mex 53.1

Mex 10.1

Mex 52.1

Mex 7.1

Mex 21.1

Mex 26.1

Bra. 4.]

Mex 25.1

Mex 44.1

Mex 50.1

Mex 1.1

Mex 2.1

Mex 4.1

Mex 47.1

Mex 55.1

Mex 3.1

Mex 6.1

Mex 14.1

Mex 16.1

Mex 18.1

Mex 19.1

Mex 20.1

Mexico
Mexico
Mexico
Mexico
Mexico
Mexico
Mexico
Brazil
Mexico
Mexico
Mexico
Durango
Durango
Durango
Mexico
Mexico
Durango
Durango
Mexico
Mexico
Mexico
Mexico

Durango

Flor de Mayo Bajio
Criollo Bayo (cream)
Puebla 493 (black)
Criollo Bayo (cream)
Pinto

Unknown

Negro Puebla (black)
FT 83120 (black)
FM x BAT (red)
Mantequilla de Calpan
Jamapa (black)

Flor de Mayo (pink)
Olathe (pinto)

Pinto

Puebla 36

Bayito Criollo

L 1213-2 (pinto)
Line # 213

CIAT 2

Azufrado (yellow)
Flor de Mayo (pink)
Canario 107 (yellow)

black

EEEEE

EEEEEEEEEEE'

§>§>§

 

130

 

449

449

453

453

457

457

457

457

465

469

521

833

1033

1165

1344

1431

1472

1472

1545

1600

1601

1673

1673

Mex 22.1

Mex 39.1

Mex 5.1

Mex 11.1

Mex 8.1

Mex 17.1

Mex 23.1

Mex 51.1

Mex 42.1

Mex 38.1

Hon 16.1

Mex 13.1

Hon 15.1

Hon 10.1

Mex 43.1

Mex 48.1

Mex 9.1

Mex 41.1

Hon 12.2

Mex 54.1

Hon 22.1

Hon 1.]

Hon 13.]

Mexico
Durango
Durango
Mexico
Mexico
Mexico
Mexico
Mexico
Mexico
Mexico
Honduras
Mexico
Honduras
Honduras
Mexico
Mexico
Mexico
Mexico
Honduras
Mexico
Honduras
Honduras

Honduras

Cacahuate (cranberry)
Negro (black)
Unknown

URG 4516

Queretaro 34

URG 3252

black

Garbancillo (brown)
Negro Criollo (black)
Negro (black)

Vaina Blanca
Unknown

Desarrural (red)
Unknown

Bayo Blanco

Frijol Negro (black)
Bayo Madero (cream)
Azufrado Tapatio
Frijol Enredadera
Criollo Negro (black)
Phaseolus vulgaris (wild)
Frijol Negro

Guatemalita

§>

EEEEEEEE'

E

EEEEEEEEE

 

131

 

1677 Hon 27.] Honduras DOR 304 (red) MA
1741 Hon 6.] Honduras Catrachita (red) MA
1929 Hon 18.] Honduras Unknown -

1993 Hon 8.] Honduras 11 - 4 MA
1993 Hon 11.1 Honduras Vaina Blanca MA
1993 Hon 20.1 Honduras Unknown -

1993 Hon 23.] Honduras Yeguare (red) MA

 

aRace identification according to binary nomenclature system (Pastor-Corrales, 1991)
b Identification of isolates in the permanent collection of the Bean Breeding and Genetics
Laboratory at the Michigan State University, Crop and Soil Sciences Department
considering the country where samples were collected followed by the entry number.
° Origin: Arg (Argentina), Bra (Brazil), DR (Dominican Republic), Hon (Honduras), Mex
(Mexico), and US (United States).

Host seed class and/or color in parenthesis.
'” Host Gene Pool: A - Andean; MA - Middle American.
f“ Poroto del zorro (wild Andean).

APPENDIX B

APPENDIX B

Table B 1. Anthracnose differential series and the binary number of each cultivar.

 

 

Cultivar Binary number'
Michelite 1
Michigan Dark Red Kidney 2
Perry Marrow 4
Cornell 49242 8
Widusa 16
Kaboon 32
Mexico 222 64
PI 207262 128
TO 256
TU 5 12
AB 136 1024
G 2333 2048

 

' Binary number: 2“, being 11 equivalent to the place of the cultivar within the series. The
sum of cultivars with susceptible reaction will give the binary number of one race.
Ex: race 17, virulent on Michelite (1) and Widusa (16).

