<%BANNER%>

Molecular Phylogenetics and Population Genetics of Fusarium Oxysporum F. Sp. Radicis-Lycopersici and Its Management by S...

Permanent Link: http://ufdc.ufl.edu/UFE0041036/00001

Material Information

Title: Molecular Phylogenetics and Population Genetics of Fusarium Oxysporum F. Sp. Radicis-Lycopersici and Its Management by Silicon Amendment
Physical Description: 1 online resource (128 p.)
Language: english
Creator: Huang, Cheng-Hua
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: admixture, ancestor, audpc, bayesian, bioinformatics, compatibility, control, crown, diagnosis, disease, diversity, dna, dynamics, ef, elongation, evolution, factor, fertilizer, flow, fol, forl, forma, fst, fungal, fusarium, gene, genetic, genetics, genotype, identification, igs, inoculation, inoculum, lycopersici, management, mat, mating, mcmc, microsatellite, migration, molecular, mrca, nucleotide, oxysporum, pathogen, pathogenicity, phylogenetics, phylogeny, population, primer, progress, radicis, rate, root, rot, sequencing, silicon, soilborne, speciales, speciation, ssr, substitution, tomato, type, vcg, vegetative, virulence, wilt
Plant Pathology -- Dissertations, Academic -- UF
Genre: Plant Pathology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Fusarium oxysporum f. sp. radicis-lycopersici, the causal agent of Fusarium crown and root rot of tomato, is an important soilborne pathogen. The objectives of this study were to investigate the population genetics of the pathogen, to conduct phylogenetic and mating-type analyses, and to evaluate the effect of silicon (Si) on disease severity. Twenty seven microsatellite loci were acquired from a bioinformatics approach and a microsatellite enrichment procedure. Ten of these 27 microsatellites along with vegetative compatibility group (VCG) assays revealed migration of the pathogen among three main tomato-growing regions in Florida. Hendry County had a higher overall average gene diversity than Manatee and Collier Counties. However, the highest mean number pairwise differences and average gene diversity of either VCG 0094 or 0098 were exhibited in Collier County, suggesting that these two VCGs might have migrated from Collier County or other regions to Manatee and Hendry Counties. VCG 0098 probably diverged from VCG 0094 according to VCG and phylogenetic analyses. Although VCG 0094 is still predominant in Florida, VCG 0098 possesses a higher virulence and increased frequency, suggesting that VCG 0098 may be helpful for screening resistant tomato lines. The complete region of intergenic spacer (IGS) provided more phylogenetic resolution than translation elongation factor (EF-1 alpha) and FOL185, a noncoding microsatellite locus. At least two evolutionary origins were revealed based on these three loci: VCGs 0094, 0098, and 0099 likely originated independent of the other VCGs. Each VCG carried a unique mating-type idiomorph but no perithecia were found after crossing isolates with opposite mating-type idiomorphs, suggesting that other factors may be required for sexual recombination. Si amendment significantly reduced disease severity of the stem due to delaying initial infection of roots and the basipetal movement of the pathogen from infected roots to stems. The increase in the Si content of roots correlated significantly with the reduction of disease severity. This study suggested restricting genotype flow of F. oxysporum f. sp. radicis-lycopersici for disease management, regularly monitoring the population structure of the pathogen, and further studying the application of Si fertilizers for controlling Fusarium crown and root rot in field-grown tomatoes.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Cheng-Hua Huang.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Roberts, Pamela D.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041036:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041036/00001

Material Information

Title: Molecular Phylogenetics and Population Genetics of Fusarium Oxysporum F. Sp. Radicis-Lycopersici and Its Management by Silicon Amendment
Physical Description: 1 online resource (128 p.)
Language: english
Creator: Huang, Cheng-Hua
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: admixture, ancestor, audpc, bayesian, bioinformatics, compatibility, control, crown, diagnosis, disease, diversity, dna, dynamics, ef, elongation, evolution, factor, fertilizer, flow, fol, forl, forma, fst, fungal, fusarium, gene, genetic, genetics, genotype, identification, igs, inoculation, inoculum, lycopersici, management, mat, mating, mcmc, microsatellite, migration, molecular, mrca, nucleotide, oxysporum, pathogen, pathogenicity, phylogenetics, phylogeny, population, primer, progress, radicis, rate, root, rot, sequencing, silicon, soilborne, speciales, speciation, ssr, substitution, tomato, type, vcg, vegetative, virulence, wilt
Plant Pathology -- Dissertations, Academic -- UF
Genre: Plant Pathology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Fusarium oxysporum f. sp. radicis-lycopersici, the causal agent of Fusarium crown and root rot of tomato, is an important soilborne pathogen. The objectives of this study were to investigate the population genetics of the pathogen, to conduct phylogenetic and mating-type analyses, and to evaluate the effect of silicon (Si) on disease severity. Twenty seven microsatellite loci were acquired from a bioinformatics approach and a microsatellite enrichment procedure. Ten of these 27 microsatellites along with vegetative compatibility group (VCG) assays revealed migration of the pathogen among three main tomato-growing regions in Florida. Hendry County had a higher overall average gene diversity than Manatee and Collier Counties. However, the highest mean number pairwise differences and average gene diversity of either VCG 0094 or 0098 were exhibited in Collier County, suggesting that these two VCGs might have migrated from Collier County or other regions to Manatee and Hendry Counties. VCG 0098 probably diverged from VCG 0094 according to VCG and phylogenetic analyses. Although VCG 0094 is still predominant in Florida, VCG 0098 possesses a higher virulence and increased frequency, suggesting that VCG 0098 may be helpful for screening resistant tomato lines. The complete region of intergenic spacer (IGS) provided more phylogenetic resolution than translation elongation factor (EF-1 alpha) and FOL185, a noncoding microsatellite locus. At least two evolutionary origins were revealed based on these three loci: VCGs 0094, 0098, and 0099 likely originated independent of the other VCGs. Each VCG carried a unique mating-type idiomorph but no perithecia were found after crossing isolates with opposite mating-type idiomorphs, suggesting that other factors may be required for sexual recombination. Si amendment significantly reduced disease severity of the stem due to delaying initial infection of roots and the basipetal movement of the pathogen from infected roots to stems. The increase in the Si content of roots correlated significantly with the reduction of disease severity. This study suggested restricting genotype flow of F. oxysporum f. sp. radicis-lycopersici for disease management, regularly monitoring the population structure of the pathogen, and further studying the application of Si fertilizers for controlling Fusarium crown and root rot in field-grown tomatoes.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Cheng-Hua Huang.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Roberts, Pamela D.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041036:00001


This item has the following downloads:


Full Text

PAGE 1

1 MOLECULAR PHYLOGENETICS AND POPULATION GENETICS OF Fusarium oxysporum f. sp. radicis lycopersici AND ITS MANAGEMENT BY SILICON AMENDMENT By CHENG-HUA HUANG A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Cheng-Hua Huang

PAGE 3

3 To my grandparents, parents, wife, and son for their support

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank m y major advisor, Dr Pamela Roberts who provided the opportunity for me to conduct my doctoral studies and to comp lete my dissertation under her direction here at the University of Florida. I am grateful for he r guidance, patience, and support in research. I would also like to thank my co-advisor, Dr Lawrence Datnoff for his encouragement and suggestions in assisting me to pass through many difficulties during my study here. My sincere gratitude is extended to my supervisory co mmittee members: Dr. John Scott for providing tomato seeds and inoculation approaches, Dr. R obert McGovern for sharing his isolates and inoculation experience, Dr. Wade Elmer for guiding my vegetative compatibility study, and Dr. Jeffrey Rollins for advising me in phyl ogenetics and molecular techniques. I would also like to thank Drs. Kerry ODonnell, H. Corby Kistler, James Correll, Raymond Schneider, and Jaacov Katan for providi ng isolates, Dr. Monica Ozores-Hampton, Dr. Rosa Muchovej, Brenda Rutherford, and Glades Crop Care, Inc. for collecting isolates, and Norma Flor, Ginger Clark, Drs. Monica Arakaki, Matt Gitzendanner, Doug Soltis, Pamela Soltis, Asha Brunings, Linley Dixon, and Liane Gale fo r guidance in laborator y techniques. I would also like to thank Eldon Philman and Herman Brown for their help in the greenhouse tasks. I would like to express my sincere gratitude a nd appreciation to Steve Hardy at the Baptist Student Center in Gainesville, FL. His intelligent and insightful advice helped me to survive in the United States. I would especia lly like to thank my grandparents, parents, brother and sisters for providing me with love, encouragement, and support. Most importantly, I thank my wife and son for encouraging me and praying for me to co mplete my doctoral degree. Finally, I thank God for His mercy, righteousness and love.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4 LIST OF TABLES ...........................................................................................................................8 LIST OF FIGURES .......................................................................................................................10 ABSTRACT ...................................................................................................................... .............12 CHAP TER 1 LITERATURE REVIEW .......................................................................................................14 Taxonomy ...................................................................................................................... .........14 Vegetative Compatibility Group (VCG) and Population Structure ........................................15 Phylogenetics of F oxysporum f. sp. radicis lyco persici .......................................................17 Silicon and Plant Disease ..................................................................................................... ...19 Hypotheses .................................................................................................................... ..........20 2 POPULATION STRUCTURE OF Fusarium oxysporum f. sp radicis-lycopersici IN FLORIDA INFER RED FROM VEGETA TIVE COMPATIBILITY GROUPS AND MICROSATELLITES ............................................................................................................ 23 Introduction .................................................................................................................. ...........23 Materials and Methods ...........................................................................................................25 Fungal Collections ...........................................................................................................25 Pathogenicity Tests ..........................................................................................................26 Molecular Differentiation of F oxysporum f. sp. lycopersici and radicis lycopersici ...26 Vegetative Compa tibility Tests .......................................................................................27 Development of Microsatellite Primers ........................................................................... 28 Microsatellite Data Collection .........................................................................................29 Microsatellite Data Analysis ...........................................................................................30 Results .....................................................................................................................................32 Pathogenicity Tests ..........................................................................................................32 Molecular Identification of F oxysporum f. sp. radicis lycop ersici ...............................33 Vegetative Compa tibility Tests .......................................................................................33 Development of Microsatellite Primers ........................................................................... 34 Molecular Diversity Indices ............................................................................................ 35 Hierarchical Distribution of Total Gene Diversity ..........................................................36 Linkage Disequilibrium ................................................................................................... 36 Population Admixture ..................................................................................................... 36 Historical Migration and Population Growth Rate ..........................................................37 Discussion .................................................................................................................... ...........37 Pathogenicity Tests ..........................................................................................................38 Molecular Identification .................................................................................................. 38

PAGE 6

6 VCG Frequency ...............................................................................................................39 Microsatellites ............................................................................................................... ..41 Molecular Diversity Indices ............................................................................................ 42 Hierarchical Distribution of Total Gene Diversity ..........................................................43 Linkage Disequilibrium ................................................................................................... 44 Population Admixture and Historical Migration ............................................................. 45 Population Growth Rate .................................................................................................. 45 3 PHYLOGENETIC AND MATING-TYPE ANALYSES OF Fusarium oxysporum f. sp. radicis lyco persici AND THEIR ASSOCIATI ON WITH VIRULENCE ............................. 65 Introduction .................................................................................................................. ...........65 Materials and Methods ...........................................................................................................68 Collection of Isolates .......................................................................................................68 DNA Extraction and Polymerase Chain Reaction ........................................................... 68 Direct Sequencing ...........................................................................................................69 Phylogenetic Analyses .....................................................................................................69 Mating Type Determination ............................................................................................ 70 Crossing ...................................................................................................................... .....71 Virulence Test .................................................................................................................71 Nucleotide Substitution Rates and Population Dynamics ............................................... 72 Results .....................................................................................................................................74 Phylogenetic Analysis .....................................................................................................74 Mating-Type Analyses and Mating .................................................................................75 Virulence ..................................................................................................................... ....75 Nucleotide Substitution Rates and Population Dynamics ............................................... 76 Discussion .................................................................................................................... ...........76 4 EFFECT OF SILICON ON FUSARIUM CROWN AND R OOT ROT OF TOMATO CAUSED BY Fusarium oxysporum f. sp. radicis lycopersici ...............................................88 Introduction .................................................................................................................. ...........88 Materials and Methods ...........................................................................................................89 Plant Growth and Silicon Amendment ............................................................................ 89 Inoculum Production and Inoculation Procedure ............................................................90 Disease Assessments ....................................................................................................... 91 Effects of Silicon and Inoculum Concentration .............................................................. 91 Effect of Silicon on Disease Progress over Time ............................................................ 92 Determination of Dry Root and Shoot Weight and Silicon Quantification ..................... 92 Statistical Analysis .......................................................................................................... 92 Results .....................................................................................................................................93 Effects of Silicon and Inoculum Concentration .............................................................. 93 Effect of Silicon on Disease Progress .............................................................................94 Discussion .................................................................................................................... ...........95 5 CONCLUSION .................................................................................................................... .108

PAGE 7

7 REFERENCES .................................................................................................................... ........114 BIOGRAPHICAL SKETCH .......................................................................................................128

PAGE 8

8 LIST OF TABLES Table page 2-1 Tester strains of vegetative compatibility groups of Fusarium oxysporum f. sp. radicis lyco persici and F oxysporum f. sp. lycopersici used in this study ....................... 47 2-2 Isolates used for develo ping m icrosatellite primers ...........................................................48 2-3 Isolates used for studying the population genetics of Fusarium oxysporum f. sp. radicis lyco persici in Florida .............................................................................................50 2-4 Frequency of vegetative compatibility groups (VCGs) of Fus arium oxysporum f. sp. radicis lycopersici sampled from 2006 to 2008 in Florida ................................................ 53 2-5 Characteristics of 27 micr osate llite loci de rived from a bioinformatics approach and a microsatellite enrichment pr ocedure for 13 isolates of Fusarium oxysporum f. sp. lycopersici and 32 isolates of F oxysporum f. sp. radicis lycopersici ..............................54 2-6 Pairwise Fst values and their P values between populations of Fusarium oxysporum f. sp. radicis lycopersici V CG 0094 from three counties in Florida ..................................... 57 2-7 Pairwise Fst values and their P values between populations of Fusarium oxysporum f. sp. radicis lycopersici V CG 0098 from three counties in Florida ..................................... 57 2-8 Mean number of pairwise difference and averag e nucleotide diversity of microsatellite haplotyp es for isolates of Fusarium oxysporum f. sp. radicis lycopersici from three counties in Florida ......................................................................... 58 2-9 Hierarchical distribution of gene diversity am ong populations of Fusarium oxysporum f. sp. radicis lycopersici from Florida ............................................................. 59 2-10 Tests for random association of alleles with in each locus and between pairs of loci in the population of Fusarium oxysporum f. sp. radicis ly copersici from three counties in Florida ............................................................................................................................60 2-11 Pairwise numbers of m igrants per generation inferred from Bayesian analyses implemented in MIGRATE-N ........................................................................................... 60 3-1 Mean pairwise nucleotide sequence diffe rences for each locus using p-distances am ong isolates of Fusarium oxysporum f. sp. radicis lycopersici (FORL), F oxysporum f. sp. lycopersici (FOL), and both formae speciales .......................................81 3-2 Results of partition homogeneity test among three data sets of intergenic spacer (IGS), partial elongation factor 1-alpha (EF-1 ) and a noncoding m icrosatellite locus, FOL185 ........................................................................................................................ ......81 3-3 Data set properties and nucleotide subs titu tion models used in phylogenetic analyses .... 81

PAGE 9

9 3-4 Bayesian Markov chain Monte Carlo (MCMC) estim ates of evolutionary dynamics of Fusarium oxysporum f. sp. radicis lycopersici using the uncorrelated relaxed lognormal clock model im plemented in BEAST ............................................................... 82 4-1 Effect of silicon (Si) on Si concentra tion, dry weight, and Si u ptake of tomato inoculated with Fusarium oxysporum f. sp. radicis lycopersici at the time of inoculation (week 0) and the time of harvest (week 4) ...................................................... 99 4-2 Analysis of variance for effects of sili con (Si) supply and inoculum concentration (IC) on plant components................................................................................................. 100 4-3 Analysis of variance of effects of silicon (Si) supply and inoculum concentration (IC) on disease com ponents ..................................................................................................... 101 4-4 Disease severity of tomato plants ame nded with or without sodium metasilicate (Na2SiO3) and inoculated with Fusarium oxysporum f. sp. radicis lycopersici 4 weeks after inoculation ....................................................................................................101 4-5 Effect of inoculum concentration on disease severity of tom ato plants inoculated with Fusarium oxysporum f. sp. radicis lycopersici 4 weeks after inoculation .............. 101 4-6 Area under disease progress curve (AUDPC ) of tom ato plants amended with or without sodium metasilicate (Na2SiO3) and inoculated with Fusarium oxysporum f. sp. radicis lycopersici ......................................................................................................102

PAGE 10

10 LIST OF FIGURES Figure page 2-1 Map of three Flor ida populations of Fusarium oxysporum f. sp. radicis lycopersici analyzed in this study and distribution of vegetative com patibility groups (VCGs) ......... 61 2-2 Identification of Fusarium oxysporum f. sp. radicis lycopersici (FORL) and Fusarium oxysporum f. sp. lycopersici (FOL) using FORL-specific prim ers sprlf and sprlr and FOL-specific prim ers P12-F2B and P12-R1 ...................................................... 62 2-3 Neighbor-joining tree based on Neis m inimum genetic dist ance between i ndividuals .... 63 2-4 Estimated population structure inferred from multilocus micros atellite data of Fusarium oxysporum f. sp. radicis lycopersici using STRUCTURE according to membership coefficient ......................................................................................................64 3-1 Phylogeny for Fusarium oxysporum f. sp. radicis lycopersici and F oxysporum f. sp. lycopersici inferred from the intergenic spacer (IGS) region of nuclear ribosomal DNA ........................................................................................................................... ........83 3-2 Phylogeny for Fusarium oxysporum f. sp. radicis lycop ersici and F oxysporum f. sp. lycopersici inferred from partial elonga tion factor 1-alpha (EF-1 ) ................................. 84 3-3 Phylogeny for Fusarium oxysporum f. sp. radicis lycop ersici and F oxysporum f. sp. lycopersici inferred from a noncoding microsatellite locus, FOL185 ............................... 85 3-4 Cluster analysis of virulence for ve getative com patibility groups (VCGs) of Fusairum oxysporum f. sp. radicis lycopersici according to Wards minimum variance method implemented in JMP ............................................................................... 86 3-5 Bayesian skyline plots of Fusarium oxysporum f. sp. radicis lycopersici derived from IGS (A), EF-1 (B), and FOL185 (C) data sets using BEAST ................................ 87 4-1 Effect of silicon (Si) on Si concentration (A), d ry weight (B), and Si uptake (C) of the tomato shoot .............................................................................................................. .103 4-2 Effect of silicon (Si) on Si concentration (A), d ry weight (B), and Si uptake (C) of the tomato root ............................................................................................................... ..104 4-3 Effect of silicon on symptom development of Fusarium crown and root rot expressed as root infection (A), dis ease severity of crown (B) and of stem (C) on tomato cultivar Bonny Best over 6 weeks after inoculation ........................................................105 4-4 Relationship between root infection (A), disease severity of cr own (B) an d stem (C) and silicon concentration of tomato root 4 weeks after inoculation with Fusarium oxysporum f. sp. radicis lycopersici ................................................................................106

PAGE 11

11 4-5 Relationship between disease severity of stem and silicon concentration of tomato shoot 4 weeks after inoculation with Fusarium oxysporum f. sp. radicis lycopersici ... 107

PAGE 12

12 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy MOLECULAR PHYLOGENETICS AND POPULATION GENETICS OF Fusarium oxysporum f. sp. radicis lycopersici AND ITS MANAGEMENT BY SILICON AMENDMENT By Cheng-Hua Huang December 2009 Chair: Pamela D. Roberts Major: Plant Pathology Fusarium oxysporum f. sp. radicis-lycopersici the causal agent of Fusarium crown and root rot of tomato, is an impor tant soilborne pathogen. The obj ectives of this study were to investigate the population geneti cs of the pathogen, to conduc t phylogenetic and mating-type analyses, and to evaluate the effect of s ilicon (Si) on disease severity. Twenty seven microsatellite loci were acquire d from a bioinformatics approach and a microsatellite enrichment procedure. Ten of these 27 mi crosatellites along with vegeta tive compatibility group (VCG) assays revealed migration of the pathogen among three main tomato-growing regions in Florida. Hendry County had a higher overall average gene di versity than Manatee and Collier Counties. However, the highest mean number pairwise differences and average gene diversity of either VCG 0094 or 0098 were exhibited in Collier Count y, suggesting that these two VCGs might have migrated from Collier County or other re gions to Manatee and Hendry Counties. VCG 0098 probably diverged from VCG 0094 accordi ng to VCG and phylogenetic analyses. Although VCG 0094 is still predominant in Florida, VCG 0098 possesses a higher virulence and increased frequency, suggesting that VCG 009 8 may be helpful for screening resistant tomato lines. The complete region of intergenic spacer (IGS) provided more phylogenetic resolution than translation elongation factor (EF-1 ) and FOL185, a noncoding microsat ellite locus. At least two

PAGE 13

13 evolutionary origins were revealed based on these three loci: VCGs 0094, 0098, and 0099 likely originated independent of the other VCGs. E ach VCG carried a unique mating-type idiomorph but no perithecia were found after crossing isol ates with opposite mating-type idiomorphs, suggesting that other factors may be requi red for sexual recombination. Si amendment significantly reduced disease severity of the stem due to delaying initial infection of roots and the basipetal movement of the pathogen from infected roots to stems. The increase in the Si content of roots correlated significantl y with the reduction of diseas e severity. This study suggested restricting genotype flow of F oxysporum f. sp. radicis-lycopersici for disease management, regularly monitoring the population structure of the pathogen, a nd further studying the application of Si fertilizers for controlling Fusari um crown and root rot in field-grown tomatoes.

PAGE 14

14 CHAPTER 1 LITERATURE REVIEW Taxonomy Fusarium oxysporum (Sacc.) Snyder & Hans., a cosm opolitan soilborn e haploid fungal plant pathogen, produces spores only through mitosi s and causes vascular wilt or root rot in both agricultural crops and ornamental plan ts (120,121,132). Pathogenic isolates of F oxysporum show a high level of host specificity and are classified into more than 120 formae speciales and races according to the plant species and cultivars in which they are able to cause disease (5). In contrast, those isolates unable to induce symp toms on a given plant species are considered nonpathogens of the given host species (120). Fusarium crown and root rot of tomato caused by F oxysporum Schlechtend:Fr. f. sp. radicis-lycopersici W. R. Jarvis & R. A. Shoemaker, is an important disease on tomato worldwide (74) and the most serious soilborne disease limiting tomato production in southern Florida (27). The disease was firs t identified in Japan in 1969 (146) and thought to be a new race (J3) of F oxysporum f. sp. lycopersici the causal agent of Fusarium wilt on tomato (92,146). In Florida, Fusarium crown and root rot was first noted in 1975 in Palm Beach County but symptoms were thought to be caused by salt damage (159). Jarvis and Shoemaker (72) proposed that the causal agent of Fusarium cr own and root rot is distinct from F oxysporum f. sp. lycopersici based on symptoms, disease development, and physiologica l characteris tics. Genetic difference has been suggested in that F oxysporum f. sp. lycopersici and radicis-lycopersici are vegetatively incompatible, which means the inabi lity to form heterokaryons between these two pathogens (128). Neither the divergence time and or igin nor the existence of physiological races have been described for F oxysporum f. sp. radicis-lycopersici (80).

PAGE 15

15 Formae speciales and races of F oxysporum cannot be identified according to morphological characteristics unless cumbersome pathogenicity tests are carried out (103). However, a particular forma specialis may be assigned preliminarily based on the host from which a pathogenic isolate of F oxysporum was recovered (177). Likewise, without determining pathogenicity, pathogenic and nonpathogenic F oxysporum species cannot be distinguished morphologically. Because nonpathogenic F oxysporum species isolated from a given host may infect other plant species, these nonpathogenic isolates are defined as those that fail to cause disease on a given plant species (51). There is a need to develop rapid, precise molecular techniques for identif ication of a given forma specialis of F oxysporum Recently, specific primers for F oxysporum f. sp. radicis lycopersici were developed based on a comparison of the nucleotide diversity of exo polygalacturonase gene (pgx4 ) among F oxysporum isolates (63). However, the specificity of the primers has not neen evaluated completely for geographically distinct isolates of F oxysporum f. sp. radicis lycopersici Vegetative Compatibility Group (VCG) and Population Structure Vegetative c ompatibility refers to the abil ity of any two isolates to form stable heterokaryons, which are said to be vegetativel y compatible and belong to the same vegetative compatibility group (VCG) (128). VCG tests may be used for studying population genetics of pathogenic fungi (95,96). It is suggested that VCG could be correlated with pathogenicity (79,104,128). However, the correlation may be st rong, weak, or nonexistent depending on each forma specialis (95,96). For example, nonpa thogenic isolates of F oxysporum are not vegetatively compatible with pathogenic isolates of F oxysporum f. sp. cyclaminis (176). In contrast, nonpathogenic isolates of F oxysporum are vegetatively compatible with VCG 0031 but not with VCGs 0030 and 0032 (19). However, th e correlation of VCG w ith pathogenicity in F oxysporum f. sp. radicis-lycopersici has not been addressed.

PAGE 16

16 Although F oxysporum f. sp radicis-lycopersici was newly recognized in 1978 (72), considerable genetic variation has been shown to exist (80,139) A previous study (139) showed that VCG 0094 of F oxysporum f. sp. radicis-lycopersici was predominant in Florida and that isolates of VCG 0094 in Europe probably migrat ed from Florida based on VCG and restriction fragment length polymorphism (RFL P) analyses. Thus, Florida is th e probable center of origin of VCG 0094 although this VCG was firs t described in Belgium (139). A higher genetic diversity of F oxysporum f. sp. radicis-lycopersici in Florida suggests that rapid changes in population structure may occur while facing selection pressure such as the resistane gene Frl deployed in the field. The evolution of a lo cal pathogen population may result from the selection for mutants, recombinants, or immigrants (107,108). The a ppearance of a new pat hogenic race may be a result of either introduction or local selection (55). Given th at the dominant resistance gene Frl has been deployed since 1983 (150), it is relevant to evaluate whether a physiological race has appeared in the field. VCGs 0094, 0098, and 0099 were reported in Florida in the 1990s (139) but the virulence of these three VCGs is not we ll known. Moreover, the frequency of these three VCGs may have changed in Florida due to se lection processes. A survey of the population structure of F oxysporum f. sp. radicis-lycopersici in Florida seems to be pertinent following a previous study (139). The genetic structure for a sp ecies is defined as the amount and distribution of genetic variation within and among populations of that species. Also, the genetic structure of a population is affected by the evolutionary hist ory of that population (107,108). Knowledge of genetic structure is useful to understand the evolutiona ry process of pathogen populations in the past and also provides insight fo r their future evolutionary cha nge. The genetic structure of a pathogen population is a consequence of the in teractions among five evolutionary forces:

PAGE 17

17 mutation, genetic drift, gene/genotype flow reproduction/mating system, and selection. Understanding the evolutionary potential may provide a suitable strategy to manage plant disease (107,108). Molecular markers such as RFLP (139) a nd random amplified polymorphic DNA (RAPD) (10), along with VCG, have been employe d for studying the population genetics of F oxysporum f. sp. radicis-lycopersici. However, highly polymorphic markers, such as microsatellites, have not been evaluated for their efficacy fo r understanding the population genetics of F oxysporum f. sp. radicis-lycopersici. Microsatellites or simple sequence repeats (SSRs) are defined as at least five runs of tandemly repeated motifs that range from 1-6 bases found in both coding and noncoding regions and are usually characterized by extensive levels of length polymorphism (6,182). Due to the high variability and codomin ance, microsatellites have been used in investigating population genetics of plant pathogens, such as Sclerotinia sclerotiorum (152), Phaeosphsaeria nodorum (160), and Phytophthora ramorum (127). Since the genomic sequence of race 2 of F oxysporum f. sp lycopersici was released in 2007 (available at http://www.broad.mit.edu), bioinformatics appr oaches may likely be useful to obtain microsatellite loci instead of de novo isolation of F oxysporum f. sp. radicis-lycopersici. Phylogenetics of F oxysporum f. sp. radicis lycopersici Multip le gene genealogies are suggested fo r recognizing phylogenetic species in fungi (169). The rDNA internal transcribed spacer (ITS) region has been extensively used in specieslevel phylogenetics due to its higher evolutionary rate. However, many Fusarium species including the F oxysporum species complex possess non-orthologous copies of the ITS2 due to interspecific hybridization or ge ne duplication prior to the ev olutionary radiation (118). Therefore, ITS may lead to inco rrect phylogenetic inferences for Fusarium In contrast, nonorthologus copies of the tr anslation elongation factor 1(EF-1 ) have never been detected.

PAGE 18

18 Also, EF-1 shows high polymorphism among closely rela ted species (119), suggesting that it may serve as an excellent gene for phylogenetics and even as a single-locus identification marker in Fusarium (52). The intergenic spacer (IGS) of the rDNA has also been proven to resolve phylogenetically intraspecifi c relationships within F oxysporum (4,19). Moreover, speciesspecific primers for Fusarium species have been developed ba sed on the alignment of the IGS region (147), suggesting that this is a suita ble candidate locus for phylogenetic analyses. Although mitochondrial small subunit (mtSSU) is not phylogenetically informative at the subspecies level, combined mtSSU and EF-1 may provide much better resolution among and within lineages (clades) (119). Recently, sequencing microsatellite loci ha s received a growing interest in phylogenetic species recognition of fungi due to thei r polymorphisms (29,30). However, this approach has not been examined for F oxysporum f. sp radicis-lycopersici. Phylogenetic analyses based on DNA sequences of mtSSU and EF-1 suggest that the F oxysporum species complex is monophyletic but many formae speciales in the complex are polyphyletic. Fusarium oxysporum f. sp. lycopersici and radicis-lycopersici are proposed to be paraphyletic (119), suggesting that they evolved from a common ancestor. Some VCGs between these two formae speciales may be more closely related than others within the same forma specialis For example, Cai et al. (19) showed that VCG 0035 of F oxysporum f. sp. lycopersici is more closely related to F oxysporum f. sp. radicis-lycopersici than to other VCGs of F oxysporum f. sp. lycopersici based on partial IGS sequences. However, Kawabe et al. (83) suggested that three isolates of F oxysporum f. sp. radicisi-lycopersici without VCG information were phylogenetically closer to VCGs 0030 and 0032 of F oxysporum f. sp. lycopersici than the others. Without including most known VCGs of F oxysporum f. sp. lycopersici and radicisilycopersici phylogenetic analyses may not uncover th e real relationship among VCGs of these

PAGE 19

19 two formae speciales. Moreover, multiple gene analyses have to be employed to construct genealogies and to recognize phylogenetic species (169). To initiate sexual recombination, two isolates must be in the same biological species and carry distinct mating-type (MAT) idiomorphs, MAT-1 or MAT-2 whereas vegetative incompatibility does not necessa rily prevent the sexual cycle (97,181). These two mating types were found in F oxysporum f. sp. lycopersici correlated with phylogene tic lineage and VCG but not with race (83). Although F oxysporum f. sp. lycopersici carries two mating type idiomorphs, successful crosses have never been observed (83) So far, little attention has been paid to determine the correlation between mating types and VCGs in F oxysporum f. sp. radicislycopersici Silicon and Plant Disease Silicon (Si) is the second most prevalent element in the earth s crust, comprising more than 25% of the earths crust ( 158). Monosilicic acid, Si(OH)4, is the main form of silicon in soils available to plants at a typical concentration of 0.1-0.6 mM (41), suggesting the low solubility of Si (60). Plants accumulating Si range from 1% to more than 10% of the dry biomass (41,42). Si accumulators are used to describe plants when a c oncentration of Si is grea ter than 1 % of the dry weight (43) and primarily accumulate Si in leaves as a result of the tran spiration stream (7). Recently, the Low silicon rice 1 ( Lsi1 ) gene encoding a Si uptake tran sporter has been isolated in rice, a typical Si-accumulating plant (101). In genera l, dicots such as tomato and soybean have a lower ability to accumulate Si than monocots (26) Si has not been considered as an essential element, although some plants absorb Si at levels equal to or greater th an essential elements (26,42,43). Beneficial effects of Si have been reported on the growth an d development of plants and enhancing plant resistance to various biotic and abioti c stresses (20,26,41,42,43,48,135). Effects

PAGE 20

20 of Si on plant disease are mainly observed in sh oot organs due to the movement of Si through transpiration (7,60). The correlation between the content of Si in r oots and soilborne diseases has not been well addressed. Interestingly, tomato, a Si excluder, accumulates more Si in roots than in shoots, and Si deposits mainly in the cell-wa ll-fraction (60). The reinforcement of cell walls has been suggested to reduce fungal penetration, whereas Si can also trigger defense responses (20,49). However, Diogo and Wydra (32) suggested that Si-mediated re sistance in tomato against Ralstonia solanacearum was not located in roots but in stems as a result of reinforcing the pectic polysaccharide structure of stem ce ll walls and impeding the bacterial movement to stems. Surprisingly, a resistant cultivar, Hawaii 7998, tr eated with Si signif icantly reduced the bacterial population in both root s and stems compared to non-treated plants suggesting that Simediated resistance might also loca te in roots. No effect of Si was suggested in tomato against Pythium aphanidermatum and the accumulation of Si in root s did not display a physical barrier for restricting the basipetal spread of the pathog en (61). It is not well known whether the efficacy of Si-mediated resistance in tomato depends on pathosystems investig ated. Although Menzies et al. (112) suggested no effect of Si on Fusarium crown and root rot of tomato, the relationship between the Si effect and inoculum concentratio n and the effect of Si on disease progress have not been studied. It may be pertinent to further re search whether Si can re duce disease severity of Fusarium crown and root rot of tomato as a result of either induced defe nse resistance and/or decrease in the root penetration by F oxysporum f. sp. radicis lycopersici that does not form appressoria prior to penetration but directly penetrate the juncti on of epidermal cells along roots or at the crown (178). Hypotheses This dissertation addressed evolutionary questions in regard to the population structure and phylogenetics of F oxysporum f. sp. radicis lyco persici and how this relates to disease

PAGE 21

21 management. Moreover, incongruencies concerning the effect of silicon on Fusarium crown and root rot of tomato were evalua ted. Chapters two, three, and four present research results that address these questions. Chapter 2 determined the population structure of F oxysporum f. sp. radicis lycopersici using VCGs and microsatellites. The first hypothesi s was that more than one VCG coexist in the three main tomato-growing counties of Florida: Manatee, Hendry, and Collier Counties, but only one VCG predominated due to selective fit. Although no physiological race has been reported, the frequency of VCGs may help breeders to choose representative isolates for screening resistant tomato lines (125). The second hypothesis was that microsatellite primers acquired from the genome sequence of F oxysporum f. sp. lycopersici can cross-amplify F oxysporum f. sp. radicis lycopersici since these two formae speciales are phylogenetically related (10,19). Bioinformatics approaches for developing micr osatellite primers from the published genome sequence of closely related species may reduce the time needed to develop these primers for studying the population genetics of a given species. The final hypothesis of chapter 2 was that genotype flow shaped the population structure of F oxysporum f. sp. radicis lycopersici in Florida. A pathogen with a high degree of genotype flow may have greater genetic diversity and pose a higher potential in break ing down resistance (107,108). If gene/genotype flow in F oxysporum f. sp. radicis lycopersici is high, disease management strategies should involve restricting genotype flow of the pathogen. Chapter 3 investigated phylogenetics and mating types of F oxysporum f. sp. radicis lycopersici and their association w ith virulence. The first hypothesis was that the forma specialis is polyphyletic or of multiple evolutionary orig ins. Identifying the predominantly phylogenetic lineage may be helpful for breeding and for unde rstanding the evolutionary history of the

PAGE 22

22 pathogen. The second hypothesis was that each VCG carried only one mating-type idiomorph but the pathogen possesses two distinct idiomo rphs. If two distinct idiomorphs are found, crossing isolates carrying opposite idiomorphs would be a direct approach to evaluate the possibility of sexual recomb ination in this pathogen. Th e third hypothesis was that F oxysporum f. sp. radicis lycopersici may have co-evolved with the domesti cation of its host due to selection by environmental and genetic un iformity of the agricultural ecosystem (161). The final hypothesis in chapter 3 was that virulence in various VCGs of F oxysporum f. sp. radicis lycopersici is not identical even within the same phylogenetic lineage. Highly virulent isolates may be used for acquiring more consistent pl ant infection since the uncertainty of seedling infection by the pathogen has been reported (80). Data presented in chapter 4 evaluated the e ffect of Si on the development of Fusarium crown and root rot of tomato. The first hypothesis was that Si can reduce disease severity due to a delay in initial infection and the basipetal movement of the pathogen. The effect of Si on disease severity can be specifically investig ated through observing di sease progress among Si treatments over time, resulting in an objective evalua tion of the potential app lication of Si in the integrated pest management (IPM) of Fusarium crown and root rot of tomato. The second hypothesis was that the decrease in disease seve rity may be associated with increasing Si contents in roots. If the Si content of the r oots influences disease development in tomato, an appropriate application of Si fertilizers may be recommended.

PAGE 23

23 CHAPTER 2 POPULATION STRUCTURE OF Fusarium oxysporum f. sp radicis-lycopersici IN FLORIDA INFERRED FROM VEGETATIVE COMPAT I BILITY GROUPS AND MICROSATELLITES Introduction Fusarium oxysporum Schlechtend:Fr. f. sp. radicis-lycop ersici W. R. Jarvis & R. A. Shoemaker, the causal agent of Fusa rium crown and root rot of tomato is an important disease on tomato worldwide (74) and limits tomato production in southern Florid a (27). The pathogen is distinct from F oxysporum f. sp. lycopersici the causal agent of Fusarium wilt on tomato (92,146), based on symptoms, disease developmen t, and physiological ch aracteristics (72). Fusarium oxysporum f. sp. radicis-lycopersici is favored by cool weather (below 20C) but Fusarium wilt of tomato is favored by temperatures around 27 C (110,141). Moreover, F oxysporum f. sp. radicis-lycopersici has a wide range of hosts in cluding species in Solanaceae, Leguminosae, Cucurbitaceae, and Chenopodiaceae, whereas F. oxysporum f. sp. lycopersici is host-specific to members of the genus Lycopersicon (110,141). Also, F oxysporum f. sp. radicislycopersici causes cortical discoloration that usua lly extends no more than 25 cm above the crown, whereas vascular discoloration caused by F. oxysporum f. sp. lycopersici extends further. The pathogen can attack young tomato seedlings but infected plants usually do not wilt until the first full fruiting stage (72). Genetic differen ces have been revealed between these two formae speciales since F oxysporum f. sp. lycopersici and radicis-lycopersici are vegetatively incompatible (128). Vegetative compatibility, also known as heterokaryon compatibility or heterokaryon incompatibility (97), has been useful in characterizing genotypic diversity in F oxysporum f. sp. radicis lycopersici (80,82,125,139). Strains that can form stab le heterokaryons are said to be vegetatively compatible and grouped into the same vegetative compatibility group (VCG)

PAGE 24

24 (95,128). The VCG assay, although laborious, has helped to elucidate populat ion structure and to differentiate F oxysporum f. sp. lycopersici and radicis lycopersici (128,139). Due to the absence of known physiological races for F oxysporum f. sp. radicis lycopersici the frequency of VCG may help breeders to choo se representative isolates of the pathogen for screening tomato lines for disease resistance (125). It has been suggested that Florida is the center of origin for the cosmopolitan VCG 0094 of F oxysporum f. sp. radicis lycopersici along with two recently found VCGs 0098 and 0099 (139). However, the distri bution of these three VCGs in Florida has not been well evaluated since they were first reported in the 1990s. Because resistant cultivars carrying the Frl resistance gene have been releas ed since the 1980s (50,148,150,151), it is pertinent to investig ate the population structure of the pathogen for tomato breeding. In addition to VCG, molecular markers such as restriction fragme nt length polymorphism (RFLP) (139) and random amplified polymorphi c DNA (RAPD) (10) have been used for studying the population genetics of F oxysporum f. sp. radicis lycopersici. Microsatellites, also known as simple sequence repeats (SSR), consist of tandemly genetic loci of one to six base pairs and are found more abundan tly in noncoding regions than in exons (58,167). They have been used to study population genetics of impor tant plant pathogens and contribute to the knowledge of disease management due to thei r high polymorphism and rapid mutation rates (23,127,183). However, microsatellite primers for F oxysporum f. sp. radicis lycopersici have not been developed and used to study evolutiona ry forces. Such studies can help model the breakdown process of introdu ced resistance genes (107,108). The main drawback of microsatellites is that they have to be isolated de novo from the species being examined if no genome sequen ce is available (182). Directly isolating microsatellites from F oxysporum f. sp. radicis lycopersici is likely laborious due to the

PAGE 25

25 procedure of screening genomic libraries with appropriate probes (182). However, a recently released genome sequence of race 2 of F oxysporum f. sp. lycopersici may help in the development of microsatellite primers by means of bioinformatics. The microsatellite primers designed from the genome sequence of F oxysporum f. sp. lycopersici may cross-amplify F oxysporum f. sp. radicis lycopersici since these two formae speciales are phylogenetically related (19,119). The genetic structure of a pat hogen population is a result of the interactions among five evolutionary forces: mutation, genetic drift, gene/genotype flow, reproduction/mating system, and selection (107,108). Of these five fo rces, determining genotype flow of F oxysporum f. sp. radicis lycopersici may be most appropriate for inves tigating the genetic structure of the pathogen due to its extremely mitosporic reprodu ction (97). Pathogens ex hibiting a high degree of gene/genotype flow have gr eater genetic diversity due to an increase in the effective population size, resulting in a greater risk in the breakdown of re sistance (107,108). To effectively limit genotype flow of F oxysporum f. sp. radicis lycopersici in Florida, commercial tomato-growing sites which donate and receive more migrants have to be identified. The objectives of this study we re (i) to investigate the frequency of VCGs of three main tomato-growing counties in Florida, (ii) to develop microsatellite primers for F oxysporum f. sp. radicis lycopersici (iii) to evaluate genetic structure w ithin and among populat ions, and (iv) to determine levels of migration between populat ions. Portions of this study were published previously as reports (66,67,69). Materials and Methods Fungal Collections Diseased to mato plants with symptoms of Fu sarium crown and root rot were sampled from three main Florida tomato-growing counties, Ma natee, Hendry, and Collier Counties (Fig. 2-1).

PAGE 26

26 Small pieces of stem and root tissues were st erilized in 5% sodium hypochlorite for 1 min, followed by rinsing in sterile deionized water. Th e pieces of the tissues were plated onto potato dextrose agar (PDA), water agar (WA), and Koma das agar (89). Morphological characteristics of F oxysporum were examined based on the methodology of Nelson et al. (116). A monosporic isolate from each diseased plant was stored on st erile paper and in 15% glycerol for long-term storage. Pathogenicity Tests Pathogenicity tests were carri ed out as previously descri bed (125,145). However, culturing F oxysporum f. sp. radicis lycopersici on rich media such as PDA m ay cause loss in its pathogenicity (94). Therefore, mono sporic isolates were cultured on carnation leaf agar (CLA) at 25C (12 h light photoperiod with photonflux of 40.8 mol/m2s) for 2 wk to obtain a massive production of conidia. Twenty tomato seeds (cv. Bonny Best) were dipped in a conidial suspension (106 mL-1) for 3 min, placed on WA plates, incubated in the dark at 25 C for three days, and then moved to 25 C for additional four days. Isolates causing both dark-brown lesions on the root-stem transition region and more than 50% of the seedlings after inoculation were identified as F oxysporum f. sp. radicis lycopersici (125). Molecular Differentiation of F oxysporum f. sp. lycopersici and radicis lycopersici Since F oxysporum f. sp. lycopersici and radicis lycopersici are m orphologically indistinguishable, molecular discriminations were conducted to further confirm that all isolates used in this study belonged to F oxysporum f. sp. radicis lycopersici For DNA extraction, single-spored isolates were grown on CLA before transferring mycelia plugs to PDA. Mycelia on PDA were harvested to extract DNA using DNeasy Plant Minikits (Qiagen, Inc., Valencia, CA). A 947 bp fragment specific for F oxysporum f. sp.

PAGE 27

27 radicis lycopersici was amplified using primers sprlf (GATGGTGGAACGGTATGACC) and sprlr (CCATCACACAAGAACACAGGA), which targeted the exo polygalacturonase gene ( pgx4 ) (63). Polymerase chain reaction (PCR) cond itions were performed with 50 cycles of denaturation at 94 C (1min), annealing at 62 C (1 min), and elongation at 72 C (2 min), and finally 72 C for 10 min. F oxysporum f. sp. lycopersici was identified using primer pairs P12F2B (TATCCCTCCGGATTTGAGC) and P12-R1 (AATAGAGCCTGCAAAGCATG) to amply an 1 kb fragment of SIX1 a virulence locus (172). Amplifica tions were carried out using the following PCR conditions: 94 C for 2 min, 30 cycles at 94 C for 45 s, 64C for 45 s, and 72C for 45 s and a final elongation step at 72 C for 10 min (172). Vegetative Compatibility Tests Methods for m edia preparation, procedures for generating nit mutants and complementation tests were performed as previo usly described (24). Briefly, each isolate was cultured on minimal medium (MM) for 5-7 days, and then forty plugs were transferred to plates filled with either PDA or MM supplemented with 25 g of KClO4 per liter (PDC and MMC media). After 7-14 days, rapid growing sectors we re transferred to MM, nitrite agar with 0.5 g NaNO2 L-1, and hypoxanthine agar with 0.2 g NaNO2 L-1. Based on growth characteristics, three phenotypic classes of n itratenon-utilizing ( nit ) mutants, nit1 nit3 and NitM, were assigned. Two mutants each of nit1 and nit3 were selected in complementation tests on MM with NitM testers, representing prev iously reported VCGs of F oxysporum f. sp. radicis lycopersici (Table 2-1). Plates were incubated at 25 C and evaluated daily for 2 wk. The presence of aerial mycelia at the point of hyphal contact was evidence of vegetative compatib ility, and these isolates were assigned to the same VCG (128).