132

APPENDIX C

APPENDIX C

Contruction of a SCAR PCR-based marker for differentiation among Andean and
Middle American races

Andean Middle American

 

abcdefghijklmnopqrstuvxy
I

0.60 kb

 

Figure C. 1. Randomly amplified polymorphic DNA (RAPD) amplicons obtained with
primer 2 of Operon kit G for 24 single-spore isolates of Colletotrichum lindemuthianum:
a- 2 (Peru), b- 3 (Peru), c- 5 - (Peru), d- 7 (Peru), e- 38 (Dominican Republic), f- 55
(Dominican Republic), g- 15 (Colombia), h- 102 (Brazil), i- 130 (United States), j- 23
(Brazil), k- 87 (Brazil), 1- 31 (Brazil), m- 17 (Brazil), n- 65 (Brazil), 0- 8] (Brazil), p-
89(Brazil), q- 453 (Mexico), r- 73 (Mexico), s- 449 (Mexico), t- 457 (Mexico), u- 1673
(Honduras), v- 1993 (Honduras), x- 20] (Honduras), y- 2047 (Costa Rica). The
polymorphic amplicon indicated by an arrow distinguished among Andean and Middle
American races. The construction of a SCAR PCR-based marker was attempted isolating
this band fi'om a 0.005% EtBr-stained agarose gel under UV light after electrophoresis.
The isolated PCR product was cloned into Escherichia coli competent cells using the
vector pCR® 2.1 (Invitrogen Co., Carlsbad, CA). The fragment was sequenced using the
fluorescent dye dideoxy nucleoside triphosphate terminator method (MSU-DNA
Sequencing Facility, East Lansing MI). Sequencing reactions were run on a
polyacrylamide gel using the ABI 373A DNA Sequencer. Primers were constructed
considering the primer 2 of Operon kit G as starting sequence. No polymorphism was
observed using 22-, 24 and 26-base primers.

133

APPENDIX D

APPENDIX D

Table D. 1. Reaction of 62 genotypes of P. vulgaris to 20 races of C. lindemuthianum.

 

Racesc

US

Mex

Bra

 

 

GP'Arg

Germplasm'

81 l 17 31 87 89 337453 2 257 321357449457 7 23 65 73 102130

 

37
53
21

RRRSSSRR

I

I

RSRSS

ISSII

1

MARSSSRSR
MAR

Carioca

RSSSRR

RSSRSSSSRS

RR

-5

I

Rio Tibagi

ISRI
IRRR
SSR
RSSSS

SSRSSSSSRSSSRR

SSRSSSSSSSSSR

I

I
I

R

I
I

I

FT 83-120

74
26

IRRRRSSRR

MARR I

Macanudo

Danli 46

~

I
I

I

188811
I

I
I

SSSSRSS

I
I
I
I

34

S

I

88888
888

SSS
MARRRRRRRRRRRRRRRRRRRR

MARRRR

I
I
I

Desarrural IR
Dorado

32

I
I

16
100

I

Tio Canela 75
Catrachita

100
21

RRRRRRRSSRRRRRR

I

Yeguare

SSRSSSSSSSSSRR

I
I

1888
RS

I
I
I

MD 3037

S S S S I I I I I S S I I 10
I I I S S S R R 42
I R S S

I

I

S

I

Zamorano

MD 2324

I SS
RSSSSR

RR

RSSSSS

MARRSRSR

53

I

I

I

Criollo Negro

 

 

Table D.] (cont’d)

Amarillo 169

RSSSRRRSSSRR 58

I

R

I

IRSSS

MA

 

6
58
58
63

R

I
SRSSSSSRSSSRR

I
I

SRSSSSRSSSSSSSSS
I

8888

SRS
MARRSRS
MARRSR

I
I
I

MA
MA

Negro Durango
Gilguerillo
Puebla 36

R

IRSSSSSISSSI
RRRRRRRRSRRRR

I
I

RR

R

Bayo Victoria

68

R

I

RSSSISRSI
SRSSSSSR

I

I

Amarillo de Calpan

Durango 32
Azufrado

S S R 42

I

SSRRR

I
I

I

8888

I
I

R

I

MARR

42

SRR

SSSSSSSRRRSSRR

MA

32
63

1888881

188
I

S
S

I
I

Flor de Mayo

Pinto Villa

RRRR

SRRRRRRRR

MARRS

135

89
95

RRRRR
SRRRRRR

RSRSSSSRRSSSRR

S

MARRRRRRRSRSSSSSI
MARRRRRRRRRSRR
MA

MA

Zacatecas 15
FM M38

1

42

21

I
I

1888
RR

I

Michoacan 8-A

RSSSSSRSSSRR

R888

1
I

Bayo Berrendo

Negro 150

79
37
58
32
63

IRRR
IRRR

R
S
I

I
I
I
I

R

I

I I

RRSSRRSRRRRSS

MARRR

MA

I

Seafarer
C-20

ISSRISSSS SSRR
SSRR

I

SRR

S
RSSSSSRSSSSS

I

MARSS

MA

I

I

I

Schooner

I
I

IRRRSSSR
RRRR

IRR

I

MARRR

Newport

79

SRR

I

I

I

I

MARRRRR

Blackhawk

 

 

SSSSISSSR126
100

I

SSSSSSR
MARRRRRRRRRRRRRRRRRRRR

I

I

MA

T-39

SEL 1308
SEL 1360

100

MARRRRRRRRRRRRRRRRRRRR

MARRRR

100
16
68

RRRRRRRRRRRRRRR
68

I

BAT 93

SSSSSSSSRSSSSSSSSSRR

MARRRSRSRRRRRRRSRRRSRR

MA
MA
MA

Michelite

Cornell 49242

Widusa

SRSSSSSRRRRRRRRSRRRR

SRRRSSSSRRSSSSRRSSSR

84

Mexico 222
P1 207262
TO

100

MARRRRRRRSRRRRSSRRRRRS

100

MARRRRRRSSRSSSSSRRRRRR

136

100
100
100

61

MARRRRRRRRRRRRRRRRRRRR

MARRRRRRRRRRRRRRRRRRRR

AB 136

MARRRRRRRRRRRRRRRRRRRR

G 2333

I

IRR

R

I
I

SRRSRRI
ssss

R

I
S

R

IRRRI

A
A
A
A
A

Pompadour Checa 50

Chihuahua 2]
Bayomex

10
10
16

8888
31

I S

I

I

I
I
I

S

I

RRS

888888888 8888

RRSSS
RRSSS

SRSSSSRRSSSSSS

Cacahuate 72
Montcalm

Ruddy

S
SRRRSSSRR

I
SSSSSSSSSSSSSS

IRRSSSSSRR

RRRSSRRS

42

SSRSI

I
I

888
8888
RRSSSRRRRRRRRRRRRRRR

I

R

I

0
63

A
A

Isabella
Isles

 

.wfimmmfi 3.5n

6.525% :08 no .833er «82 me 8:85: .33 05 8 meme.— .cN \ AM - 3363.. «summon we .2. 3.5 u x35 353%on

.885 .883 m: 8. .8821»: .Afiamv cm .Afiasgv mi ”88.:
M? :35 Age .33. 2.. 53.6 9 38 - m9. .23 £5.23. - a .3 =8 - 5 .3 85...... - S .3 «Ba: - an .3 2% - S
1 $532 335 .3 wean—Emma“. £333.:— mooaM .223 .moiheoeemamv Eon? age—088°: Ea.— 2: .8 v9.2 moon.— .«e ease—$89

.32— 25» Sow—23x 22:: I <2 gee; 28m 5823. I <«
8.3 was: .3885 88m swsogv acme.— assse s: 8:85 8m 83825 €58 cases.

«gum—wax ow Eamon—:33 e580 30¢on AAAEHZC 83.35 v5 523 Eu 33am:— AocaomoM 1:232 $.56an .eqaaeaaN

2.
3V
NV
m
n
NV
NV

M

MUJUJUJUJM

m

newt/imam

M

MMCDUJMM
MM—‘UJMM

M

M

UJUIUJUIUJUJ

M

mmmmmm

M

HMUJWMM

HMUJmMMM

mmmmmmm

M

MMWWMM
M—‘UJUJMM
Marco—mg.

M

M

mmmmMmM

M

~~mwMM

MMWVJMMM
mmmmmmm

mmmmmmg

M

MMmUJMM

M

MMUJWMM

M

MMMMMM

<

<<<<<<

888845 58.84.. :85 agate .Aoecfimb «5.8083. 838.8 2. 338m 88qu ”.3 83,88 .88 a

£3on
$2.32 Ea.—

MM;

13:.qu
Essa—33.8mm .315.
Ego—~25

:35 3M

APPENDIX E

APPENDIX E

Table E. 1. Reaction of 62 genotypes of P. vulgaris to 14 races of C. lindemuthianum.