PAGE 28

28 Development of Microsatellite Primers Micros atellite loci were derived using a bioinformatics a pproach and a microsatellite enrichment procedure. For the bioinfor matics approach, the genome sequence of F oxysporum f. sp. lycopersici (strain 4287, race 2, VCG 0030) was downloaded from http://www.broad.mit.edu. Sequences with a microsatellite motif were identified using TANDEM REPEAT FINDER (13). Three alignment parameters, match, mismatch, and indel were set at 2, 7, and 7, respectively. The microsatellite enrichment protocol wa s based on the method of Edwards et al. (35) except a biotinylated (GA)17 probe was used in this study. Briefly, genomic DNA of F oxysporum f. sp. radicis lycopersici (isolate HE-0631, VCG 0094) was ex tracted from mycelia using DNeasy Mini Kit (Qiagen, Valencia, CA), digested with Sau3AI, and ligated with Sau3AI linkers. Ligation was confirmed by PCR. The fragment lib rary was enriched by hybridization with a biotinylated (GA)17 probe and then captured by VECTRE X Avidin D (Vector Laboratories, Burlingame, CA). The selected fragment was enriched by PCR using primers designed to the linker sequence. The enriched library was ligated into a TOPO TA pCR 4.0 vector, transformed into One Shot Escherichia coli competent cells (Invitrogen, Carlsbad, CA), and grown overnight on Luria broth (LB) agar plates containing 50 g mL-1 kanamycin. Transformed colonies were cultured in LB overnight and screen ed for microsatellite re peats using PCR with a (GA)10 repeat primer and M13F/M13R. The appropriate sized amplicons (>200 bp) were screened on 2% agarose gels, am plified by rolling circle amplif ication (RCA), and sequenced with the T7 primer and BigDye Terminator Cycle Sequencing Chemistry (Applied Biosystems, Foster City, CA) at the Interd isciplinary Center for Biotechnology Research (ICBR) facility, University of Florida. Sequences were edit ed using SEQUENCHER 4.2 (Gene Codes Corp., Ann Arbor, MI). Primers were designed using PRIMER3 (142). An M13 tail

PAGE 29

29 (CACGACGTTGTAAAACGAC) was added to the 5 end of the each forward primer for amplification with 6-FAM/VIC/NED/PET-labeled M13 primers (15). Fourteen isolates of F oxysporum f. sp. lycopersici and 33 isolates of F oxysporum f. sp. radicis lycopersici representing most known VCGs (Table 2-2) were used to verify designed microsatellite primers. PCR amp lification was carried out in a 20 L reaction mixture containing 0.5 U of GoTaq Flexi DNA polymerase (Promega Corp., Madison, WI) or Taq DNA polymerase (Bioline USA Inc. Taunton, MA), 1 Promega Colorless GoTaq Flexi Buffer or 1 Bioline KCl buffer, 2.5 mM MgCl2, 0.01 M of each forward primer labe led with the M13 tail, 0.15 M of each reverse primer and 6-FAM/VI C/NED/PET-labeled M13, and 50 M dNTP. Cycling conditions were: 94 C for 3 min, 35 cycles at 94 C for 30 s, 52C for 30 s, and 72 C for 45 s and a final elongation step at 72C for 20 min (35). 6-FAM/VIC/ NED/PET-labeled PCR products were mixed, diluted, and run on an ABI 3730x1DNA Analyser (Applied Biosystems) at ICBR by loading 1 L of the diluted PCRs, 9.9 L formamide, and 0.1 L LIZ 600 size standard (Applied Biosystems). Fragment sizes we re analyzed using PEAK SCANNERTM SOFTWARE version 1.0 (Applied Biosystems). Allele numbers and gene diversity were analyzed using ARLEQUIN version 3.1 (45). A neighbor-joining (NJ) tree base d on Neis minimum genetic dist ance (115) was generated using POPULATIONS (Centre National de la Recherche Sc ientifique, Paris, Fran ce) to infer genetic distance among isolates. Two isolates of F commune (Table 2-2) were also genotyped and utilized as outgroups. Microsatellite Data Collection Ten m icrosatellite loci (CH2-9, CH215, CH2-66, FOL20, FOL35, FOL99, FOL175, FOL185, FOL245, and FOL680) (Table 2-5) with hi gh gene diversity were chosen to genotype

PAGE 30

30 125 isolates (Table 2-3) of F oxysporum f. sp. radicis lycopersici identified using pathogenicity tests and the PCR-based discrimination. DNA extraction and PCR temperature cycling conditions were the same as above except that annealing temperature for FOL99 was increased to 58C to avoid nonspecific amplification. A control isolate (HE-0601) w ith a reproducible fragment analysis with the ten loci was included in every run of 95 samples. Since DNA degradation, low DNA concentration, and primer-site mutation may resulted in genotyping error, MICRO-CHECKER was used to identify a nd correct microsatellite data (173). Microsatellite Data Analysis Isolates with the sam e micr osatellite genotyping pa ttern within populations were removed and counted only once to generate clone-corr ected microsatellite da ta (139). Depending on genetic analyses, VCG affiliation alone or VCG affiliation and geographic source was predefined as populations (139). Microsat ellite haplotypes were identif ied using ARLEQUIN version 3.1 (45) and THE EXCELL MICRO SATELLITE TOOLKIT (University College Dublin, Belfield, Ireland). The existence of identic al microsatellite ha plotypes shared be tween two populations was considered to be direct evidence of genotype flow (56). Formatted files were generated for the data analyses using THE EXCELL MICROSATELLITE TOOLKIT a nd CONVERT (54). The mean number of pairwise differences and it s standard deviation (S D) were determined according to Tajima (165). The average nucleotid e diversity was calculated based on Nei (115) and Tajima (163). Fixation indices ( F statistics) were used to examine qualifying differentiation between pairs of populations and the degree of population subdiv ision according to the sum of squared size differences for mi crosatellite da ta (156). If P was 0.05, genetic differentiation between two populations was considered significan t. The distribution of variance within and among populations was based on using the analysis of molecular variance framework (AMOVA)

PAGE 31

31 implemented in ARLEQUIN. A nonparametric a pproach with 1,023 permutations was performed (44). Linkage equilibrium tests were used to dete rmine whether recombination was probable for F oxysporum f. sp. radicis lycopersici in the field. Fishers exact test implemented in GENEPOP version 3.4 (Insititut des Sciences de lEvoluti on, Universit Montpellier 2, Montpellier, France) was analyzed based on an Markov Chain Monte Carlo (MCMC) algorithm with 1,000 batches and 1,000 iteration s/batch (17,131). A pair of loci were considered at linkage disequilibrium if its P value was 0.05. Two indices of multilous linkage disequilibrium, IA and d, were calculated using MULTILOCUS version 1.3 (Department of Biology, Imperial College, Silwood Park, UK) with 1000 randomizations fo r testing statistical significance (1). d is a modification of IA and it is less sensitive to th e number of loci used (1). The population genetic structure of the microsatellite data was analyzed using the Bayesian model-based clustering program STRUCTUR E version 2.2 (46,126). STRUCTURE infers population structure based on multilocus genotyp e data and calculates the membership coefficients to determine possible admixed/hybr id individuals. The number of populations (K) was computed from 1 to 3. Ten runs with th e admixture model and a burn-in period of 10,000 generation and MCMC simulations of 100,000 iterations for each run. K for the best fit of the data was determined by estimates for the natural logarithm of the probability of the data (46,126). Graphic displays of STRU CTURE results were genera ted using DISTRUCT (138). Historical migration between three Florida tomato-growing counties was evaluated using MIGRATE-N version 3.0.3 (11,12). Briefly, MIGRATE-N jointly cal culated the mutation scaled population size ; that is, the effective population size Ne times the mutation rate per site and generation (2 Ne for haploid F oxysporum f. sp. radicis lycopersici ) and the mutation scaled

PAGE 32

32 immigration rate M ( m / ) between pairs of populations base d on the coalescence theory (85). Asymmetrical migration between population pair s was analyzed using the Bayesian inference module in MIGRATE-N. Ten repli cates of one long chain and four chains of static heating scheme with four temperatures (1.0, 1.5, 3.0, a nd 10000) were used for running microsatellite data with Brownian motion, which assumed a st epwise mutation model. 50000 recorded steps in interval of 200 with 10 c oncurrent chains (100,000,000 visits) were performed, and 1,000,000 trees per chain were discarded as burn-in. Migrants per generation between two populations were derived from multiplied by M The exponential growth rate ( g) for evaluating the population size fluctuation of each population was calculated using Bayesian MCMC implemented in LAMARC version 2.1.3 (90). One long chain and four chains of static hea ting scheme with four temperatures (1.0, 1.5, 3.0, and 10,000) were used for running microsatellite da ta using the stepwise model. Twenty-five thousand recorded steps in intervals of 40 we re performed and 100,000 samples were discarded as burn-in. Positive values of g suggested that the population size has been growing, whereas negative values indicate that the population size has been shrinking. Nevertheless, if the confidence intervals of g includes zero, the popula tion size likely has little or no growth (90,91). Results Pathogenicity Tests Of 148 isolates from Manatee, Hendry, and Co llier Counties, 125 isolates caused disease symptoms on 50-100% of the tomato seeds 7 days after inoculation. A dark brown lesion on the root-stem transition region (crown) was observed, suggesting these isolates were the pathogenic F oxysporum f. sp. radicis lycopersici (145). These 125 isolates were further characterized using primers P12-F2B and P12-R1 to exclude F oxysporum f. sp. lycopersici. The primers consistently amplified two ioaltes of F oxysporum f. sp. lycopersici (MN24 and DA1). However,

PAGE 33

33 no amplification was observed for these 125 isolat es, and thus they were used for studying the population genetics of F oxysporum f. sp. radicis lycopersici in Florida. Molecular Identification of F oxysporum f. sp. radicis lycopersici Primers sprlf and sprlr specific for F oxysporum f. sp. radicis lycopersici (63) were first tested with 24 isolates. Of these isolat es, 15 isolates represent VCGs 0090, 0091, 0092, 0093, 0094, 0096, 0098, and 0099 of F oxysporum f. sp. radicis lycopersici and nine isolates belong to VCGs 0030, 0031, 0032, 0033, and 0035 of F oxysporum f. sp. lycopersici These two primers amplified the expected 947 bp fragment for isolates of F oxysporum f. sp. radicis lycopersici in VCGs 0090, 0091, 0092, and 0096, whereas no amplifica tion was observed in the other VCGs of F oxysporum f. sp. radicis lycopersici or in the isolates of F oxysporum f. sp. lycopersici (Fig. 2-2). In contrast, F oxysporum f. sp. lycopersici specific primers P12-F2B and P12-R1 correctly confirmed that the nine isolates belonged to F oxysporum f. sp. lycopersici without further amplifying any isolates of F oxysporum f. sp. radicis lycopersici Therefore, to identify F oxysporum f. sp. radicis lycopersici pathogenicity tests along with primers P12-F2B and P12-R1 were used to exclude isolates of F oxysporum f. sp. lycopersici Vegetative Compatibility Tests One hundred and twenty five isolates collected from 2006 to 2008 were used for vegetative compatibility tests. Sixty-nine percent of these isolates could be assigned to one of three VCGs: 0094, 0098 and 0099. Thirty eight isolat es could not be assigned to a known VCG. Frequencies of VCGs 0094, 0098 and 0099 were 38.6%, 24.4%, and 6.8% respectively, indicating that VCG 0094 was predominant among these isolates asse ssed (Table 2-4). Although 77% of VCG 0094 was mainly confirmed as subgroup I, some of them also formed heterokaryons with testers of subgroup II, III, or IV. Interestingly, VCG 0098 was previously reported in Collier County, but it has now been found in Manatee and Hendry Coun ties, whereas VCG 0099 seems restricted to

PAGE 34

34 Collier County. Two isolates MN-0713 and MN-0724 from Manatee County were assigned to both 0094 and 0098 after repeating two single-spored isolations and confirming vegetative compatibility with testers of 0094I (01150-6 and 01152-31) and 0098 (CL-7/6, CL-75/4) (Table 2-1). Occasionally, all three VCGs 0094, 0098, and 0099 were recovered from a single sampling site in Collier County. On the other ha nd, VCGs 0094 and 0098 were found occasionally on single farms in both Manatee and Hendry Counties. Development of Microsatellite Primers A m icrosatellite enrichment pr ocedure and a bioinformatics appr oach were used to develop microsatellite primers for st udying the population genetics of F oxysporum f. sp. radicis lycopersici Of 48 clones sequenced from the enrichment approach, 14 included microsatellites. For the bioinformatics approach, 38 microsatelli te loci were acquired using TANDEM REPEAT FINDER. These 52 loci were used to design prim ers to initially amplify four representative isolates each of F oxysporum f. sp. lycopersici and radicis lycopersici Of these 52 original loci, 27 were amplified consistently and exhibited va riation among the eight isolates. These 27 pairs of primers (Table 2-5) were further tested on another nine isolates of F oxysporum f. sp. lycopersici and 27 isolates of F oxysporum f. sp. radicis lycopersici that included the most reported VCGs (Table 2-2). These 27 loci were successfully genotyped in both F oxysporum f. sp. lycopersici and radicis lycopersici Of these 27 loci, 13 did not encode any protein when queried by BLAST to GenBank. Allele numbers and gene diversity scor es are shown on Table 2-5. The 27 loci had 1 to 14 alleles per locus (ave rage of 6.7 alleles for F oxysporum f. sp. radicis lycopersici and 5 alleles for F oxysporum f. sp. lycopersici ). Loci FOL680 and FOL35 had the highest gene diversity for F oxysporum f. sp. lycopersici and F oxysporum f. sp. radicis lycopersici respectively, whereas gene dive rsity of the 27 loci ranged from 0 to 0.95. A neighbor-joining tree

PAGE 35

35 generated from the fragment analysis of the 27 microsatellite loci fo r these 47 isolates (Table 2-2) revealed four clusters: FORL clusters I and II, and FOL clusters I and II (Fig. 2-3). FORL cluster II and FOL cluster II represent predominant VCGs of F oxysporum f. sp. radicis lycopersici and F oxysporum f. sp. lycopersici respectively. Molecular Diversity Indices Ten of the 27 loci (CH2-9, CH2-15, CH 2-66, FOL20, FOL35, FOL99, FOL175, FOL185, FOL245, and FOL680) were selected to genotype the 125 isolates of F oxysporum f. sp. radicis lycopersici collected fro m Manatee, Hendry, and Collier Counties for population genetics analyses. Pairwise Fst values between isolates from these three counties suggested no genetic differentiation in either VCG 0094 or 0098 (Tab le 2-6 and Table 2-7). VCG 0098 was found in Collier County but not in the other two countie s in the 1990s (139) and two isolates (MN-0713 and MN-0724) in this study were vegetatively compatible with testers of VCGs 0094 and 0098. Therefore, this study tested a hypothesis that th ese two VCGs originally migrated from Collier County to Manatee and Hendry Counties according to measures of intrapopulation diversity (mean number of pairwise differences and aver age gene diversity). Th e basic assumption was that a new population exhibits a lower level of molecular divers ity as a result of the founder effect and less accumulation of mutations (107,108,139). Mean numbers of pairwise differences and aver age gene diversity were compared for three Floridia populations (Table 2-8). Hendry C ounty had a higher mean number of pairwise differences and average gene dive rsity when either whole microsat ellite or clone-c orrected data were used for analysis. However, to evaluate these two indices for each VCG, the clonecorrected data set was used to further classify each VCG and its affiliated geographical region (139). The two indices were the highest in Collier County for VCGs 0094 and 0098, suggesting

PAGE 36

36 that these two VCGs might have migrated from Collier County to the other two counties. For isolates not assigned to a known VCG, the two indices were highe st in Hendry County. Regarding individual VCGs, results of clone-cor rected data suggested that VCG 0099 had a higher mean number of pairwise differences and average gene diversity than VCGs 0094 and 0098, whereas VCG 0094 was predominant in Florida. Hierarchical Distribution of Total Gene Diversity The AMOVA for F oxysporum f. sp. radicis lycopersici was used to determine the variance within and among populations. The results demonstrated that 84% of the variance occurred within populations and that a significant partition of variance, 16% ( P <0.0001), was detected among populations. st was 0.159, suggesting a medium level of differentiation among populations (Table 2-9). Linkage Disequilibrium Fishers exact test revealed that16.7% of pair wise loci in isolates from Manatee County and 62.3% in Collier County were at linkage equ ilibrium. However, two indices of multilocus linkage disequilibrium, IA and d, showed significant linkage disequilibrium, suggesting clonal reproduction for F oxysporum f. sp. radicis lycopersici Despite the finding of up to 62.3% of pairwise loci of the Collier popul ation at linkage equilibrium, d, which is less sensitive to the number of loci sampled, for the population re vealed significant deviation from linkage equilibrium (Table 2-10). Population Admixture In order to infer the best value of K clusters (real population groups), different runs of K=1 to 3 were perform ed using STUCTURE. Mean va lues of likelihood probability of data given K=1 to 3 were -665.8, -562.9, and -533.1, respectively. Therefore, K=3 was used to determine possible admixture in the three populations, suggesting that isol ates sampled from the same

PAGE 37

37 county could be defined as a population. Eighteen likely admixed individuals were identified: three in Manatee County (12% of the total clone-corrected in dividuals), 13 in Hendry County (50% of the total), and two in Collier County (5.89% of the total) (Fig. 2-4). Highly significant admixture found in Hendry County suggests that some individuals could be immigrants from the other two counties. Historical Migration and Population Growth Rate The population param eter theta ( ) of the three countie s was very close ( = 0.09738 to 0.09755). Asymmetrically historical migration was shown according to estimates of directional genotype flow (Table 2-11 and Fig. 2-1). Most migrants migrated from Manatee and Collier Counties to Hendry County (2 NmManatee Hendry =2.75, 2NmCollier Hendry =2.15), whereas fewer migrants were found between Manatee and Col lier Counties. The historical movement of migrants was much higher between Manatee an d Hendry Counties than that between Collier and Hendry Counties no matter which direction of migration was compared. Population growth rates, g, with 95% confidence intervals were 0.10+0.98/ 22.5, 0.11+0.13/-0.14, and 0.10+1.27/ 0.03 for Manatee, Hendry, and Collier Counties, respectively. This suggests that the population size of F oxysporum f. sp. radicis lycopersici in Manatee and Hendry Counties have shown little or no growth because the confidence levels include zero (90,91). In contrast, the population size in Collier County has been increasing. Discussion This study tested the hypothesis that migrati on played an important role in shaping the population structure of F oxysporum f. sp. radicis lycopersici in the three currently major tomato-growing counties of Florida. This st udy reveals the current population structure of F

PAGE 38

38 oxysporum f. sp. radicis lycopersici inferred from VCGs and micr osatellites and supports the hypothesis. Pathogenicity Tests A laboratory pathogenicity bioass ay has been used to identify F oxysporum f. sp. radicis lycopersici according to sym pto mology (71,145). Since the F oxysporum f. sp. radicis lycopersici specific primers were unable to identi fy predominant VCGs in Florida, the pathogenicity bioassay along with the F oxysporum f. sp. lycopersici specific primers were used to confirm isolates of F oxysporum f. sp. radicis lycopersici The isolates causing disease symptoms in greater than 50% of the assay and not amplified by F oxysporum f. sp. lycopersici specific primers were defined as F oxysporum f. sp. radicis lycopersici A subset of 12 isolates were also used for the standard root-dip inocul ation method (80) and caused either typical darkbrown lesions on the root-stem transition region a nd/or cortical discolora tion, indicating that the pathogenicity bioassay was reliable (125). Molecular Identification Molecu lar distinction of F oxysporum f. sp. radicis lycopersici and F oxysporum f. sp. lycopersici was originally evaluated to select isolates of F oxysporum f. sp. radicis lycopersici for studying population genetics. Sp ecies-specific primers for F oxysporum f. sp. radicis lycopersici targeting exo polygalacturonase (pgx4 ) (63) were first tested using most of the reported VCGs of F oxysporum f. sp. radicis lycopersici However, these primers did not amplify VCGs 0093, 0094, 0098, and 0099, i ndicating high variation of pgx4 in this pathogen. Moreover, the isolates originally used for deve loping the specific primers did not cover most known VCGs but only focused on Japanese isolates without VCG information (63). This is the probable reason for the lack of amplification in the majority of isolates from this study. In contrast, another set of primers, P 12-F2B and P12-R1 correctly identified F oxysporum f. sp.

PAGE 39

39 lycopersici but did not amplify isolates of F oxysporum f. sp. radicis lycopersici in this study. P12-F2B and P12-R1 targeted a virulence ge ne that is expresse d in the xylem during colonization of F oxysporum f. sp. lycopersici in tomato plants (172). Cell-wall-degrading enzymes (CWDEs) are secreted by F oxysporum (31,77) and a comparison of their nucleotide diversity may be used to discriminate formae speciales of F oxysporum (63). Regarding specificity of primers, however, virulence gene s conferring a specific tr ait to a pathogen are likely ideal targets since th ey have subtle nucleotide difference within the same formae speciales (98). Based on the findings in the current study, there is a need to develop more specific primers for rapid diagnosis of F oxysporum f. sp. radicis lycopersici VCG Frequency A VCG predom inant in the population may be sel ectively more fit than others (95). Sixtynine percent of the 125 isolates used in this study were as signed to three VCGs 0094, 0098, 0099, which were previously reported in the 1990s (139). While VCG 0094 is still predominant in Florida, its frequency has been reduced from 70.3 to 38.6% since the 199 0s (139). On the other hand, VCG 0098, first reported in Florida, has in creased in its distribution compared to a previous study (139). For example, VCG 0098 wa s not found in Manatee and Hendry Counties previously (139) but now is present in these tw o counties, probably due to migration. Interestingly, two isolates (MN-0713 and MN-0724) found in Manatee County were compatible with testers of both VCGs 0098 and 0094 I, su ggesting some degree of genetic relatedness between these two VCGs (Fig. 2-3). Therefore, sympatric speciation may be another possibility to explain the appearance of VCG 0098 in Manatee County, which suggests VCG 0098 might have evolved from the local population of VCG 0094. It has been suggested that bridge isolates may be interpreted as a stage in a process of e ither convergence or diverg ence of VCGs (80). If a VCG is newly formed, it may exhibit low levels of molecular diversity due to the founder effect

PAGE 40

40 and few accumulated mutations (139). Based on this assumption, VCG 0098 probably diverged from VCG 0094 as evidenced by low nucleotide diversity relative to VCG 0094. Further studies in molecular evolution of these two VCGs is n ecessary to clarify whethe r the ancestor of these two isolates originated from a native VCG 0094 of Manatee County or migrated from other regions. Florida has been suggested as the probable center of origin of VCG 0094 and this VCG might have migrated from Florida to Eur ope according to restri ction fragment length polymorphism (RFLP) analyses (139). The globa l distribution of this VCG is not well known (82). This study showed that the main VCG 0094 in Florida belonged to subgroup I (77%). Of these VCG 0094 I isolates, some could form heter okaryons with testers of subgroup II, III, and IV, suggesting variation in inte risolate compa tibility (139). Although 31% of the isolates used could not be assigned to a known VCG, most of them had the same microsatellite haplotype as ei ther VCG 0094 or 0098 (59%). Moreover, some interisolate pairings re vealed several minor groups that co mpromised two to five vegetative compatible isolates. These finding suggested th at these nonassigned isolates may not comprise new VCGs. More isolates of these minor groups need to be recovered from the tomato-growing region before VCG code numbers can be assigne d to them (81,86). Mutations may occur in any of a number of genes controlli ng incompatibility, altering the VCG of an isolate (95). In addition, the selection of spontaneous mutants may also cause mutations that interfere with the heterokaryosis (97). These results may expl ain why some nonassigne d isolates were not compatible with testers of VCGs 0094 and 0098 but had the same microsatellite haplotype as these two VCGs. On the other hand, increasing the frequency of unclassified isolates may also

PAGE 41

41 suggest the emergence of another VCG within F oxysporum f. sp. radicis lycopersici Regularly screening VCGs of the pathogen appears necessary to monitor its evolving genetic variation. Microsatellites Microsatellites are valuable for both taxonom ic and population genetic studies (14). The microsatellite loci isolated from other Fusarium species may not be useful for F oxysporum f. sp. radicis lycopersici due to their high mutation rates (78). Since the genome sequence of F oxysporum f. sp. lycopersici became available in 2007, a bioinf ormatics approach was used in this study to search microsatellite loci in addition to direct isolation from F oxysporum f. sp. radicis lycopersici Surprisingly, 20 of 38 microsatellite loci selected from the genome sequence could consistently amp lify known VCGs of both F oxysporum f. sp. radicis lycopersici and F oxysporum f. sp. lycopersici. Moreover, 7 of 14 microsatellite loci isolated de novo from F oxysporum f. sp. radicis lycopersici were cross-amplified between these two formae speciales. The cross amplification suggested that the two formae speciales are closely related. Genetic analyses of 27 microsatel lite data for these two formae speciales showed that alle les per locus of F oxysporum f. sp. radicis lycopersici (6.7 alleles) were hi gher than those of F oxysporum f. sp. lycopersici (5 alleles), indicating a higher level of genetic diversity in F oxysporum f. sp. radicis lycopersici Also, the considerably larger number of VCGs reported in F oxysporum f. sp. radicis lycopersici m ay reflect its high ge netic diversity (80,81). Molecular markers, such as RFLP and RAPD s, have been used to investigate the evolutionary origins of VCGs of F oxysporum f. sp. radicis lycopersici and F oxysporum f. sp. lycopersici (10,38,139). To our knowledge, microsatellites have not been used to investigate genetic relatedness of these two formae speciales A dendrogram generated from 27 microsatellite data exhibited four clusters: FORL clusters I and II, and FO L clusters I and II. In general, isolates within the same VCG are more closely related than others within different

PAGE 42

42 VCGs, confirming previous results derived from other molecular markers (10,37,38). VCGs 0090, 0091, 0092, and 0093 of FORL cluster I were not found in this st udy, whereas VCGs 0090 and 0091 were previously reported in Nort h America and VCGs 0092, 0093, 0096 were found in Israel and Europe (37,80). In contrast, isolates of F oxysporum f. sp. radicis lycopersici used for this study were grouped into FORL cl uster II in which VCGs 0094, 0098, and 0099 are predominant. This change in population structure shows that the three VCGs replaced previous dominant VCGs 0090 and 0091 in Florida in agreem ent with a previous study by Rosewich et al. (139). According to the dendrogram analysis, FO RL clusters I and II probably evolved from different progenitor populations. Regarding the evolution of F oxysporum f. sp. lycopersici VCGs 0030 and 0032 shared higher genetic si milarity than VCGs 0031, 0033, and 0035. Like F oxysporum f. sp. radicis lycopersici two probable ancestral progenitors were suggested in F oxysporum f. sp. lycopersici, corresponding to previous stud ies (37,38). Interestingly, VCG 0033 of F oxysporum f. sp. lycopersici and VCGs 0094, 0098, and 0099 of F oxysporum f. sp. radicis lycopersici showed a high degree of genetic simila rity and were all found in this study. Molecular Diversity Indices The m ean number of pairwise differences and average gene diversity were compared for three Florida populations (Table 2-8). Isolat es from Hendry County had a higher mean number of pairwise differences and aver age gene diversity when either whole microsatellite data or clone-corrected data was used for analysis proba bly because it was the sink of migrants from the other two counties according to analysis of historical migration (Figs. 2-1 and 2-4). VCG 0098 was not previously reported in Hendry County befo re this study. Pairwise Fst values of VCGs 0098 between isolates from the three counties exhi bited no genetic differentiation. Moreover, its nucleotide diversity in Hendry County was not highe r than in the two othe r counties. Therefore, genotype flow from other tomato-growing regi ons may be the reason VCG 0098 appeared in

PAGE 43

43 Hendry County. However, the possibility of VCG 0098 arising from either another local VCG or nonpathogenic F oxysporum cannot be completely ruled out as two isolates in this study were found to form heterokaryons with testers of VCGs 0094 and 0098. This cross vegetative compatibility indicated that both of these two VCGs belong to the same clonal lineage and may have diverged by somatic processes (38,55). Furthe r analysis based on the coalescent theory (137) may provide more insights about the evolution of VCGs 0094 and 0098. Pairwise Fst values of VCG 0094 or 0098 am ong these isolates from three counties suggested no genetic differentiation (Table 26 and Table 2-7). Measur es of intrapopulation diversity (mean number of pairwise differences and average gene di versity) were used to infer the origin of these two VCGs according to a ba sic assumption that a new population exhibits a lower level of molecular diversity as a result of the founder effect and less accumulation of mutations (139). To evaluate these two indices for each VCG, the clone-corrected data set was used to further classify each VCG and its aff iliated geographical region. The two indices were the highest in Collier County for VCGs 0094 and 0098, suggesting that these two VCG may have originated in Collier County and then migr ated to Manetee and Hendry Counties. For those isolates not assigned to known VC Gs, the two indices were highe st in Hendry County. Regarding individual VCG, results of clone-corrected data showed that VC G 0099 had a higher mean number of pairwise differen ces and average gene diversity than VCGs 0094 and 0098, while VCG 0094 was predominant in Florida. VCG 0099 probably preexisted in Florida relative to VCGs 0094 and 0098; however, its frequency declined due to being le ss selectively fit in comparison to the others (139). Hierarchical Distribution of Total Gene Diversity Although no genetic differentiation was re vealed for VCGs 0094 and 0098, The AMOVA for all isolates of F oxysporum f. sp. radicis lyco persici suggested a significant partition of

PAGE 44

44 variance among counties. VCG 0099 in Collier County and other nonassigned VCG isolates in these three counties may have contributed to the significant genetic variation among counties. Moreover, most variance occurred within populati ons (84%), indicating th at genotype flow was higher among the counties than within the counties. Linkage Disequilibrium The F oxysporum species com plex has been viewed as an extremely mitosporic species, whereas it comprises a broad diversity of formae speciales that are able to infect numerous hosts (55,97). Phylogenetic analyses showed that th is species complex is monophyletic but most formae speciales are polyphyletic (119). It has been suggested that r ecombination is possible in this species complex in addition to clonal re production (168). Since Florida, where two new VCGs, 0098 and 0099, were first repor ted, has been considered as th e probable center of origin of VCG 0094, this study tried to determine linkage disequilibrium for understanding the possibility of recombination in the field although its sexual stage has not been reported. Results showed that F oxysporum f. sp. radicis lycopersici sampled from these three counties exhibited linkage disequilibrium according to index of association (Table 2-10), but up to 62.3% of pairwise loci for isolates of Collier County were at linkage equilibrium, suggesting random association of alleles. Even in a mitosporic f ungus, parasexual recombin ation and reassortment resulting from protoplast fusion can cause chro mosome rearrangement (170), contributing to alleles at one locus to be random ly associated. However, other processes, such as selection, gene/genotype flow, drift, and linkage, can al so cause linkage disequilibrium (113). Crossing isolates from Collier County carrying different mating-type idiomorphs may provide a direct confirmation of recombination.

PAGE 45

45 Population Admixture and Historical Migration The best value of K clusters (real populations) w as three, su ggesting that defining a county as a population was reasonable in this study. K = 3 was used to determine possible admixture in the three populations. Eighteen likely admixed i ndividuals were identified, and most of them were found in Hendry County (50% of the total) (Fig. 2-4). Highly sign ificant admixture found in Hendry County suggests that some alleles fou nd in isolates were most likely drawn from Manatee and Collier Counties and that alleles were exchanged among populations. Moreover, historical migration revealed that more immi grants moved into Hendry County than into the other two counties (Table 2-11). Because high gene /genotype flow can enhance genetic diversity of a population as a result of increasing the size of ge netic neighborhood (107,108), it may partially explain why Hendry County had a higher genetic diversity of F oxysporum f. sp. radicis lycopersici Given that a population wi th high genetic diversity ma y have more alleles to overcome a resistance gene (107,10 8), lowering genotype flow of F oxysporum f. sp. radicis lycopersici among these three counties is necessary to re duce the break-down risk of a resistance gene. Moreover, F oxysporum f. sp. radicis lycopersici has been considered as a soilborne, airborne, and waterborne pa thogen (59,74,133,140). Therefore, me thods involved in reducing genotype flow would have to th oroughly evaluate unique epidemio logical characteristics of this pathogen (82). Natural dispersal of pathogen prop agules by wind, water, a nd insects is difficult to restrict. However, long distance dispersa l by man may be limited by means of inspecting infected plant material, soil, or contaminated equipment (107,108). Population Growth Rate Since population growth rates, g, with 95% confidence interval s, of Manatee and Hendry Counties included zero, the population size of F oxysporum f. sp. radicis lycopersici in these two counties have experienced little or no growth (90,91). In contrast, g of Collier C ounty did

PAGE 46

46 not contain zero, suggesting a growing population. The genetic structure of a population is dynamic and determined by five evolutionary for ces: mutation, genetic drif t, gene/genotype flow, reproduction/mating system, and selection ( 107,108). A pathogen population with a high population size and high standing genetic diversity may have greater potential to break down a host resistant gene thro ugh mutation. Therefore, in addition to limiting genotype flow among these three counties, redu cing the population size of F oxysporum f. sp. radicis lycopersici through disease management in these three coun ties is another important consideration to decrease the evolutionary potential of the pathogen. Utilizing regular crop rotations and avoiding extremely susceptible cultivars are two simple ways to minimize pathogen population size (107,108). Lettuce has been suggested for using in rotation and intercroppi ng with tomato due to evidence of phenolic compounds from lettuce inhibiting the growth of F oxysporum f. sp. radicis lycopersici (73,74).

PAGE 47

47 Table 2-1. Tester strains of ve getative compatibility groups of Fusarium oxysporum f. sp. radicis lycopersici and F oxysporum f. sp. lycopersici used in this study VCG Tester strainsz References F oxysporum f. sp. radicis lycopersici 0090 I FRC-O-1090/1, FRCO-1090/B (80), (128) 0090 II FORL-C-407/1, FORL-C415/3 (80) 0090 III FORL-C696A/3, FORL-C710/A62 (82) 0091 II FORL-C69E3 (80) 0094 I 01150-6, 01152-31 (82) 0094 II D69/7, D69/19 (82) 0094 III 26787/6, 26787/10 This study 0094 IV CL-2/36, CL-2/40 This study 0096 FORL-C624A/3, FORL-C622A/6 (80) 0098 CL-7/6, CL-75/4 (139) 0099 LE-20/11, LE-20/12 This study F oxysporum f. sp. lycopersici 0033 DA-1/7 (139) z Testers were provided by H. C. Kistler, except those of VCGs 0094 III, 0094 IV, and 0099 developed in this study.

PAGE 48

48 Table 2-2. Isolates used for de veloping microsatellite primers Isolatex Speciesy VCG Race Origin Sourcez Sampling year NRRL 26771 (0-1090) FORL 0090 I Canada e 1975 NRRL 26772 (FORL-C63F) FORL 0090 I Israel e 1988 NRRL 26773 (CRNK-11C) FORL 0090 II Israel e 1988 NRRL 26774 (FORL-C726) FORL 0090 II Israel e 1990 NRRL 26775 (FORL-C696A) FORL 0090 III Israel e 1990 NRRL 26776 (FORL-C710B) FORL 0090 III Israel e 1990 NRRL 26777 (FORL-1) FORL 0091 I Israel e 1986 NRRL 26778 (FORL-C75) FORL 0091 I Israel e 1988 NRRL 26779 (FORL -C69E) FORL 0091 II Israel e 1988 NRRL 26780 (FORL-FA2) FORL 0091 II France e 1985 NRRL 26781 (CRNK-78) FORL 0092 Israel e 1988 NRRL 26782 (FORL C201) FORL 0093 Israel e 1988 NRRL 26783 (FORL-C202) FORL 0093 Israel e 1988 NRRL 26884 (01152) FORL 0094 I Belgium e 1987 HE-0616 FORL 0094 I Florida a 2006 HE-0631 FORL 0094 I Florida a 2006 HE-0611 FORL 0094 I Florida a 2006 NRRL 26786 (FORL-D69) FORL 0094 II UK e 1990 NRRL 26787 (FORL149/74R/89) FORL 0094 III UK e 1988 CL-2 FORL 0094 IV Florida c 1995 MN-0713 FORL 0094I/0098 Florida a 2007 MN-0724 FORL 0094I/0098 Florida a 2007 NRRL 26788 (FORL-C 623) FORL 0096 Israel e 1990 NRRL 26789 (FORL-C 624A) FORL 0096 Israel e 1990 CL-0601 FORL 0098 Florida a 2006 CL-06122 FORL 0098 Florida a 2006 MN-0630 FORL 0098 Florida a 2006 CL-0626 FORL 0099 Florida a 2006 CL-06196 FORL 0099 Florida a 2006

PAGE 49

49 Table 2-2. Continued Isolatex Speciesy VCG Race Origin Sourcez Sampling year CL-06202 FORL 0099 Florida a 2006 CL-06220 FORL 0099 Florida a 2006 93-193 FORL Florida d 1993 JBF6 FOL 0030 2 Florida b BE1(5397) FOL 0030 3 Florida b F189 FOL 0031 2 California b OSU451 FOL 0031 2 Ohio b MM59 FOL 0032 2 Arkansas b 1993 MM61 FOL 0032 2 Arkansas b 1993 MM62 FOL 0032 2 Arkansas b 1993 MM64 FOL 0032 2 Arkansas b 1993 MN-24 FOL 0033 3 Florida c 1996 DA-1 FOL 0033 3 Florida c 1996 MN-0619 FOL 0033 3 Florida a 2006 MN-0805 FOL 0033 3 Florida a 2008 DF0-23 FOL 0035 2 California f 1989 NRRL 22903 F communce e NRRL 28387 F communce e x Previous isolate designa tion in parentheses (80,105). y FORL = Fusarium oxysporum f. sp. radicis lycopersici ; FOL = F oxysporum f. sp. lycopersici. z a = this study; b = J. C. Correll; c = H. C. Kistler; d= R. J. McGovern; e = K. ODonnell; f = R. W. Schneider.

PAGE 50

50 Table 2-3. Isolates used for st udying the population genetics of Fusarium oxysporum f. sp. radicis lycopersici in Florida Isolate VCG Tomato cultivar County MN-0630 0098 FL 91 Manatee MN-0631 0094 FL 91 Manatee MN-0632 0094 FL 91 Manatee MN-0633 0098 FL 91 Manatee MN-0634 0094 FL 91 Manatee MN-0635 0098 FL 91 Manatee MN-0701 0098 ?x Manatee MN-0702 0094 ? Manatee MN-0703 0094 ? Manatee MN-0704 0098 ? Manatee MN-0705 Nonassigned ? Manatee MN-0706 Nonassigned ? Manatee MN-0707 0098 ? Manatee MN-0708 0098 ? Manatee MN-0709 Nonassigned ? Manatee MN-0713 0094/0098 Marriana Manatee MN-0714 Nonassigned Marriana Manatee MN-0715 0098 Marriana Manatee MN-0716 Nonassigned ? Manatee MN-0717 0098 ? Manatee MN-0718 0098 ? Manatee MN-0719 0094 ? Manatee MN-0720 Nonassigned ? Manatee MN-0721 0094 ? Manatee MN-0722 0098 ? Manatee MN-0723 0098 ? Manatee MN-0724 0094/0098 ? Manatee MN-0729 0094 Beauty Manatee MN-0730 0094 Beauty Manatee MN-0731 Nonassigned Beauty Manatee MN-0732 0094 Tygress Manatee MN-0733 0094 Tygress Manatee MN-0734 0094 BHN-745 Manatee MN-0735 Nonassigned BHN-745 Manatee MN-0801 Nonassigned FL47 Manatee MN-0802 0094 FL47 Manatee MN-0803 Nonassigned FL47 Manatee HE-0601 0094 Marriana Hendry HE-0602 0094 Marriana Hendry HE-0603 0098 Marriana Hendry HE-0604 0094 Marriana Hendry HE-0605 0094 Marriana Hendry

PAGE 51

51 Table 2-3. Continued Isolate VCG Tomato cultivar Origin HE-0606 Nonassigned Marriana Hendry HE-0607 Nonassigned Marriana Hendry HE-0609 0094 Marriana Hendry HE-0610 0098 Marriana Hendry HE-0611 0094 Marriana Hendry HE-0612 0094 Marriana Hendry HE-0614 0098 Marriana Hendry HE-0615 Nonassigned Marriana Hendry HE-0616 0094 Marriana Hendry HE-0619 0094 Marriana Hendry HE-0621 0094 Marriana Hendry HE-0622 Nonassigned Marriana Hendry HE-0623 Nonassigned Marriana Hendry HE-0624 0094 Marriana Hendry HE-0625 Nonassigned Marriana Hendry HE-0626 0094 Marriana Hendry HE-0627 0094 Marriana Hendry HE-0628 0094 Marriana Hendry HE-0630 Nonassigned Marriana Hendry HE-0631 0094 FL 47 Hendry HE-0632 Nonassigned FL 47 Hendry HE-0633 0098 FL 47 Hendry HE-0634 0094 FL 47 Hendry HE-0635 0094 FL 47 Hendry HE-0636 Nonassigned FL 47 Hendry HE-0637 0098 FL 47 Hendry HE-0638 Nonassigned FL 47 Hendry HE-0639 Nonassigned FL 47 Hendry HE-0640 0094 FL 47 Hendry HE-0641 0094 FL 47 Hendry HE-0801 Nonassigned FL 47 Hendry HE-0802 Nonassigned FL 47 Hendry HE-0803 Nonassigned FL 47 Hendry HE-0804 Nonassigned FL 47 Hendry CL-0601 0098 Grape tomato Collier CL-0602 0098 Grape tomato Collier CL-0603 0094 Grape tomato Collier CL-0612 0098 Grape tomato Collier CL-0613 Nonassigned Grape tomato Collier CL-0618 0094 FL 47 Collier CL-0620 0098 FL 47 Collier CL-0623 ? FL 47 Collier CL-0626 0099 FL 47 Collier

PAGE 52

52 Table 2-3. Continued Isolate VCG Tomato cultivar Origin CL-0668 0098 Grape tomato Collier CL-0672 0098 Grape tomato Collier CL-0673 0094 Grape tomato Collier CL-0678 Nonassigned Grape tomato Collier CL-0686 Nonassigned Grape tomato Collier CL-0690 0094 Grape tomato Collier CL-0692 Nonassigned Grape tomato Collier CL-0696 Nonassigned Grape tomato Collier CL-0697 0099 Grape tomato Collier CL-06118 0098 ? Collier CL-06120 Nonassigned ? Collier CL-06122 0098 ? Collier CL-06124 0094 ? Collier CL-06128 0098 ? Collier CL-06135 0094 ? Collier CL-06136 0094 ? Collier CL-06137 0094 ? Collier CL-06139 Nonassigned ? Collier CL-06140 0098 ? Collier CL-06142 Nonassigned ? Collier CL-06152 0094 ? Collier CL-06171 0094 FL 47 Collier CL-06175 0094 FL 47 Collier CL-06176 0099 FL 47 Collier CL-06182 Nonassigned FL 47 Collier CL-06186 Nonassigned ? Collier CL-06190 Nonassigned ? Collier CL-06191 0099 ? Collier CL-06196 0099 ? Collier CL-06197 0099 ? Collier CL-06201 Nonassigned Grape tomato Collier CL-06202 0099 Grape tomato Collier CL-06203 0094 Grape tomato Collier CL-06210 0099 Grape tomato Collier CL-06212 0099 Grape tomato Collier CL-06214 0098 Grape tomato Collier CL-06220 0099 FL 47 Collier CL-06222 0098 FL 47 Collier CL-06224 0094 FL 47 Collier CL-06230 Nonassigned FL 47 Collier x ? = unknown cultivar due to confiden tiality of the tomato growers.

PAGE 53

53 Table 2-4. Frequency of vegetative compatibility groups (VCGs) of Fusarium oxysporum f. sp. radicis lycopersici sampled from 2006 to 2008 in Florida VCG (%) Population Number of isolates 0094 0098 0099 Nonassigned Manatee 37z 38.5 35.9 0 25.6 Hendry 39 48.7 12.8 0 38.5 Collier 49 28.6 24.5 20.4 26.5 Overall 125 38.6 24.4 6.8 30.2 z Isolates MN-0713 and MN-0724 were assigned to 0094 and 0098 after repeating two singlespored isolations and confirming vegetative compatibility with test ers of 0094I (01150-6 and 01152-31) and 0098 (CL7/6 and CL-75/4).