 

Races‘

ru‘
°/.
40
47
33
53
7

RMMMM-
gMMmMm
g Scream—m
anagram;—
aMM-‘MM
ee—u—w—ng
g W—MmMM
HM—tMM

 

 

Hon
15 2047 9 5211673 1993 3

5 ...............
e Harma—
Z‘s §§§§§
”a
a.
E

 

FT 83-120
Macanudo
Danli 46

Rio Tibagi

Carioca

138

Desarrural IR

20
14
87
54
27

Dorado

Tio Canela 75
Catrachita

RRRRRRRR
RRRRRRRR

S
S

R

Yeguare

MD 3037

Zamorano
MD 2324

34
14

I

I R R R R

R

Criollo Negro

 

 

Table E.l (cont’d)

Amarillo 169

R R R R R R R R 34

S

8

MA 8 S S S I

Negro Durango

Gilguerillo
Puebla 36

R R R R R R I 14

S
8

MA R S I S S

R R R R R R R R

S
RRRRRRR

73

S

I
R R R R R

Bayo Victoria

27

R

I

I

MA

Amarillo de Calpan

Durango 32
Azufrado

I
R R R R R

R R R R R

I

40

S

14
73

Flor de Mayo

Pinto Villa

1
R R R R R R R

R R R R R R R

S

139

40

I

Zacatecas 15

FMM38

67
27

R R R R R R R R

SRRRS

8

MA

R R R R R R

I

MA

Michoacan 8-A

14
40

Bayo Berrendo

Negro 150

R R R R R R R R

S

60
20

Seafarer
C-20

R R R R R R R R

S

14
54
54

Schooner

I

R R R R

S
S

R
R

Newport

R R R R R R R R

Blackhawk

 

 

20
93

S

RRRRRRRR

RRR
RRRRRRRR

I

T-39

I
I
S

RRRRR

R R R R R

MA
MA

SEL 1308
SEL 1360

93

67

R R R R R R R R

BAT 93

14
60
73

Michelite

RRRS
SRRR

S

R
RRRRRRR

R.RR

Cornell 49242

Widusa

S

R R R R R R R R
R R R R R R R R
R R R R R R R R

R R R R R R R R

S
S

27
53
40

S
S
S
S

Mexico 222
P1 207262

TO
TU

S

R R R
S

140

74
80
100

67

S

S
RRRRRRRRRRRRRR

S
8

EE

RRRRRRRR

8

AB 136

MA

Pompadour Checa 50 A

Chihuahua 21
Bayomex

G 2333

S

RRRRRRRRS

S

R

20
40

Cacahuate 72
Montcalm

Ruddy

53

S

R R R R R R R R S

S

S

Isabella
Isles

I R R R R S 87

R R R R

S

 

6.58% :63 :e 882.58. moo... me 898:: .83 o... v. wfioa .3 2M - 3388.. «5.6.85: 6: 18: u 58:. 856.35

.888... 80.8.8. ...: ...: .38... 8.. .858... 8.. .8... 388 :o 3.838 ...: “moo:

M. .o 8.5 .33 ....3. ...: 5.53 G. 38 - m... .3 3.5-2:... - s. .3 as - .. .3 8:2.» - 8 .3 :55. - n. .3 2...... - 2
1 ”Eato— x35 .3 32.5.8: image...— moo... .29... .m2:h:0..3mamv 808% 2320388 3:3 05 :e :83 m3... we 3.33.39
.2... 0:0» 585:3. 23.2 I <2 ace: 0:0» 5.3.5.. I <8

.85. 3.5 .3825 88m 3:28.28 858.. 8.88: ...: ~58... as: 83825 .358 .885

23.3% o: $363.8:— ebgo 80.on A3575 2330.3... :5. Each ...... 33.6:— goSomoM 3:232 $2.68: .efiueEaN
88888.... -..88... :88: .38... 35.58 338.2»: 838...... 8.. .338: 8.35.. s. 838.. .38. .

8 m m m m : .. 3. ... e. 3. ... .. x e. < .523.
S m m m m .. ... m ... ... ... .. 3. m m < 32:2 PE.
.5 m m m m m : .. m .. ... .. .. m m < 5.5,.
o m m m m m m m . m m m m m m < ......50
o m m m m m m m m m m m m m m < 1.33:3: 3......
S. m m m m m : .. m . .. .. .. m m < 528.25
2. . m m m M. m m . m .. .. .. m m < U.2... .5.

 

"IWflfiflilflflml‘ifl!“fi‘l‘lflflflfiifiifir