PAGE 54

54 Table 2-5. Characteristics of 27 mi crosatellite loci derived from a bioinformatics approach and a microsatellite enrichment pro cedure for 13 isolates of Fusarium oxysporum f. sp. lycopersici and 32 isolates of F oxysporum f. sp. radicis lycopersici Locusv Primer sequence (5 -3 )w Repeat motif Size (bp)x Number of alleles Gene diversityy GenBank accession no. FOL FORL FOL FORL CH2-9 F: GGGCTTCAAGGTGCTGAGTA R: TAAACGAAGCTGGGAATGGA (GTGA)7 FJ882019 216 2 6 0.36 0.73 CH2-15 F: ATCTTCCTCACGGTTTTGGA R: TGTAGCGTAGCACAACAGTGG (CT)8 FJ882020 203 8 7 0.92 0.82 CH2-16 F: GGGCTTCAAGGTGCTGAGTA R: AGCTGGGAATGGAAATTTGA (TGAG)8 FJ882021 209 1 2 0 0.51 CH2-31 F: CGACAGGAGGCTGAGGAGTA R: CGTCAATTGAGAACCATCCA (CT)5CG(CT)5FJ882022 191 4 4 0.74 0.77 CH2-51 F: ATACGAGCACAAGGGACGAG R: CATCCATTCCGTCTCCATTT (AG) 8 FJ882023 243 5 7 0.76 0.77 CH2-66 F: GAAGCGCTTACAGTGCCAAT R: CCCTTGACTCTCCACGAAAC (AG) 14 FJ882024 236 4 10 0.76 0.88 CH2-71 F: TGTAGCGTAGCACAACAGTGG R: ATCTTCCTCACGGTTTTGGA (AG)8G(GA)3 FJ882025 203 4 2 0.74 0.52 FOL15 F: TATGGACGGATCAGGAAAGG R: TCAACAACGCACTGAAGACC (AAG)17 -z 236 6 9 0.83 0.85 FOL20 F: CATTGAGGAAGAGCGGAAAG R: CACATTTGGCACAGCAATCT (AGCAC)19 269 9 9 0.88 0.79 FOL35 F: GTCGTTTTCAAGGACGCACT R: GGTGGCAGTTTCCTCCTTTT (GAA)19 269 7 14 0.88 0.91 FOL89 F: ATATCCCACCCTCCTTGCAT R: TGCTCAGTCTCGTCACAACC (TGTAT)9 263 4 5 0.78 0.71 FOL99 F: AGTTGAGGTTGTCGCTGGTT R: CTATTCCTCCCGCTGCAC (CTC)11CCC( CTC)3 257 6 4 0.88 0.60 FOL104 F: GGAACCCGAAACCACCTTAT R: ATTGGCACTTGCTTTGCTTT (GTGCCT)8 242 2 4 0.53 0.68

PAGE 55

55 Table 2-5. Continued Locusv Primer sequence (5 -3 )w Repeat motif Size (bp)x Number of alleles Gene diversityy GenBank accession no. FOL FORL FOL FORL FOL118 F: GAGGCCACAGAAATGAGAGC R: GGTTGGTTGGAAGGATTCAG (CAGGA)9 -z 196 3 3 0.67 0.17 FOL175 F: ACCAGCAAGCAGCTTCATTT R: ACGAGTCGCAGGGTATCAAC (GAAT)13 255 6 6 0.87 0.79 FOL185 F: TCCTTG TAGCGCCTTGCTAT R: AAATGCAACACCGCACTGTA (AGGT)15 267 4 6 0.74 0.81 FOL245 F: TGAAAGGCGCGCTATTTAGT R: GAGAGGCGGAGGAAGAAGA (CTT)14 258 5 10 0.80 0.87 FOL293 F: AAGACTCGGGCAAGTCAGAA R: ATTTTCGTAAACCCCATCCA (TTTCT)29 379 7 3 0.88 0.57 FOL296 F: CACTGAAGGAAATGCAGCAG R: TAGGCTCTGGAGATGCTTGG A23(AAG)22 234 6 10 0.86 0.72 FOL338 F: GAACCCTTTCCCACGAGAC R: AACTCGCTGTTGGTGATGTG (TGATT)10 216 3 6 0.67 0.81 FOL356 F: CCTCCTGCTCTTCCTCATCTT R: CGGTATTGTTGGGGGTTTAG (CAA)14 261 6 9 0.87 0.87 FOL602 F: CTCGTCACTGCTGGAATCAA R:TGTCAAAGAATGGCCCATATTA (AGT)15 248 6 6 0.81 0.69 FOL624 F: CAAGAGGCCAGCGATAGTGT R: AGCTTTTGATACCCCATTCG (GTA)31 240 4 12 0.70 0.73 FOL638 F: GAAGCACTCGCTACGTGTCA R: CGGTTGTGCAGCTCAAATAA (TGAGA)11 244 4 5 0.78 0.73 FOL665 F: ACCCTGGGTACTCCGGTTAT R: GGCGCAGCTTCAAGACTAAT (GT) 25 213 2 8 0.54 0.81

PAGE 56

56 Table 2-5. Continued v Locus indicated by a CH2-number was isolated directly from isolate HE-0631 (VCG 0094) of F oxysporum f. sp. radicis lycopersici (FORL). Loci indicated by an FOL-contig number were identified by a search of the genome sequence of F oxysporum f. sp. lycopersici (FOL) at http:// www.broad.mit.edu. w M13 tag (CACGACGTTGTA AAACGAC) added to 5 end of forward primers for amplifica tion with fluorescently labeled M13. x Based on the genome sequence of isolate HE-0631 and F oxysporum f. sp. lycopersici. y Based on Nei (115). z = microsatellite loci derived fr om searching the genome sequence of F oxysporum f. sp. lycopersici. Locusv Primer sequence (5 -3 )w Repeat motif Size (bp)x Number of alleles Gene diversityy GenBank accession no. FOL FORLFOL FORL FOL680 F: CGCAGAATGGCTCTTCAAAT R: TGCAACATCATCGACCACTT (TTTA)11 -z 254 10 12 0.95 0.84 FOL803 F: GTGGTAGCGTGGAGTGGATT R: GTTCGACATTCGCTCGAGTT (AGACA)11 240 7 4 0.87 0.64

PAGE 57

57 Table 2-6. Pairwise Fst values (above diagonal) and their P values (below diagonal) between populations of Fusarium oxysporum f. sp. radicis lycopersici VCG 0094 from three counties in Florida Manatee Hendry Collier Manatee -0.03 0.09 Hendry 0.55 0.06 Collier 0.12 0.11 Table 2-7. Pairwise Fst values (above diagonal) and their P values (below diagonal) between populations of Fusarium oxysporum f. sp. radicis lycopersici VCG 0098 from three counties in Florida Manatee Hendry Collier Manatee -0.10 0.18 Hendry 0.52 0.05 Collier 0.15 0.30

PAGE 58

58 Table 2-8. Mean number of pairwise differe nce and average nucleotide di versity of microsatellite haplotypes for isolates of Fusarium oxysporum f. sp. radicis lycopersici from three counties in Florida VCGw Locality Sample size Polymorphic locix Mean number of pairwise differences and SDy Nucleotide diversity and SDz Manatee 15 1 1.00 (0.79) 0.10 (0.10) 0094 Hendry 19 2 1.27 (0.92) 0.13 (0.09) Collier 14 4 2.20 (1.45) 0.22 (0.11) Total Florida clone-corr ected 10 4 1.75 (1.11) 0.18 (0.10) Manatee 14 1 1.00 (0.89) 0.10 (0.10) 0098 Hendry 5 1 1.00 (1.00) 0.10 (0.10) Collier 12 2 1.33 (1.09) 0.13 (0.09) Total Florida clone-corr ected 6 2 1.47 (1.03) 0.15 (0.10) 0099 Collier 10 3 2.00 (1.30) 0.20 (0.15) Total Florida clone-corr ected 5 3 2.20 (1.45) 0.22 (0.12) Manatee 10 10 6.21 (3.30) 0.62 (0.03) Nonassigned Hendry 15 10 6.48 (3.30) 0.65 (0.06) Collier 13 5 3.00 (1.71) 0.30 (0.11) Manatee 37 10 2.12 (1.21) 0.21 (0.05) Overall Hendry 39 10 3.44 (1.80) 0.34 (0.06) Collier 49 6 1.99 (1.14) 0.19 (0.08) Overall clonecorrected Manatee 25 10 2.90 (1.58) 0.29 (0.06) Hendry 26 10 3.99 (2.06) 0.39 (0.07) Collier 34 6 2.45 (1.36) 0.24 (0.10.) w Vegetative compatibility group x Among a total of ten microsatellite loci. y Mean number of pairwise differences and standard deviation (SD) (in parenthe ses) based on Tajima (165). z Average nucleotide diversity and SD (in pare ntheses) according to Nei (115) and Tajima (163).

PAGE 59

59 Table 2-9. Hierarchical di stribution of gene diversity among populations of Fusarium oxysporum f. sp. radicis lycopersici from Florida Source of variancey df Sum of squares Variance components Percentage of variation Fixation index Pz Among populations 2 18378 276 16 s t =0.159 <0.0001 Within populations 82 118887 1450 84 Total 84 137266 1726 y Molecular analysis of varian ce (AMOVA) was performed using ARLEQUIN versio n 3.1 (45). Distance method was according to the sum of squared size differences (R st) between two haplotypes for microsatellite data (156). z Based on 1,023 permutations.

PAGE 60

60 Table 2-10. Tests for random association of alleles within each locus and between pairs of loci in the population of Fusarium oxysporum f. sp. radicis lycopersici from three counties in Florida Population IA x P d xy P GDz Manatee 3.19 <0.001 0.36 <0.001 42/45 Hendry 3.56 <0.001 0.40 <0.001 45/45 Collier 1.26 <0.001 0.15 <0.001 17/45 Overall 2.40 <0.001 0.27 <0.001 45/45 x IA and d are indices of multilous linka ge disequilibrium (157). y d is a modification of IA and independent of sample numbers of loci (1). z Pairs of loci at significant linkage disequilibr ium according to Fishers exact test implemented in GENEPOP version 3.4 (131). Table 2-11. Pairwise numbers of mi grants per generation inferred from Bayesian analyses implemented in MIGRATE-N Population receiving migrantsz Population donating migrants Manatee Hendry Collier Manatee 2.75 (0-5.50) 1.33 (0-3.50) Hendry 2.18 (0-4.50) 1.65 (0-3.50) Collier 1.94 (0-4.50) 2.15 (0-4.00) z Values in parentheses indi cate 95% confidence intervals.

PAGE 61

61 Figure 2-1. Map of three Florida populations of Fusarium oxysporum f. sp. radicis lycopersici analyzed in this study and distribution of vegetative compatibility groups (VCGs). Width of section is proportional to the per centage of isolates a ssigned in each VCG. Population differentiation ( st) is labeled on each population. Migrants per generation between populations are labeled ne xt to arrows. Directions of arrows indicate migration between sink and source populations.

PAGE 62

62 Figure 2-2. Identification of Fusarium oxysporum f. sp. radicis lycopersici (FORL) and Fusarium oxysporum f. sp. lycopersici (FOL) using FORL-specific primers sprlf and sprlr (63) (top panel) an d FOL-specific primers P12-F2B and P12-R1(172) (bottom panel), repectively. Vegetative compa tibility groups (VCGs) are shown in parentheses of isolates. Sprlf and sprlr amplify a fragment of FORL exo polygalacturonase gene ( pgx4 ), and P12-F2B and P12-R1 target a virulence gene of FOL, SIX1 (secreted in xylem 1).

PAGE 63

63 Figure 2-3. Neighbor-joining tree based on Neis minimum gene tic distance (115) between individuals. Four clusters are indicated including Fusarium oxysporum f. sp. radicis lycopersici (FORL) clusters I and II, and Fusarium oxysporum f. sp. lycopersici (FOL) clusters I and II.

PAGE 64

64 Figure 2-4. Estimated population structure inferred from multilocus micr osatellite data of Fusarium oxysporum f. sp. radicis lycopersici using STRUCTURE according to membership coefficient. Each individual is represented by a thin vertical line, which is partitioned into three colored segments fr om which alleles are most likely derived. Black lines separate individuals of differe nt populations that are labeled below the figure. The bar length suggests its membersh ip coefficient (Q) to variously colored populations. Individuals marked with an as terisk along the top ar e of statistically significant admixture.

PAGE 65

65 CHAPTER 3 PHYLOGENETIC AND MATING-TYPE ANALYSES OF Fusarium oxysporum f. sp. radicis lycopersici AND T HEIR ASSOCIATION WITH VIRULENCE Introduction Fusarium oxysporum Schlechtend.: Fr. is a cosm opolitan soilborne plant pathogen infecting a wide range of economically impor tant plant hosts (8,31,97). More than 100 hostspecific formae speciales have been described in the F oxysporum species complex (97). Fusarium oxysporum f. sp. radicis lycopersici and lycopersici are two important fungal plant pathogens of tomato and cause Fusarium crown and root rot and Fusarium wilt, respectively (72,74,174). Three races have been reported for F oxysporum f. sp. lycopersici whereas no physiological races are currently known for F oxysporum f. sp. radicis lycopersici (75,125). Vegetative compatibility groups (VCGs) have been useful in char acterizing the population structure of these two formae speciales (19,139). Compared to F oxysporum f. sp. lycopersici a number of VCGs have been described in F oxysporum f. sp. radicis lycopersici (36,80). It has been suggested that isolates within the same VCG belong to the same clonal lineage (55). A forma specialis is often assumed to be monophyletic in which all VCGs and races have been derived from the same common ancestor. However, multiple VCGs and races within a given forma specialis could evolve from multiple independent origins, which shows paraor polyphyly (8,119). In other words, isolates of a forma specialis could be more closely related to members of other formae speciales of F oxysporum than those of the same forma specialis Previous studies showed that F oxysporum f. sp. radicis lycopersici and lycopersici are paraphyletic, suggesting that VCGs of these two formae speciales within the same phylogenetic clade (lineage) may have been derived from a common ancestor. These studies inferred the phylogeny of these two formae speciales according to partial regi ons of the gene encoding translation elongation factor 1 (EF-1 ), m itochondrial small subunit (mtSSU) rDNA, and

PAGE 66

66 itergenic spacer region (IGS) (19,119). It may be of interest to further investigate whether adding more characters, such as the entire IGS region and microsatellite loci, increases the phylogenetic resolution for these two formae speciales Moreover, the inclusion of too few VCGs of F oxysporum f. sp. radicis lycopersici in previous phylogenetic studi es might have resulted in an incomplete inference in their phylogeny. Phylogenetic analyses has revealed that the F oxysporum species complex is closely related to the Gibberella clade, even though the teleomorph of the F oxysporum species complex is unknown (117,118). Results of RT-PCR analysis re vealed that mating type (MAT) genes of F oxysporum are still expressed and pro cessed correctly (181). Two isol ates must be in the same biological species and carry distinct mating-type alleles, MAT-1 or MAT-2 for a sexual cross to happen, whereas vegetative incomp atibility does not necessarily pr event the sexual cycle (95,97). Mating types of various VCGs of F oxysporum f. sp. radicis lycopersici have not been well investigated. Moreover, crossing F oxysporum f. sp. radicis lycopersici isolates with opposite mating types has not been done. Pathogenicity is the qualitative ab ility of a parasite to infect and cause disease on a host. In contrast, virulence is defined as the degree of damage caused by a parasite to a host and it is suggested to be negatively correlated with host fitness (143). Some comme rcial tomato cultivars carry the Frl resistance gene against Fusarium crown a nd root rot, but physiological races have not been found for F oxysporum f. sp. radicis lycopersici (151). It is not clear how the pathogen adapts to different resistant tomato cultivars by means of adjusting its virulence. There is no universal relationship between parasite reproduc tive capacity and virulenc e (143), but a positive relationship has been observed in some pat hosystems (65,114). Therefore, pathogens may increase virulence for accelerating their spread (65). Virulence tests and phylogenetic analyses

PAGE 67

67 for reported VCGs of F oxysporum f. sp. radicis lycopersici would shed insight into how virulence evolves among different phylogenetic lin eages and within a phylogenetic lineage while under the selection pressure from the host. This study tested the hypothesis that virulence was significantly different among VCGs within the sa me phylogenetic lineage despite already being pathogenic on tomato. Fusarium crown and root rot was first found in greenhouses of tomatoes in Japan in 1969. The causal agent was identifie d as a new race (J3) of F oxysporum f. sp. lycopersici (146). However, according to symptomology, etiology and pathogenesis, it was renamed as a new forma specialis of F oxysporum and designated as F oxysporum Schlecht. f. sp. radicis lycopersici Jarvis & Shoemarker (72). Although F oxysporum f. sp. radicis lycopersici is known as a relatively new pathogen, its divergence time has not been estimated based on sequence data. Moreover, its demographic hist ory has not been well known si nce it was identified in 1969. Recently, a new approach to address these evol utionary questions has been proposed using Bayesian inference for microbial pathogens sample d at different dates (153). The approach might provide insights into the evolutionary history of F oxysporum f. sp. radicis lycopersici. The purpose of this study was four-fold (i) to investigate classification of VCGs of F oxysporum f. sp. radicis lycopersici and lycopersici according to multilocus DNA sequence data; (ii) to evaluate the occurrence of opposite ma ting types within VCGs and the possibility of sexual recombination based on crossi ng isolates carrying distinct ma ting-type idiomorphs; (iii) to compare association of virulence within VCGs w ith phylogenetic an alyses; (iv) to determine the nucleotide substitution rate and population dynamics of F oxysporum f. sp. radicis lycopersici. A preliminary report of this work has been published (68).

PAGE 68

68 Materials and Methods Collection of Isolates Thirty one isolates of F oxysporum f. sp. radicis lycopersici and 13 isolates of F oxysporum f. sp. lycopersici that comprised m ost of the known VCGs of these two formae speciales (Table 2-2) were included in the molecular studies. In addition, thre e isolates of VCGs 0098, CL-0620, HE-0610, and HE-0620 were also incl uded as ingroups. Six outgroups including F commune NRRLs 22903 and 28387, F redolens NRRL 31075, F hostae NRRL 29889, F subglutinans NRRL 22016, and F foetens NRRL 31852 were used for phylogenetic analyses. Each isolate was single-spored and st ored on sterile paper and/or at -80 C in 15% glycerol. DNA Extraction and Polym erase Chain Reaction Monosporic isolates from the culture collect ion were grown on carna tion leaf agar (CLA) before transferring m ycelia plugs to potato de xtrose agar (PDA) for DNA extraction. About 200 mg mycelia were ground in liquid nitrogen befo re using DNeasy Plant Minikits (Qiagen, Inc., Valencia, CA) to extract DNA. Amplification of an approximate 690-bp fragment of the EF-1 gene was performed using primers ef1 and ef2, and polymerase chain reaction (PCR) as previously described (119). The nuclear ribosomal IGS region ( 3 kb) was amplified using primers NL11 and CNS1 (3). The sequence of a noncoding microsatellite locus FOL185 (Table 2-5) was used to redesign primers AGGTf (CCATCTTTCCGTCTCCACAT) and AGGTr (TTGCGCAAAGTTTGAATGAG) A fragment of 786 bp was amplified using the following PCR conditions: 94C for 3 min, 34 cycles at 94 C for 30 s, 56.5 C for 30 s, and 72 C for 45 s and a final elongation step at 72 C for 10 min.

PAGE 69

69 Direct Sequencing Prim ers ef1 and ef2 were used to sequen ce the amplification products of the EF-1 gene. Four internal sequencing primers were used to sequence the IGS rDNA region: iNL11 (5AGGCTTCGGCTTAGCGTCTTAG-3), NLa (5-TCTAGGGTAGGCKRGTTTGTC-3), iCNS1 (5-TTTCGCAGTGAGGTC GGCAG-3), and CNSa (5TCTCATRTACCCTCCGAGA CC-3) (K. ODonnell, unpublished data ). Primers AGGTf and AGGTr were used for sequencing the microsatel lite locus. Sequences were obtained using BigDye Terminator Cycle Sequencing Chemis try and ABI 3730 XL DNA Sequencer (Applied Biosystems, Foster City, CA) at the Interdisci plinary Center for Biot echnology Research (ICBR) facility, University of Florida, Gainesville. Phylogenetic Analyses Sequences were edited using SequencherTM version 4.6 (Gene Codes Corporation, Ann Arbor, MI) and aligned using Clustal X version 2.0.6 (93). The alignment was adjusted by eye using Se-Al version 2.0a11 (Univers ity of Oxford, Oxford, UK). Gaps were considered missing data. Prior to phylogenetic analyses, sequence divergence was determined among isolates of F oxysporum f. sp. radicis lycopersici F oxysporum f. sp. lycopersici and both formae speciales according to p-distance implement ed in MEGA version 4 (166). Phylogenetic analyses were first performed on DNA sequences of EF-1 FOL185, and IGS as individuals. To test whet her the three data sets could be combined, incongruence length difference (ILD) was assessed (47). ILD implem ented as the partition homogeneity test in PAUP* 4.0b10 (162) was performed with 1000 data partitions using heur istic search with 1000 replications of random stepwise addition and branch swapping algorithm using tree bisection-

PAGE 70

70 reconnection (TBR). If the null hypoth esis of homogeneity is rejecte d, the data sets should not be combined without further justification (18). Parsimony analyses were conducted using PAUP*. Heuristic searches for most parsimonious trees were performed with random stepwise addition (1000 replications) and TBR. Clade stability was evaluated using 1000 bootst rap replicates with arrangements limited to 1,000,000 per replicate. For maximum likelihood (ML) and Bayesi an analysis, MODELTEST version 3.7 (124) was used to determine appropria te models of nucleotide subs titution. The best-fit model of sequence evolution was chosen based on the Akaike Information Criterion (70). ML was analyzed using GARLI version 0.96 (184) in which a stochastic genetic algorithm-like approach was used to simultaneously estimate the topology, branch lengths, and substitution model parameters that maximize the log-likelihood. A bootstrap analysis was performed using 1000 replicates. Bayesian analysis was performed usi ng MRBAYES version 3.1.2 (136). The Markov Chain Monte Carlo (MCMC) was run with f our chains for 10,000,000 generations, sampling every 100 generations and starting with a random tree. Stationarity was reached at approximately generation 30,000; thus the first 300 trees were the burn-in of the chain, and phylogenetic inferences were based on these trees sa mpled after generation 30,000. The remaining 99,700 trees were imported to PAUP* 4.0b10 to generate a 90% majority-rule consensus tree rooted with outgroups. Mating Type Determination The m ating type (MAT) of each isolate was de termined as previously described (84). Isolates from which a fragment size of 200 bp was amplified with fusALPHAfor and

PAGE 71

71 fusALPHArev were typed as MAT-1 Isolates were determined as MAT-2 if a fragment of 260 bp was amplified with fusHMGfor and fusHMGrev. Crossing Sexual crosses were carried out as previous ly described (87,88). For all m atings, each isolate was started from a single conidium Two isolates, CL-06191 (VCG 0099) and CL-06196 (VCG 0099), carrying MAT-2 acted as females and were cultured on carrot agar. Five isolates, CL0620 (VCG 0098), CL06122 (VCG 0098) CL-06124 (VCG 0094), CL-06125 (VCG 0094), and CL-06128 (VCG 0098), with MAT-1 acting as males (the spermatizing agent) were cultured on CLA. Matings were conducted 10 days after the cultures were initiated. A conidial suspension was made from the male parent using sterile water and spread evenly over the surface of the female culture. After mating, these plates were incubated at 25 C with a 12-h-photoperiod (photonflux of 40.8 mol/m2s). Observation for perithecia by mi crosacope was carried out every week until three months after mating. All male pare nts were crossed with the female isolates at least two times. Virulence Test Seedlings of the cultivar Bonny Best (suscep tible to Fusarium crown and root rot) and Fla. 7781 (resistant to Fusarium crown and root rot) (149) were grown in 17 25 6 cm trays with a commercial potting mix (Metro Mix 300, pH1:2 = 5.27 0.16). Seedlings at the cotyledon stage were gently uprooted, washed, and then replanted into 15-cm-diameter pots with 1.5 kg sand (University of Florida Turfgrass Research Envirotron) for approximately 2 weeks or until the first two true leaves had emerged. Each pot contained two tomato plants. Since soil pH affects the development of Fusarium crown a nd root rot (76), a tota l of 600 ml Hoaglands solution (64) adjusted to pH 5 was applied to each pot before inocula tion to favor infection.

PAGE 72

72 Eleven isolates of F oxysporum f. sp. radicis lycopersici that represented VCGs 0090, 0094 and 0098, 0099 and nonassigned VCGs includi ng a control of plants inoc ulated with sterile water were used for the virulence test (Fig. 3-4). Ten ml of a conidial and mycelial fragment suspension with 106 conidia mL-1, recovered from 14-day-old cultures grown on CLA, was placed on crowns using a pipette with sterile tip s. Ten seedlings were inoculated with each isolate. The inoculated seedlings were then placed in an incubator at 20C with a 12 h photoperiod (photonflux of 70.7 mol/m2s). Disease severity was assessed 35 days after inoculation. The inoculated plants were gently uprooted and washed with tap water. The length of discolored area in stem was measured, and it was divided by the total stem length to represen t the percentage of di sease severity. Disease severity in crowns was determined according to a 0-4 scale, where 0 = no symptom; 1 = 1-25 % discoloration; 2 = 26-50 %; 3 = 51-75%; and 4 = 76-100%. The percentage of infected root was also recorded. The inoculation experiment, repeated twice, was arranged in a completely randomized design with five replicates (pots) per isolate per cultivar and two plants per pot. Statistical analyses were performed with PROC GLM im plemented in SAS version 9.2 (SAS Institute, Cary, NC). The difference in virulence between is olates was compared using the least significant difference test (LSD) at p =0.05. The three disease severity ratings were used for cluster analysis according to Wards minimum variance method (175) implemented in JMP version 7 (SAS Institute, Cary, NC). Nucleotide Substitution Rate s and Population Dynamics To estim ate rates of nucleotide subs titution and population dynamics in F oxysporum f. sp. radicis lycopersici a Bayesian MCMC approach was performed using BEAST version 1.4.8

PAGE 73

73 (34). The method relied on the samp ling date for each isolate of F oxysporum f. sp. radicis lycopersici and explored evolutionary models wi th tree topology, substitution rate, and population size changes. Sequence data of F oxysporum f. sp. radicis lycopersici were realigned, and nucleotide substitution models for IGS, EF-1 and FOL185 were determined using MODELTEST. The uncorrelated relaxed lognorma l clock implemented in BEAST was used since variation in the nucleotide substitution rate among branches was detected by BEAST in these three data sets as a result of the standard deviation in br anch rates greater than the mean rate. The result suggested that these three data se ts exhibit very substantial rate heterogeneity among lineages, rejecting the st rict molecular clock (33). Bayesian MCMC was run for 1,000,000,000 generations with sampling every 100,000 generations. The first 1000 samples were discarded as burn-in. Default priors were used to analyze nucleo tide substitution rates ( ) and most recent common ancestor (MRCA). At least three independent runs were analyzed to corroborate these results. TRACER version 1.4.1 (University of Oxford, Oxford, UK) was used to visually examine posterior probabilities fo r Markov chain stationar ity and to summarize population parameters. Bayesian skyline plots with te n population groups of unique si zes were used to investigate historical changes in effective population size ( Ne) of F oxysporum f. sp. radicis lycopersici Priors for substitution rates with the 95% highprobability density (HPD) were derived from above analyses. All other parameters of these BEAST runs were the same as above. TRACER was used to generate Bayesian skyline plots.

PAGE 74

74 Results Phylogenetic Analysis Mean pairwise sequence divergences (uncor rected p-distances) were lower than 1% (ranging from 0.32-0.97%) among IGS, EF-1 and FOL185 for F oxysporum f. sp. radicis lycopersici and F oxysporum f. sp. lycopersici. IGS was the fastest evolving marker and FOL185 was the slowest (Table 3-1). Since the ILD test rejected th e null hypothesis of homogeneity, the three data sets were not combined but analyzed separately (Table 3-2) ( 18). The three loci provided different levels of phylogenetic resolution among isolates used. However, F oxysporum f. sp. radicis lycopersici and F oxysporum f. sp. lycopersici formed a strongly supported clade (96-100% bootstrap and 100% Bayesian posterior probability) in these three phylogenies (Figs. 3-1, 3-2, and 3-3). The IGS data set provided more resolution than either the EF-1 or FOL185 data set due to considerably more informativ e characters (Table 3-3).The IGS data set consisted of 2419 nucleotide characters, 314 of which were clad istically informative. In contrast, the EF-1 data set included 86 informative sites among 721 char acters, and the FOL185 data set contained 86 informative sites among 798 characters. Maximum parsimonious trees derived from these three data sets showed very low homoplasy as a resu lt of high consistency and retention indices (CI and RI). The IGS phylogeny revealed four clades (Fig. 31). Clade 1 consisted of three predominant VCGs, 0094, 0098, and 0099, of F oxysporum f. sp. radicis lycopersici in Florida, whereas in the EF-1 (Fig. 3-2) and FOL185 (Fig. 3-3) phylogenies these three VCGs were placed in a clade with VCGs 0031, 0033, and 0035 of F oxysporum f. sp. lycopersici Clade 2 contained VCGs 0030 and 0032 of F oxysporum f. sp. lycopersici and VCGs 0090, 0091, 0092, 0093, and

PAGE 75

75 0096 of F oxysporum f. sp. radicis lycopersici, which was mostly congruent with the EF-1 and FOL185 phylogenies except that VCG 0091 in the EF-1 phylogeny fell into an individual clade. A methodological incongruence was found in clade 2 of the IGS phylogeny as maximum likelihood analysis showed VCG 0093 as a distinct clade, contradicting with maximum parsimonious and Bayesian analyses. The IGS phylogeny showed that VCGs 0031 and 0035 were grouped in clade 3, and VCG 0033 was placed in clade 4. The stability of these two clades was strongly supported (bootstrap = 100%; Bayesian posterior pr obability = 100%), but these three VCGs were not well resolved in the EF-1 and the FOL185 phylogeny. The evolution of races in F oxysporum f. sp. lycopersici was not completely related to phylogenetic lineage. Clade 2 of the IGS phylogeny consisted of isolates of race 2 and BE1 which belongs to race 3 (105), whereas clades 3 and 4 seemed restricted to race 2 and 3, respectively (Table 2-2, Fig. 3-1). Mating-Type Analyses and Mating The mating-type analysis showed that each VCG carried a single unique MAT idiomorph (Fig. 3-1). Within the same clade of the IGS phylogeny, MAT-1 and MAT-2 were not found in F oxysporum f. sp. lycopersici but in F oxysporum f. sp. radicis lycopersici clade 1 and 2 both mating types were found. However, no perithecia were observed after crossing isolates carry distinct mating-type idiomorphs in clade 1of the IGS phylogeny. Virulence No isolate o f F oxysporum f. sp. radicis lycopersici was observed to infect the resistant cultivar, Fla. 7781. However, cluster analysis of virulence on the sus ceptible cultivar, Bonny Best, for VCGs of F oxysporum f. sp. radicis lycopersici showed that VCG 0098 had a higher

PAGE 76

76 virulence than the other two VCGs in Clade 1, and VCG 0090 in Clade 2 of the IGS phylogeny (Figs. 3-1 and 3-4). Nucleotide Substitution Rate s and Population Dynamics Mean nucleotide substitution rate varied am ong IGS, EF-1 and FOL185 data sets and ranged from 1.27 10-4 to 4.92 10-4, with a 95% credible interval of 2.74 10-5 9.80 10-4. The mean estimate of a most r ecent common ancestor (MRCA) of F oxysporum f. sp. radicis lycopersici ranged from 38 to 129 years ago (Table 34). However, the MRCA estimated from the IGS data set was older than from the other two data sets. These rate estimates with 95% credible interval s from each data set were used as priors of the relaxed molecular clock rate to generate Baye sian skyline plots. The increase in the scaled population size (equivalent to the effective population size ( Ne) multiplied by generation length ( g)) were revealed by IGS, EF-1 and FOL185 data sets, sugges ting a population expansion of F oxysporum f. sp. radicis lycopersici in the last 15 years (1992-2007) (Fig. 3-5). Discussion Although the phylogeny of F oxysporum f. sp. radicis lycop ersici and lycopersici have been previously studied ba sed on sequence data, phylogenetic resolution for these two formae speciales was not well resolved and the evolution of VCGs is not well known. This study used more characters than previous studies for phyl ogenetic analyses (19,171). One of the goals of this study was to evaluate polymorphisms of IGS, EF-1 and a noncoding microsatellite locus, FOL185 among VCGs of F oxysporum f. sp. radicis lycopersici for developing diagnostic markers. Of the three loci examined, EF-1 has been used most co mmonly for distinguishing Fusarium species (52). However, it did not provide good resolution for the two closely related

PAGE 77

77 formae speciales in this study. In contra st, the complete region of IGS was phylogenetically informative due to its fast-evolving nature (T ables 3-1 and 3-3). Moreover, the phylogenetic resolution derived from the complete region of IGS for these two formae speciales is better than previous studies using the partial region ( 19,83,171), indicating that se quencing the complete IGS region may be useful for studying phylogenies of formae speciales of F oxysporum According to this study, IGS may be used to de sign primers for rapid diagnosis of predominant VCGs of F oxysporum f. sp. radicis lycopersici FOL185 was evaluated for constructing the phylogeny of F oxysporum f. sp. radicis lycopersici and lycopersici as it revealed high gene diversity for these two formae speciales (Table 2-5). However, FOL185 showed the lowest average pairwise sequence divergence (Table 3-1), suggesting a phylogeneti cally uninformative marker. Based on sequence alignment, insertion and deletion were frequent in the micr osatellite motif, but transition and transversion rarely occurred in the flanking region of the microsatellite motif. This mutation type likely makes it more appropriate for stud ies of population genetics (39). The ILD test implemented in PAUP* suggest ed that the evoluti onary rate was not homogeneous among IGS, EF-1 and FOL185 data sets. Therefor e, combining data sets did not correctly improve the phylogeneti c estimation. Instead, these three data sets were subjected to separate phylogenetic analyses (18). Phylogenetic congruence can be defined as two identical trees obtained from different data sets or different methodol ogies (57). Instead, phylogenetic incongruence was shown in this study as topologies derived from IGS, EF-1 and FOL185 were different, suggesting unique evolut ionary histories of these thr ee loci. However, a species tree reveals most parts of the genetic history of a species and has a variance well represented by the diversity of trees derived from different genes (102). Therefore, the phylogenetic incongruence

PAGE 78

78 revealed in this study is not unc ommon. Other fast-evolving loci, if any, need to be developed for further elucidating the phylogeny and for designing diagnostic markers of these two formae speciales Fusarium oxysporum f. sp. radicis lycopersici and lycopersici have multiple evolutionary origins (polyphylectic) but share a common ancestor (paraphyletic ), whereas an incongruence was shown within topology of three phylogenies derived from IGS, EF-1 and FOL185. These three phylogenies revealed that F oxysporum f. sp. radicis lycopersici and F oxysporum f. sp. lycopersici formed a strongly supported clade, sugges ting that they evolved from the same common ancestor. At least two evol utionary origins were shown for F oxysporum f. sp. radicis lycopersici The evolutionary origin of VCGs 0094, 0098, and 0099 c ould be independent of other VCGs of F oxysporum f. sp. radicis lycopersici. VCG 0094 is cosmopolitan, and Florida has been suggested as its proba ble center of origin. Moreover, VCG 0094 in Europe might have migrated from Florida (139). P hylogenetic analyses showed that an isolate from Belgium was placed in a clade with Florida VCG 0094 isolates suggesting that Florida and European VCG 0094 are closely related even t hough geographically separated. Fu rther phylogenetic analyses involving more VCG 0094 isolat es worldwide may uncover its phylogeographical distribution and probable center of origin. VCGs 0092, 0093, and 0096 were only found in Israel (82). Like VCGs 0094, 0098, and 0099 in Florida, they were phyl ogenetically close and fell into clade 2 of the IGS phylogeny, probably suggesting sympatric speciation, which describes two or more descendant species inhabiting th e same geographic location and deriving from a single common ancestor species (161). VCGs 0090 and 0091 were widely distributed in North America and placed in clade 2 of the IGS phylogeny. However, they were not found in this study (Fig. 2-1)

PAGE 79

79 but replaced by VCGs 0094, 0098, and 0099 of cl ade 1, suggesting another independent evolutionary event of F oxysporum f. sp. radicis lycopersici No perithecia were observed after mating isolat es with different mating-type idiomorphs in clade 1of the IGS phylogeny. This finding suggests th at other factors may be required for sexual recombination since RT-PCR analysis re vealed that mating type genes of F oxysporum are still expressed and processed correctly (181). Virulence tests showed that VCG 0098 had a higher level of virulence than the other VCGs. VCG 0098 was recently found in Florida and phylogenetically close to VCG0094 according to sequence data and VCG assays, suggesting that the pathogen may have diverged toward higher virulence. However, Validov et al. (171) suggested that virulence is not associated with the phylogenetic clade of the IGS phylogeny. No VCG information was available and partial IGS was used in their study, as a consequence our data could not be directly compared with theirs. Further studies including all the VCGs in different clades will be necessary to better understand the evolution of virulence in F oxysporum f. sp. radicis lycopersici However, based on this study, VCG 0098 may be useful for br eeding against Fusari um crown and root rot of tomato due to its high virulence and evolutionary origin. Estimates of mean most recent common ancestor (MRCA) derived from IGS, EF-1 and FOL185 data sets ranged from 38 to 129year s ago (1969-1878), suggesting that rate heterogeneity may be across regions of the genome and that Fusarium oxysporum f. sp. radicis lycopersici is a relatively new pathogen compared to the invention of agri culture approximately 10,000 years ago (161). Based on the mean MRCA es timates, the pathogen might have diverged from its ancestor around the ninet eenth or twentieth centu ry. It has been suggested that tomatoes had undergone considerable domestication be fore being taken to Europe in the 15th century and a

PAGE 80

80 much more intensive level of domestication occu rred throughout Europe in the eighteenth and nineteenth century (9,155). Moreover, tomato only gained economic importance by the end of the nineteenth century or begi nning of the twentieth century (1 23). These results suggest that F oxysporum f. sp. radicis lycopersici might have coevolved with domestication of its host and has been selected by the environmental and genetic un iformity of the agricult ural ecosystem (161). IGS, EF-1 and FOL185 data sets reveal ed the population expansion in F oxysporum f. sp. radicis lycopersici in the last 15 years ( 1992-2007). Numerous resistan t cultivars have been developed since 1983 (150), probably causing a considerable decrease in the population size of F oxysporum f. sp. radicis lycopersici However, the population size ha s been increasing since the 1990s according to these three data sets (Fig. 3-5). It is not clear whether other factors, such as the phase-out of methyl bromide and the Frl resistance gene not introgressed into most commercial cultivars, may also have caused this increase in the populat ion size. Using regular crop rotations and avoiding extremely suscepti ble cultivars are suggested to minimize the pathogen population size (107,108) Regularly monitoring the population structure of F oxysporum f. sp. radicis lycopersici is also necessary for disease management.

PAGE 81

81 Table 3-1. Mean pairwise nucleotide sequence di fferences for each locus using p-distances (in percentages) among isolates of Fusarium oxysporum f. sp. radicis lycopersici (FORL), F oxysporum f. sp. lycopersici (FOL), and both formae speciales Locus name FORL FOL FORL+FOL IGS 0.76 0.97 0.88 EF-1 0.73 0.74 0.77 FOL185z 0.36 0.32 0.39 z A noncoding microsatellite locus obtained from contig number 185 of the genome sequence of F oxysporum f. sp. lycopersici at http:// www.broad.mit.edu. Table 3-2. Results of partition homogeneity test among three data sets of intergenic spacer (IGS), partial elongation f actor 1-alpha (EF-1 ) and a noncoding microsat ellite locus, FOL185 P values (1000 partitions)y Locus name IGS EF-1 IGS EF-1 0.001 FOL185 z 0.004 0.638 y Partition homogeneity test was performed using heuristic search with 1000 replications of random stepwise addition and branch swappi ng algorithm using tree bi section-reconnection (TBR) implemented in PAUP* 4.0b10 (162). z A noncoding microsatellite locus obtained from contig number 185 of the genome sequence of F oxysporum f. sp. lycopersici at http:// www.broad.mit.edu. Table 3-3. Data set properties and nucleotide subs titution models used in phylogenetic analyses Data set Substitution model Characters PIw sites Tree length CIx RIy IGS TrN + G 2419 314 796 0.895 0.929 EF-1 GTR + G 721 86 151 0.914 0.955 FOL185z TrN + G 798 86 187 0.904 0.922 w Number of potentially pars imonious informative sites. x Consistency index. y Retention index. z A noncoding microsatellite locus obtained from contig number 185 of the genome sequence of F oxysporum f. sp. lycopersici at http:// www.broad.mit.edu.

PAGE 82

82 Table 3-4. Bayesian Markov chain Monte Carlo (MCMC) estimates of evolutionary dynamics of Fusarium oxysporum f. sp. radicis lycopersici using the uncorrelated re laxed lognormal clock model implemented in BEAST Data set Model selected Characters Isolation date range Mean rate (subs/site/yr) HPDx of rate (subs/site/yr) Mean MRCAy (yr) HPD of MRCA (yr) IGS HKY+I+ 2335 1975-2007 1.68 10-4 2.84 10-5, 3.46 10-4 129 32, 310 EF-1 HKY 699 1975-2007 1.27 10-4 2.74 10-5, 2.57 10-4 81 32, 150 FOL185z GTR+I+ 752 1975-2007 4.92 10-4 8.44 10-5, 9.80 10-4 38 32, 56 x 95% high-probability density. y Most recent common ancestor. z A noncoding microsatellite locus obtained from contig number 185 of the genome sequence of F oxysporum f. sp. lycopersici at http:// www.broad.mit.edu.

PAGE 83

83 Figure 3-1. Phylogeny for Fusarium oxysporum f. sp. radicis lycopersici and F oxysporum f. sp. lycopersici inferred from the intergenic spacer (IGS) region of nuclear ribosomal DNA. Numbers on nodes represent bootstrap support values for maximum parsimony (front), maximum likelihood (middle), and Bayesian posterior probabilities presented as percentage (back). Values represented by an were less than 50% for bootstrap. Maximum likelihood analysis revealed VC G 0093 was an individual clade, which was the only incongruence compared to th e other two analyses. The mating type (MAT) of each isolate was determined as previously described (84).

PAGE 84

84 Figure 3-2. Phylogeny for Fusarium oxysporum f. sp. radicis lycopersici and F oxysporum f. sp. lycopersici inferred from partial elonga tion factor 1-alpha (EF-1 ). Numbers on nodes represent bootstrap support values fo r maximum parsimony (front), maximum likelihood (middle), and Bayesian posterior probabilities presented as percentage (back). Values represented by an were less than 90% of Bayesian posterior probabilities.

PAGE 85

85 Figure 3-3. Phylogeny for Fusarium oxysporum f. sp. radicis lycopersici and F oxysporum f. sp. lycopersici inferred from a noncoding micros atellite locus, FOL185. Numbers on nodes represent bootstrap support values fo r maximum parsimony (front), maximum likelihood (middle), and Bayesian posterior probabilities presented as percentage (back). Values represented by an were less than 50% for bootstrap.

PAGE 86

86 Figure 3-4. Cluster analysis of virulence for vegetative compatibility groups (VCGs) of Fusairum oxysporum f. sp. radicis lycopersici according to Wards minimum variance method (175) implemented in JMP. Th ree disease severity ratings were used including the percent of co rtical, crown, and root discolorations. represents nonassigned VCGs.

PAGE 87

87 Figure 3-5. Bayesian skyline plots of Fusarium oxysporum f. sp. radicis lycopersici derived from IGS (A), EF-1 (B), and FOL185 (C) data sets using BEAST version1.4.8. The xaxis is time as measured in years before present and the y-axis is the scaled population size ( which is equivalent to the effective population size ( Ne ) multiplied by generation length (g)). Each curve is a plot of the median, with its 95% credible interval estimates indicated by the blue lines.

PAGE 88

88 CHAPTER 4 EFFECT OF SILICON ON FU SAR IUM CROWN AND ROOT ROT OF TOMATO CAUSED BY Fusarium oxysporum f. sp. radicis lycopersici Introduction Fusarium oxysporum f. sp. radicis lycopersici the causal agent of Fusarium crown and root rot, is an im portant soil borne pathogen of tomato (72,80). This pathogen has been found in many tomato-growing regions of the world although it is a relatively newl y identified pathogen (80). No physiological races have been reported, but considerable genetic diversity is suggested by the existence of many vegetative compatibil ity groups (VCGs) (80,82). The pathogen can be introduced to new tomato-growing regions by means of infected seeds, transplants, soil and media (59,74,111). Once introduced, this polycyclic pathogen can be disseminated via root-toroot contact, dispersal of ai rborne conidia, water flow, and fungus gnats of the genus Bradisya (53,59,74, 133,140), making control of this disease mo re difficult. Soil fumigation has been investigated to manage this disease (62,106,109) but social and environmental concerns have caused the phaseout of certain chemicals such as methyl bromide. Further research in biological control needs to be conducted for application in the field (2 7). Thus, the use of resistance cultivars seems logical and effective for controlling the disease. However, resistance may be broken down due to the evolutionary potential of the pathogen and e nvironmental conditions (107,108). Moreover, the Frl resistance gene to F oxysporum f. sp. radicis lycopersici is only deployed in some newer cultivars (151). Altern ative and environment-friendly approaches for managing Fusarium crown and root rot of tomato need to be evaluated. Silicon (Si) has been suggested to moderate biotic and abiotic stresses on tomato (2,25,32, 122), although tomato has been defined as being a Si excluder (99). Si accumulates in the cytoplasmic fraction of Si-accumul ator plants but it is mainly de posited in the cell-wall fraction in roots of Si-excluder plants such as tomato (60). The reinforcement of root cell walls may

PAGE 89

89 affect the penetration of F oxysporum f. sp. radicis lycopersici because the pathogen penetrates the epidermis of tomato roots di rectly and then produ ces intracellular and intercellular hyphae in the outer parenchyma of cortical tissues beneath the penetrated sites (16). Resistant cultivars can form a defensive barrier in the parenchyma cells and prevent the pathogen from spreading toward the central vascular bundl e (16,178). Like the defensive barr ier induced in the resistant cultivar, Si may influence the creation of a physical barrier and also induce other defense responses in the host (20,26). Furt her research in the relationshi p between the penetration caused by this pathogen and the structure of cell wall st rengthened by Si needs to be conducted, whereas no effect of Si on Fusarium crown and root rot has been previously suggested (112). Si-mediated resistance in tomato may not be lo cated in the roots. For example, Si-mediated resistance in tomato against Ralstonia solanacearum was likely located in stems due to changes in the pectic polysaccharide structure of stem cell walls, restricting the bacterial movement to the stems (32). However, Si significantly decreased the bacterial population in roots and stems of a resistant cultivar, Hawaii 7998, compared to non-tr eated plants. These resu lts suggested that Simediated resistance may also exis t in tomato roots. It is not well known whether Si content of roots correlates with disease severity of Fusarium crown and r oot rot of tomato. The objectives of this study were (i) to investigate effects of Si and inoculum concentrations on disease severity of Fusarium crown and root rot (ii) to eval uate the effect of Si on disease progress (iii) to determine whether there was an association of disease severity with Si content of roots and shoots. Materials and Methods Plant Growth and Silicon Amendment Seeds of Bonny Best, susceptible to Fusarium crown and root rot, were soaked in 5% bleach for 2 m in for surface steriliz ation and then washed several times with sterile water. The

PAGE 90

90 sterilized seeds were sowed in a co mmercial growth medium (Metro Mix 300, Palmetto, FL). One week after sowing, seedlings at the cotyledo n stage were transplanted to 15-cm-diameter plastic pots (Hummert Internatinal, Earth City, Missouri) filled with 1.5 kg sand provided by the University of Florida Turfgrass Research Envi rotron. Each pot included two tomato plants. The seedlings were grown in a growth chamber held at 20 C (12 h light photoperiod with photonflux of 70.7 mol/m2s). Hoaglands nutrient solution ( 64) with (+ Si) or without ( Si) 100 mg Si L-1 (3.56 mM) as sodium meta silicate nonahydrate (Na2SiO39H2O) (42) was adjusted to pH 5 using 36 N sulfuric acid (H2SO4) before applying to tomato plants since lowering the pH of fine sand increases disease severity of Fusarium crown a nd root rot (76). The nutri ent solution contained N 224 mg, P 62 mg, K 235 mg, S 32mg, B 0.5 mg, Mn 0.5, Zn 0.05, Cu 0.02, Mo 0.01 mg, and Fe 1.56 mg L-1and was prepared using deionized water. Ea ch pot was fertigated with a total of 600 ml of the nutrient solutio n with or without Si during 3 weeks after transplanting. Fifty mL of the nutrient solution was given to each pot every other day for the first 18 days after transplantation. After which, 50 mL of the solution was applied to each pot for 3 consecutive days. Deionized water was used to irrigate tomato plants as needed. Inoculum Production and Inoculation Procedure Isolate CL-0601, belonging to VCG 0098, of F oxysporum f. sp. radicis lycopers ici was used for inoculation due to its high virulence re vealed by earlier studies (Chapter 3). Depending on experiments, 10 ml of a conidial a nd mycelial fragment suspension with 105 and/or 106 conidia mL-1, recovered from 14-day-old cultures grown on carnation leaf agar (CLA), was placed on crowns using a pipette with sterile tip s 3 weeks after transplanting. The inoculated plants were placed in a comple tely randomized design in an in cubator at 20C with 12 h light photoperiod (photonflux of 70.7 mol/m2s).

PAGE 91

91 Disease Assessments Three dis ease severity ratings were made on roots, crowns, and stems. Root infection was rated visually as the percentage of roots show ing discoloration. Brown discoloration in crowns was determined according to a 0-4 scale: 0 = no symptom; 1 = 1-25 %; 2 = 26-50 %; 3 = 51-75%; and 4 = 76-100%. Disease severity in stems was defined as the ra tio of the lesion length divided by the stem length. The area under disease progress curve (AUDPC) was calculated using the method of Shaner and Finney (154) for studying th e effect of silicon on disease progress over time. Diseased plants were sampled to confirm the presence of the causal agent, isolate CL-0601, identified using vegetative compatibility grouping (24). Effects of Silicon and Inoculum Concentration A factorial design of six treatm ents with five replicates (10 plants ) was arranged using a completely randomized design in the incubator: (i ) plants with silicon, inoculated with sterile water (+Si FORL), (ii) plants with silicon, inoculated with F oxysporum f. sp. radicis lycopersici at a concentration of 106 conidia/plant (+Si+FORL1), (iii) plants with silicon, inoculated with F oxysporum f. sp. radicis lycopersici at a concentration of 107 conidia/plant (+Si+FORL2), (iv) plants without sili con, inoculated with sterile water ( Si FORL), (v) plants without silicon, inoculated with F oxysporum f. sp. radicis lycopersici at a concentration of 106 conidia/plant ( Si+FORL1), and (vi) plants wi thout silicon, inoculated with F oxysporum f. sp. radicis lycopersici at a concentration of 107 conidia/plant ( Si+FORL2). Four weeks after inoculation, all plants were harv ested, washed and rated for disease severity and then divided into shoots and roots for silicon quantification.

PAGE 92

92 Effect of Silicon on Disease Progress over Time To f urther determine and confirm the effect of silicon on Fusarium crown and root rot of tomato, disease progress over time was determined using partial treatments of the above factorial design: +Si FORL, +Si+FORL1, Si FORL, and Si+FORL1. Ten plants of each treatment were evaluated for disease severity and divided into shoots and roots for silicon concentration analysis 2, 3, 4, and 6 w eeks after inoculation. Determination of Dry Root and Shoot Weight and Silicon Quantification After rating for disease severity, plants were divided into shoots and roots, washed in deionized water, and dried separate ly in paper bags for 72 h at 80 C (Isotem p Oven, Fisher Scientific). Dry roots and shoots were we ighed, ground using a Cyclotec 1093 sample mill (FOSS, Denmark), passed through a 40-mesh scree n, and stored in 20 ml plastic scintillation vials (Fisher Scientific, Pittsbu rgh, PA). Si analysis was based on the method of Elliott and Snyder (40) using a colorimetric analysis with a modification of the digestion procedure of plant tissues. One hundred mg plant tissues were used for digestion in a 100 ml plastic high-speed polypropylene copolymer tube (Nal gene) using 2 ml of 30% H2O2 and 3 ml of 100% NaOH. The tube was then placed in a 100 C water bath for one hour to initiate the tissue digestion before autoclaving for 20 min. If necessa ry, additional amount of 30% H2O2 was added and the autoclave cycle was repeated until complete digestion. Statistical Analysis All data collected were subjected to analysis of a factorial experim ent in SAS v. 9.2 (SAS Institute, Cary, NC) using PROC GLM to evaluate effects of Si and inoculum concentration and their interaction. Standard analysis of varian ce (ANOVA) was also performed. Treatment mean comparisons were analyzed using Fishers Protec ted Least Significant Difference test (FLSD) at

PAGE 93

93 P 0.05. Regression analysis was performed to determine the relationship between silicon content and disease severity using SIGMAPLOT version 10.0 (S ystat Software, Chicago, IL). Results Effects of Silicon and Inoculum Concentration The application of Si signifi cantly increased the concentratio n and uptake of Si in shoots and roots at the tim e of inoculation (3 weeks af ter transplanting) and at the time of harvest (4 weeks after inoculation) (Table 4-1). Si supply also si gnificantly enhanced dr y weight of roots by 20.8%, whereas no significant difference was shown fo r dry weight of shoot and total dry weight per plant between +Si and Si treatments (Tables 4-1 and 4-2). The effect of inoculum concentration significantly affect ed dry weight, silicon uptake of roots and shoots, and Si concentration in shoots except the Si content in roots. However, no significant interaction was revealed between Si and inoculum effects for these plant components estimated except the Si uptake of the shoot (Table 4-2). The application of Si significantly decreased disease severity of the stems 4 weeks after inoculation, although disease seve rity of the root and crown was not affected by Si supply (Tables 4-3 and 4-4). No significant interac tion between Si application and inoculum concentration was detected, suggesting the respon se to inoculum concentration was consistent between +Si and Si treatments. Four weeks after inoculation, dis ease severity of both the root and crown was not significantly different be tween two the inoculum concentrations, 106 and 107 conidia plant-1, but there was significantly larger stem lesions with the higher concentration (Table 4-5).

PAGE 94

94 Effect of Silicon on Disease Progress Sim ilar Si effects were observed as in the above experiment. The application of Si significantly increased dry weight of roots, and the concentration and uptake of Si in shoots and roots, whereas the +Si treatment did not significan tly affect dry shoot weight (Figs. 4-1 and 4-2). Root discoloration appeared earlier for the Si treatment than for the +Si treatment. Moreover, the +Si treatment significantly decreased root in fection by 82.1% 3 weeks after inoculation (Fig. 4-3). No significant difference was detected between Si and +Si treatments for disease severity of roots 4 and 6 weeks after inoc ulation. Crown discolor ation was also shown to occur earlier in tomato plants without Si. Three and 4 weeks afte r inoculation, disease seve rity of the crown in tomato plants amended with Si was significantly lower than those without Si. Like the other two rating systems, stem discoloration was first obs erved in tomato plants without Si. The +Si treatment significantly decreased disease severity of the stems 4 weeks after inoculation but no significant difference was shown between Si and +Si treatments 6 w eeks after inoculation (Fig. 4-3). Si supply significantly lowered the AUDPC for disease severity of stem by 52.5%, whereas the AUDPC for disease se verity of root and crown wa s not significantly different between Si and +Si treatments (Table 4-6). Regression analysis was used to further investig ate the effect of silic on on disease severity of Fusarium crown and root rot development. The results showed that a linear model best described the relationship between silicon content of roots and disease se verity of root, crown, and stem (Fig. 4-4). Regression analysis also suggested that disease severity decreased consistently with increasing sili con content of roots. Interesti ngly, the linear model also best described the relationship between silicon conten t of shoots and diseas e severity of stem,

PAGE 95

95 showing that disease severity of stem decreased consistently with increas ing silicon content of shoots (Fig. 4-5). Discussion One of the goals of this study was to determ ine the effect of Si on Fusarium crown and root rot of tom ato through studying disease progre ss and effects of inoculum concentration, since no effect of Si was previously suggested for th is disease (112). This study also showed no significant difference between Si and +Si treatments for root in fection 4 weeks after inoculation. However, Si was able to significantly reduce the disease severity of stems 4 weeks after inoculation, whereas a significant decrease in the diseas e severity of crowns by Si supply was merely observed in the experiment on disease progr ess (Table 4-4 and Fig. 4-3). Possible causes in reducing disease sever ity by Si are discussed. Inoculum concentration significan tly affected the disease severity of roots, crowns, and stems, whereas no significant interaction was rev ealed between Si and inoculum levels (Table 43), suggesting that the response to inoculum concentrations wa s consistent between +Si and Si treatments in this study. In other words, the e ffect of Si was not signi ficantly impacted by the inoculum level 4 weeks after inoculation. When e ffects of Si and inoculum concentration were separated to analyze their impact on disease severity, the +Si tr eatment significantly reduced the disease severity of the stem but not the root or crown. This resu lt suggested that disease progress already had reached the stem 4 w eeks after inoculation and that th e pathogen had proliferated in roots and crown overwhelmingly (Table 4-4). Levels of inoculum concentration also significantly affected disease progress. The disease se verity of the stem in plants inoculated with 107 conidia plant-1 was significantly higher than in those inoculated with 106 conidia plant-1, whereas no significant difference was shown for the disease severity of either roots or crowns

PAGE 96

96 between these two inoculum concentrations (Table 4-5). These results su ggested that the higher the inoculum concentration, the faster the dise ase progress of Fusarium crown and root rot. Inoculum concentration of 106 conidia plant-1 was selected for studyi ng disease progress to corroborate the effect of Si on Fu sarium crown and root rot of to mato since the inoculum density in the root zone would be able to cause disease (134). Furthe r analysis of disease progress suggested that the decrease in di sease severity by the +Si treatm ent was probably due to delaying initial infection in roots and the movement of the pathogen fro m roots to stems (Fig. 4-3). Tomato plants treated with Si di d not show visible discoloration in roots, crowns, and stems but those without Si supply revealed some brown lesions 2 weeks after inoculation.The disease severity of roots showed a significant difference between Si and +Si treatments 3 weeks after inoculation as a result of signifi cant variance in Si content of r oots (Figs. 4-2 and 4-3). These findings suggested that soilborne diseases may be reduced using Si fertilizers for Si nonaccumulator plants with a limited ca pacity to accumulate Si in root s and inefficient translocation of this element to shoots. The reduction in disease progress of Fusarium crown and root rot was positively correlated with an increase in the Si c ontent of tomato roots (Fig. 44), suggesting a silicon-induced resistance and/or reduction of fungal colonizat ion. Formation of a physical barrier has been proposed to explain Si-mediated resistance ( 20,26,179) because Si can accumulate and deposit beneath the cuticle to form a cu ticle-Si double layer and thus preven t leaves from penetration of pathogens (26,144). Although physical barriers in r oots may not be associated with siliconinduced resistance in tomato to R solanacearum silicon-induced changes in the pectic polysaccharide structure of tomato stem cell wall s have been observed (32). Moreover, siliconinduced resistance in tomato to R solanacearum may be associated with the capability of the

PAGE 97

97 plant to restrict the bacterial movement to th e stems due to a significant correlation between the bacterial population in stem and the resistance (3 2). Our study also showed that the +Si treatment likely limited the basipetal spread of F oxysporum f. sp. radicis lycopersici from infected roots (Fig. 4-3), whereas the reinforcement in cell wa lls of tomato roots wa s not evaluated. Also, a significant relationship between si licon content of shoots and disease severity of stems was exhibited (Fig. 4-5), suggesting th at Si concentration of the stem may be associated with the movement of the pathogen in st ems. Due to Si accumulation mainly in the cell-wall fraction of tomato roots (60) and unclear mechanisms of Si-m ediated resistance in tomato, it is pertinent to further investigate physical barrier, biochemical and molecular mechanisms involved in siliconinduced resistance (20,26) to F oxysporum f. sp. radicis lycopersici Si concentrations in roots were higher than those in shoots of tomato (Figs. 4-1 and 4-2) and this is typical for a Si nonaccumulator plant in agreement w ith previous studies (25,60). Si amendment significantly increased Si contents in both roots and shoots, even though tomato is categorized as a Si non-accumulator (99). Infection by F oxysporum f. sp. radicis lycopersici did not increase the Si accumulation in tomato, whereas cucumber plants accumulated more Si around penetration sites (21). Tomato plants were pretreated with Si, but Si application was discontinued after inoculation in th is study, resulting in a rapid decr ease in Si contents of roots and shoots during incubation (Figs. 4-1 and 4-2). A rapid decline of Si-induced resistance to Sphaerotheca fuliginea was observed after transferring cucumber plants treated with Si to Si-free solution (144), and this was also found in the Pythium aphanidermatum /bitter gourd system (61). The availability of soluble silicic acid, but not the total Si concentration in roots, at the time of infection has been suggested as the main contributor to Si-media ted resistance (22). Although Si accumulates in the cell-wall fraction of roots in tomato (60), it is not clear whether the continual

PAGE 98

98 application of Si after inoculation may increase the availabi lity of soluble silicic acid and enhance Si-mediated resistance to Fusarium crown and root rot. Silicon is not defined as an essential nutrient for plants, but it has been suggested as an essential element involved in the physiology of tomato growth through phytohormone synthesis (100). To maintain normal plant biology, 1 mM Na2SiO3 is recommended to be included in a modified Hoagland solution (42). This study utili zed 3.56 mM Si in Hoaglands nutrient solution, suggesting that enough Si was utilized to s upport normal tomato growth. Si significantly increased dry weight of roots, whereas no signi ficant difference was shown for dry weight of shoots between Si and +Si treatments. In addition to Si-m ediated resistance, dry weight of roots increased by Si supply, suggesti ng that Si application may also benefit tomato plants by maintaining normal physiology (43). Plant roots release exudates in the rhizosphere which microorganisms can use as a primary food source. This influences the population density and activity of these microorganisms and the outcome of the pathogen infecti on as a result of the interaction between pathogens a nd other microorganisms (130). Sin ce Si has been reported to mitigate biotic and abiotic stresses on tomato (2,25,32,122), how to apply Si fertilizers for fieldgrown tomatoes is worth of being further investigated.

PAGE 99

99 Table 4-1. Effect of silicon (Si) on Si concentration, dr y weight, and Si uptake of tomato inoculated with Fusarium oxysporum f. sp. radicis lycopersici at the time of inoculation (week 0) and the time of harvest (week 4) Week 0 Shoot Root Treatments Dry weight (g plant-1) Si concentration (mg (g dw)-1) Si uptake (mg)Dry weight (g plant-1) Si concentration (mg (g dw)-1) Si uptake (mg) Total dry weight (g) Without Si 0.30 0.31 0.09 0.03 1.51 0.06 0.34 With Si 0.39 0.95 0.36 0.06 3.31 0.21 0.45 FLSD ( P 0.05) 0.16 0.25 0.11 0.02 0.43 0.09 0.17 Week 4 Shoot Root Treatments Dry weight (g plant-1) Si concentration (mg (g dw)-1) Si uptake (mg)Dry weight (g plant-1) Si concentration (mg (g dw)-1) Si uptake (mg) Total dry weight (g) Without Si 1.62 0.45 0.73 0.24 0.77 0.18 1.86 With Si 1.65 0.92 1.53 0.29 1.55 0.44 1.94 FLSD ( P 0.05) 0.14 0.11 0.17 0.04 0.13 0.06 0.17

PAGE 100

100 Table 4-2. Analysis of variance for effects of silicon (Si) supply and inoculum concentration (IC) on plant components F valuesy Shoot Root Source of variation dfz Dry weight (g plant-1) Si concentration (mg (g dw)-1) Si uptake (mg) Dry weight (g plant-1) Si concentration (mg (g dw)-1) Si uptake (mg) Total dry weight (g) Si 1 0.15 ns 74.3 *** 98.9 *** 5.69 156 *** 78.1 *** 0.83ns IC 2 18.5 *** 5.49 26.7 *** 24.9 *** 1.37ns 15.5 *** 23.7 *** Si IC 2 0.84 ns 2.05ns 6.75 ** 0.48ns 0.89ns 2.14ns 0.31ns y Levels of probability: ns = not significant, 0.05, ** 0.01, and *** 0.001. z df = degrees of freedom.

PAGE 101

101 Table 4-3. Analysis of variance of effects of si licon (Si) supply and i noculum concentration (IC) on disease components Source of variation dfz F valuesx Disease severityy Root Crown Stem Si 1 0.16ns 0.01ns 5.15 IC 2 55.6 *** 146 *** 215 *** Si IC 2 0.10ns 0.01ns 2.26 ns x Levels of probability: ns = not significant, 0.05, ** 0.01, and *** 0.001. y Disease severity was estimated 4 weeks after inoculat ion. Root infection was rated visually as the percentage of root system showing discoloration; disease se verity of crown was evaluated using a 0-4 scale where 0 represents health and 4 means 100% discoloration; Stem discoloration was defined as the ratio of the lesion length in stem divi ded by the stem length. z df = degrees of freedom. Table 4-4. Disease severity of tomato plants amended with or without sodium metasilicate (Na2SiO3) and inoculated with Fusarium oxysporum f. sp. radicis lycopersici 4 weeks after inoculation Disease severityz Treatments Root Crown Stem Without Si 44.7 1.80 27.7 With Si 42.3 1.80 23.7 FLSD ( P 0.05) 12.1 0.31 3.63 z Root infection was rated visual ly as the percentage of root system showing discoloration; disease severity of crown was evaluated usi ng a 0-4 scale where 0 represents health and 4 means 100% discoloration; Stem discoloration was defined as the ra tio of the lesion length in stem divided by the stem length. Table 4-5. Effect of inoculum concentration on dis ease severity of tomato plants inoculated with Fusarium oxysporum f. sp. radicis lycopersici 4 weeks after inoculation Disease severityz Inoculum concentration (conidia/plant) Root Crown Stem 0 0 0 0 106 62.0 2.6 35.9 107 68.5 2.8 41.2 FLSD ( P 0.05) 14.8 0.38 4.46 z Root infection was rated visual ly as the percentage of root system showing discoloration; disease severity of crown was evaluated usi ng a 0-4 scale where 0 represents health and 4 means 100% discoloration; Stem discoloration was defined as the ratio of the lesion length in stem divided by the stem length.

PAGE 102

102 Table 4-6. Area under disease prog ress curve (AUDPC) of tomato plants amended with or without sodium metasilicate (Na2SiO3) and inoculated with Fusarium oxysporum f. sp. radicis lycopersici AUDPCz Treatments Root Crown Stem Without Si 66.2 8.76 157 With Si 39.2 5.10 74.6 FLSD ( P 0.05) 49.1 4.47 78.3 z Root infection was rated visual ly as the percentage of root system showing discoloration; disease severity of crown was evaluated usi ng a 0-4 scale where 0 represents health and 4 means 100% discoloration; Stem discoloration was defined as the ra tion of the lesion length in stem divided by the stem length. An inoculum concentration at 106 conidia/plant was used.

PAGE 103

103 Figure 4-1. Effect of silicon (Si) on Si concentra tion (A), dry weight (B), and Si uptake (C) of the tomato shoot. Bars with the same letter at each time period do not differ significantly at P 0.05 as determined by Fishers protected least significant difference test.

PAGE 104

104 Figure 4-2. Effect of silicon (Si) on Si concentra tion (A), dry weight (B), and Si uptake (C) of the tomato root. Bars with the same letter at each time period do not differ significantly at P 0.05 as determined by Fishers protected least significant difference test.

PAGE 105

105 Figure 4-3. Effect of silicon on symptom developmen t of Fusarium crown and root rot expressed as root infection (A), disease severity of crown (B) and of stem (C) on tomato cultivar Bonny Best over 6 weeks after inoculation. A 0-4 index was used according to the infection percentage of crown. Disease sever ity of stem is defined as the lesion length in stem divided by the stem length. Bars w ith the same letter at each time period do not differ significantly at P 0.05 as determined by Fishers protected least significant difference test.

PAGE 106

106 Figure 4-4. Relationship between root infection (A), disease severi ty of crown (B) and stem (C) and silicon concentration of tomato root 4 weeks after inoculation with Fusarium oxysporum f. sp. radicis lycopersici

PAGE 107

107 Figure 4-5. Relationship between di sease severity of stem and silicon concentration of tomato shoot 4 weeks after inoculation with Fusarium oxysporum f. sp. radicis lycopersici

PAGE 108

108 CHAPTER 5 CONCLUSION Analyses of m icrosatellites and vegetative compatibility groups (VCGs) revealed the current population structure of F oxysporum f. sp. radicis lycopersici in Florida which is the probable center of origin for the cosmopo litan VCG 0094 occurred along with two newly reported VCGs 0098 and 0099 (139). A bioinformatic s method and a microsatellite enrichment approach used in this study su ccessfully developed primers for 27 microsatellite loci amplifying both F oxysporum f. sp. radicis lycopersici and lycopersici, suggesting that microsatellites may be derived from searching the published genome se quence of closely related species instead of de novo isolation from the species studied. VCG 0094 is still predominant in Florida, but its frequency has decreased since the 1990s (139). In contrast, the distribu tion of VCG 0098 has increased, sugge sting an increased selective fitness. Moreover, VCG 0098 may have migrated from Collier County or other tomato-growing regions to Manatee and Hendry C ounties since a previous study did not find this VCG in these two counties (139). Two bridge is olates, vegetatively compatible with testers of both VCGs 0098 and 0094 I, were found in Manatee County, probably resulting from a process of either convergence or divergence (80). The frequency of VCG 0098 has increased and this VCG has a lower genetic diversity than VCG 0094. Moreover, VCG 0098 is phylogenetically related to VCG 0094 (Fig. 3-1). These findings suggest th at VCG 0098 may have diverged from VCG 0094. Migration and sympatric sp eciation are two possible caus es to explain how VCG 0098 appeared in Manatee County. Although 38 isolates could not be assigned to a known VCG, most of them showed the same microsatellite haplotype as either VCG 0094 or 0098, s uggesting that these nonassigned isolates might not comprise many new VCGs. Mutations in any of the genes controlling

PAGE 109

109 incompatibility may change the VCG of an isolate (95). Therefore, nonassi gned isolates with the same microsatellite haplotype as either of VCG 0094 or 0098 may be genetically related to either these two VCGs but not compatible with test ers of VCGs 0094 and 0098. However, this study cannot totally rule out another possibility; that is, increasing the frequency of unclassified isolates may also suggest the emergence of a new VCG within F oxysporum f. sp. radicis lycopersici Thus, regularly screening VC Gs and virulence of the pathogen seems necessary to monitor its genetic variation. Fusarium oxysporum f. sp. radicis lycopersici specific primers need to be further developed as the primers targeting exo polygalacturonase ( pgx4 63) failed to amplify all known VCGs of the pathogen due to considerable geneti c diversity within this pathogen. Florida VCGs 0094, 0098, and 0099 were not identified by the specifi c primers, suggesting that a comparison of their nucleotide diversity of cell-wall-degr ading enzymes (CWDEs) may not be a reliable approach to discriminate formae speciales of F oxysporum Instead, virulence genes conferring a specific trait to a pathogen are more likely to distinguish closely related formae speciales since they may have subtle nucleotide differences within a forma specialis but show obvious dissimilarity among formae speciales (98). Before primers with a higher specificity for F oxysporum f. sp. radicis lycopersici are developed, both pathogenicity and VCG tests should be conducted for identifying the pathogen. The labo ratory pathogenicity bioassay as previously described (71,145) may be appropriate to diagnose the pathogen according to symptomology, whereas inoculum preparation is critical for the success of seed ling inoculation (80). Microsatellites developed in th is study have revealed the current population structure of F oxysporum f. sp. radicis lycopersici in Florida. Hendry County had a higher mean number of pairwise differences and average gene diversity when either whol e microsatellite data or clone-

PAGE 110

110 corrected data were used for analysis. This may be because Hendry County was the sink of migrants from Manatee and Collier Counties based on the analysis of historical migration (Figs. 2-1 and 2-4). Considering the genetic structur e of each VCG and its affiliated geographical region, pairwise Fst values of VCG 0094 or 0098 between these three tomato-growing regions suggested no genetic differentiation (Table 2-6 and Table 2-7). However, the mean number of pairwise differences and average gene diversity were highe r for either VCG 0094 or 0098 in Collier County, suggesting that these two VCGs mi ght have migrated from Collier County to the other two counties. The gene diversity of VCG 0094 was higher in Collier County contradicting an earlier study by Rosewich et al. (139) showing a higher gene diversity in Hendry County. This difference may be caused by sample size bias (180 ) changes in population size (129), and other demographic events acting as selection (28,164). Our study used a more equal sample size of VCG 0094 from each tomato-growing region than the previous work (139) and suggested an increasing population size in Collier County, corr oborating a higher gene di versity observed in this county. However, whether ot her selection factors affect th e gene diversity of VCG 0094 in Florida needs to be investigated further. VC G 0099 had a higher mean number of pairwise differences and average gene diversity than either VCG 0094 or 0098, suggesting its preexistence in Florida (139). Migration of F oxysporum f. sp. radicis lycopersici among tomato-growing regions counties was exhibited by population admixture and historical migration analyses (Figs. 2-1 and 2-4). High gene/genotype flow can enhance genetic diversity of a pathoge n population as a result of increasing the size of geneti c neighborhood, resulting in more alleles to overcome a resistance gene in the host (107,108). Moreover, F oxysporum f. sp. radicis lycopersici has been considered as a soilborne, airborne, a nd waterborne pathogen (59,74,133,140). Therefore,

PAGE 111

111 limiting gene/genotype flow of F oxysporum f. sp. radicis lycopersici within Florida and migration of the pathogen to other tomato-growi ng regions is vital to re duce the break-down risk of the resistance gene Frl Of IGS, EF-1 and the noncoding microsatellite locu s, FOL185, examined in this study, the complete region of IGS was phylogenetically informative for F oxysporum f. sp. radicis lycopersici and lycopersici due to its fast-evolving nature (T ables 3-1 and 3-3). IGS may be used to design primers for rapid diagnosis of predominant VCGs 0094, 0098, and 0099 of F oxysporum f. sp. radicis lycopersici since these VCGs formed a supported clade (Fig. 3-1). Although the evolutionary rate of IGS, EF-1 and FOL185 was not homogeneous, phylogenetic analyses showed that F oxysporum f. sp. radicis lycopersici and lycopersici have multiple evolutionary origins (polyphylectic ) but share a common ancestor (paraphyletic). At least two evolutionary origins were shown for F oxysporum f. sp. radicis lycopersici in agreement with a dendrogram generated from 27 micros atellite data (Fig. 2-3). The evolutionary origin of VCGs 0094, 0098, and 0099 could be independent of other VCGs of F oxysporum f. sp. radicis lycopersici Mating isolates with distinct mating-type idio morphs from Collier C ounty in clade 1 of the IGS phylogeny was not successful since no perithecia were observed to form, whereas up to 62.3% of pairwise loci for isolates of this county were at linkage equilibrium. Even for a mitosporic fungus, parasexual recombination and reassortment resulting from protoplast fusion can cause chromosome rearrangement (170), resulti ng in alleles at one locus to be randomly associated. Other factors may be required for sexual recombination because RT-PCR analysis revealed that mating type genes of F oxysporum are still exp ressed and processed correctly (181).

PAGE 112

112 Restricting the migration of VCG 0098 to other tomato-growing regions in the United States is critical in disease management of F oxysporum f. sp. radicis lycopersici since this VCG was recently found in Florida (139) and show ed a higher level of virulence than the other VCGs used in this study. Moreover, VCG 0098 is phylogenetically closer to VCG0094 according to sequence data and VCG assays, sugg esting that the pathogen might have diverged toward higher virulence. Ba sed on this study, VCG 0098 may be used in breeding against Fusarium crown and root rot of tomato due to its high virulence and evolutionary origin. Fusarium oxysporum f. sp. radicis lycopersici is a relatively new pathogen compared to the invention of agriculture approximatel y 10,000 years ago (161). The mean most recent common ancestor (MRCA) of this pathogen ranges from 38 to 129 years ago when estimated from IGS, EF-1 and FOL185 data sets. Based on the mean MRCA derived from these three loci, the pathogen might have diverg ed from its ancestor around the 19th or 20th century. Interestingly, tomatoes acquired a much more intensive level of domestication that occurred throughout Europe in the 18th and 19th century (9,155), and were widely cultivated by the end of the 19th century or beginning of the 20th century (123). These findings suggest that F oxysporum f. sp. radicis lycopersici might have coevolved with domestica tion of its host and been selected by the environmental and geneti c uniformity of the agricultural ecosystem (161). The global population size of F oxysporum f. sp. radicis lycopersici has been increasing since the 1990s as revealed by the IGS, EF-1 and FOL185 data sets. Factors th at caused this increase in the population size need to be furthe r investigated. Regularly monito ring the population structure of F oxysporum f. sp. radicis lycopersici is essential for disease management since a plant pathogen with a higher populati on size has greater evolutiona ry potential to break down resistance (107,108).

PAGE 113

113 Silicon (Si) significantly reduced disease seve rity of stems 4 weeks after inoculation (Tables 4-4 and 4-6), whereas no signif icant difference was revealed between Si and +Si treatments in disease severity of crown and root. Inoculum con centration significantly affected disease severity of the root, crown, and stem, but no significant interaction was revealed between Si and inoculum concentration effects (Table 4-3) suggesting the response of Si consistent over inoculum levels. The analysis of disease progress suggested that the decrease in disease severity by Si amendment probably resulted from delaying in itial infection in roots and the movement of the pathogen from roots to stems (Fig. 4-3). Si contents of roots and shoots were significantly higher in tomato plants supplied with Si than those without Si amendment. Moreover, the increase in the Si content of roots was significantly correlated with the reduction of disease severity of roots, crowns, and stems (Fig. 4-4), indicating a Si-indu ced resistance and/or reduction of fungal colonization. The +Si treatmen ts likely limited the basipetal spread of F oxysporum f. sp. radicis lycopersici from infected roots to st ems (Figs. 4-3, 4-4, and 4-5). Although laboratory experiments incl uding this study have shown Si ab le to alleviate biotic and abiotic stresses on tomato (2,25,32,122), further re search in applying Si fertilizers for fieldgrown tomatoes needs to be conducted to further elucidate the effects of Si on the development of Fusarium crown and root rot.

PAGE 114

114 REFERENCES 1. Agapow, P.-M., and Burt, A. 2001. Indices of multilocus linkage disequilibrium. Mol. Ecol. Notes 1:101-102. 2. Al-Aghabary, K., Zhu, Z., and Shi, Q. H. 2004. Influence of silicon supply on chlorophyll content, chlorophyll fluorescence, and antioxi dative enzyme ac tivities in tomato plants under salt stress. J. Plant Nutr. 27:2101-2115. 3. Aoki, T., O'Donnell, K., Homma, Y., and Lattanzi, A. R. 2003. Sudden death syndrome of soybean is caused by two morphologically an d phylogenetically distin ct species within the Fusarium solani species complex : F virguliforme in North America and F tucumaniae in South America. Mycologia 95:660-684. 4. Appel, D. J., and Gordon, T. R. 1996. Relationships among pathogenic and nonpathogenic isolates of Fusarium oxysporum based on the partial sequence of the intergenic spacer region of the ribosomal DNA. Mol. Pl ant-Microbe Inter. 9:125-138. 5. Armstrong, G. M., and Armstrong, J. K. 1981. Formae speciales and races of Fusarium oxysporum causing wilt diseas es. Pages 391-399 in: Fusarium : diseases, biology, and taxonomy. P. E. Nelson, T. A. Tousson, and R. J. Cook, eds. Pensylvania State University Press, University Park, PA. 6. Ashley, M. V., and Dow, B. D. 1994. The us e of microsatellite analysis in population biology: background, methods, and potential application. Pages 185-201 in: Molecular Ecology and Evolution: Approach es and Applications. B. Schi erwater, B. Streit, G. P. Wanger, and R. DeSalle, eds. Birkhuser Verlag, Basel. 7. Aston, M. J., and Jones, M. M. 1976. A study of transpiration surfaces of Avena sterlis L. var. algerian leaves using monosilicic aci d as a tracer for water movement. Planta 130:121-129. 8. Baayen, R. P., O'Donnell, K., Bonants, P. J. M., Cigelnik, E., Kroon, L., Roebroeck, E. J. A., and Waalwijk, C. 2000. Gene gen ealogies and AFLP analyses in the Fusarium oxysporum complex identify monophyletic and nonmonophyletic formae speciales causing wilt and rot disease. Phytopathology 90:891-900. 9. Bai, Y. L., and Lindhout, P. 2007. Domestica tion and breeding of tomatoes: What have we gained and what can we gain in the future? Ann. Bot. 100:1085-1094. 10. Balmas, V., Scherm, B., Di Primo, P., Rau, D., Marcello, A., and Migheli, Q. 2005. Molecular characterisation of ve getative compatibility groups in Fusarium oxysporum f. sp. radicis lycopersici and f. sp. lycopersici by random amplification of polymorphic DNA and microsatellite-primed PCR. Eur. J. Plant Pathol. 111:1-8. 11. Beerli, P. 2004. Effect of unsampled populat ions on the estimation of population sizes and migration rates between sample d populations. Mol. Ecol. 13:827-836.

PAGE 115

115 12. Beerli, P. 2006. Comparison of Bayesi an and maximum-likelihood inference of population genetic parameters. Bioinformatics 22:341-345. 13. Benson, G. 1999. Tandem repeats finder: a program to analyze DNA sequences. Nucl. Acids Res. 27:573-580. 14. Bogale, M., Wingfield, B. D., Wingfield, M. J., and St eenkamp, E. T. 2005. Simple sequence repeat markers for species in the Fusarium oxysporum complex. Mol. Ecol. Notes 5:622-624. 15. Boutin-Ganache, I., Raposo, M., Raymond, M., and Deschepper, C. F. 2001. M13-tailed primers improve the readability and usability of microsatellite anal yses performed with two different allele-sizing me thods. Biotechniques 31:25-28. 16. Brammall, R. A., and Higgins, V. J. 1988. A histological co mparison of fungal colonization in tomato seedlings susceptible or resistant to Fusarium crown and root rot disease. Can. J. Bot. 66:915-925. 17. Brown, A. H. D., Feldman, M. W., and Ne vo, E. 1980. Multilocus structure of natural populations of Hordeum spontaneum Genetics 96:523-536. 18. Bull, J. J., Huelsenbeck, J. P., Cunningham, C. W., Swofford, D. L., and Waddell, P. J. 1993. Partitioning and combining data in phy logenetic analysis. Syst. Biol. 42:384-397. 19. Cai, G., Gale, L. R., Schneider, R. W., Kistle r, H. C., Davis, R. M., Elias, K. S., and Miyao, E. M. 2003. Origin of race 3 of Fusarium oxysporum f. sp. lycopersici at a single site in California. Phytopathology 93:1014-1022. 20. Cai, K., Gao, D., Chen, J., and Luo, S. 2009. Probing the mechanisms of silicon-mediated pathogen resistance. Plan t Sig. Behav. 4:1-3. 21. Chrif, M., Benhamou, N., and Belanger, R. R. 1991. Ultrastructura l and cytochemical studies of fungal development and host reac tions in cucumber plants infected by Pythium ultimum Phys. Mol. Plant Pathol. 39:353-375. 22. Chrif, M., and Belanger, R. R. 1992. Use of potassium silicate amendments in recirculating nutrient solutions to suppress Pythium ultimum on long English cucumber. Plant Dis. 76:1008-1011. 23. Ciampi, M. B., Meyer, M. C., Costa, M. J. N., Zala, M., McDonald, B. A., and Ceresini, P. C. 2008. Genetic structure of populations of Rhizoctonia solani anastomosis group-1 IA from soybean in Braz il. Phytopathology 98:932-941. 24. Correll, J. C., Klittich, C. J. R., and Lesl ie, J. F. 1987. Nitrate non-utilizing mutants of Fusarium oxysporum and their use in vegetative compatibility tests. Phytopathology 77:1640-1646.

PAGE 116

116 25. Dannon, E. A., and Wydra, K. 2004. Interact ion between silicon amendment, bacterial wilt development and phenotype of Ralstonia solanacearum in tomato genotypes. Phys. Mol. Plant Pathol. 64:233-243. 26. Datnoff, L. D., Rodrigues, F. ., and S eebold, K. W. 2007. Silicon and plant disease. Pages 233-246 in: Mineral Nutrition and Plant Dis ease. L. E. Datnoff, W. H. Elmer, and D. M. Huber, eds. The American Phytopa thological Society Press, St. Paul, MN. 27. Datnoff, L. E., Nemec, S., and Pernezny, K. 1995. Biological control of Fusarium crown and root rot of tomato in Florida using Trichoderma harzianum and Glomus intraradices Biol. Cont. 5:427-431. 28. Depaulis, F., Mousset, S., and Veuille, M. 2003. Power of neutrality tests to detect bottlenecks and hitchhiking. J. Mol. Evol. 57:S190-S200. 29. Dettman, J. R., Jacobson, D. J., and Taylor, J. W. 2003. A multilocus genealogical approach to phylogenetic species r ecognition in the model eukaryote Neurospora Evolution 57:2703-2720. 30. Dettman, J. R., Jacobson, D. J., Turner, E., Pringle, A., and Taylor, J. W. 2003. Reproductive isolation and phylogenetic divergence in Neurospora : Comparing methods of species recognition in a mode l eukaryote. Evolution 57:2721-2741. 31. Di Pietro, A., Madrid, M. P., Caracuel, Z., Delgado-Jarana, J., and Roncero, M. I. G. 2003. Fusarium oxysporum : exploring the molecular arse nal of a vascul ar wilt fungus. Mol. Plant Pathol. 4:315-325. 32. Diogo, R. V. C., and Wydra, K. 2007. Silicon-i nduced basal resistance in tomato against Ralstonia solanacearum is related to modification of pectic cell wall polysaccharide structure. Phys. Mol. Plant Pathol. 70:120-129. 33. Drummond, A. J., Ho, S. Y. W., Phillips, M. J., and Rambaut, A. 2006. Relaxed phylogenetics and dating with confidence. PLoS Biol. 4:699-710. 34. Drummond, A. J., and Rambaut, A. 2007. BEAS T: Bayesian evolutionary analysis by sampling trees. BMC Evol. Biol. 7:214. 35. Edwards, C. E., Arakaki, M., Quintana-Ascencio, P. F., Soltis, D. E., and Soltis, P. S. 2007. Isolation and characterization of microsat ellite loci from th e endangered highlands scrub hypericum ( Hypericum cumulicola ). Mol. Ecol. Notes 7:1135-1137. 36. Elias, K. S., and Schneider, R. W. 1991. Vegetative compatibility groups in Fusarium oxysporum f. sp. lycopersici Phytopathology 81:159-162. 37. Elias, K. S., and Schneider, R. W. 1992. Genetic diversity within and among races and vegetative compatibility groups of Fusarium oxysporum f. sp. lycopersici as determined by isozyme analysis. Phytopathology 82:1421-1427.

PAGE 117

117 38. Elias, K. S., Zamir, D., Lichtmanpleban, T., and Katan, T. 1993. Population structure of Fusarium oxysporum f. sp. l ycopersici : Restriction fragment length polymorphisms provide genetic evidence that vegetative compatibility group is an indicator of evolutionary origin. Mol. Pl ant-Microbe Inter. 6:565-572. 39. Ellegren, H. 2004. Microsatellites: Simple sequences with complex evolution. Nature Rev. Genetics 5:435-445. 40. Elliott, C. L., and Snyder, G. H. 1991. Auto clave-induced digestion for the colorimeteric determination of silicon in rice straw. J. Agric. Food Chem. 39:1118-1119. 41. Epistein, E., and Bloom, A. J. 2005. Minera l Nutrition of Plants: Principles and Perspectives. Sinauer Asso ciations, Sunderland, MA. 42. Epstein, E. 1994. The anomaly of s ilicon in plant biology. PNAS 91:11-17. 43. Epstein, E. 1999. Silicon. Ann. Rev. Plant Physiol. Plant Mol. Biol. 50:641-664. 44. Excoffier, L., Smouse, P. E., and Quattro, J. M. 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics 131:479-491. 45. Excoffier, L., Laval, G., and Schneider, S. 2005. Arlequin (version 3.0): an integrated software package for population genetics data analysis. Evol. Bioinf orm. Online 1:47-50. 46. Falush, D., Stephens, M., and Pritchard, J. K. 2003. Inference of population structure using multilocus genotype data: Linked loci and correlated allele frequencies. Genetics 164:1567-1587. 47. Farris, J. S., Kallersjo, M., Kluge, A. G ., and Bult, C. 1995. Testing significance of incongruence. Cladistics 10:315-319. 48. Fauteux, F., Remus-Borel, W., Menzies, J. G., and Belanger, R. R. 2005. Silicon and plant disease resistance ag ainst pathogenic fungi. FEMS Microbiol. Lett. 249:1-6. 49. Fauteux, F., Chain, F., Belzile, F., Menz ies, J. G., and Belanger, R. R. 2006. The protective role of silicon in the Arabidopsis-powdery mildew pathosystem. PNAS 103:17554-17559. 50. Fazio, G., Stevens, M. R., and Scott, J. W. 1999. Identification of RAPD markers linked to fusarium crown and root rot resistance (Frl ) in tomato. Euphytica 105:205-210. 51. Fravel, D., Olivain, C ., and Alabouvette, C. 2003. Fusarium oxysporum and its biocontrol. New Phytol. 157:493-502. 52. Geiser, D. M., Jimenez-Gasco, M. D., Kang, S. C., Makalowska, I., Veeraraghavan, N., Ward, T. J., Zhang, N., Kuldau, G. A., and O'Donnell, K. 2004. FUSARIUM-ID v. 1.0: A DNA sequence database for identifying Fusarium Eur. J. Plant Pathol. 110:473-479.

PAGE 118

118 53. Gillespie, D. R., and Menzies, J. G. 1993. Fungus gnats vector Fusarium oxysporum f. sp. radicis lycopersici Ann. Appl. Biol. 123:539-544. 54. Glaubitz, J. C. 2004. CONVERT: A user-frie ndly program to reformat diploid genotypic data for commonly used population genetic so ftware packages. Mol. Ecol. Notes 4:309310. 55. Gordon, T. R., and Martyn, R. D. 1997. The evolutionary biology of Fusarium oxysporum Annu. Rev. Phytopathol. 35:111-128. 56. Grnig, C. R., Du, A., and Sieber, T. N. 2006. Population genetic analysis of Phialocephala fortinii s.l. and Acephala applanata in two undisturbed forests in Switzerland and evidence for new cryptic species. Fungal Genet. Biol. 43:410-421. 57. Graur, D., and Li, W.-H. 2000. Fundamentals of molecular evolution. Sinauer Associates, Inc., Sunderland, MA. 58. Hancock, J. M. 1995. The contribution of slippage-like processes to genome evolution. J. Mol. Evol. 41:1038-1047. 59. Hartman, J. R., and Fletcher, J. T. 1991. Fusa rium crown and root rot of tomatoes in the UK. Plant Pathol. 40:85-92. 60. Heine, G., Tikum, G., and Horst, W. J. 2005. Silicon nutrition of tomato and bitter gourd with special emphasis on silicon distribution in root fractions. J. Plant Nutr. Soil Sci. 168:600-606. 61. Heine, G., Tikum, G., and Horst, W. J. 2007. The effect of silicon on the infection by and spread of Pythium aphanidermatum in single roots of tomato and bitter gourd. J. Exp. Bot. 58:569-577. 62. Hibar, K., Daami-Remadi, M., Ayed, F., and El-Mahjoub, M. 2007. Fusarium crown and root rot of tomato and its chemical c ontrol. Int. J. Agric. Res. 2:687-695. 63. Hirano, Y., and Arie, T. 2006. PCR-based differentiation of Fusarium oxysporum ff. sp. lycopersici and radicis lycopersic i and races of F oxysporum f. sp. lycopersici. J. Gen. Plant Pathol. 72:273-283. 64. Hoagland, D. R., and Arnon, D. I. 1950. The water-culture method for growing plants without soil. Circ. 347. Univ. of Ca lif. Agric. Exp. Station, Berkley. 65. Hovmller, M. S., Yahyaoui, A. H., Milus, E. A., and Justesen, A. F. 2008. Rapid global spread of two aggressive strains of a wheat rust fungus. Mol. Ecol. 17:3818-3826. 66. Huang, C.-H., Datnoff, L. D., Gale, L. R ., and Roberts, P. D. 2009. Development of microsatellite markers for Fusarium oxysporum f. sp. lycopersici and Fusarium oxysporum f. sp. radicis lycopersici Mol. Ecol. Res. (Accep ted).

PAGE 119

119 67. Huang, C., Roberts, P. D., and Datnoff, L. E. 2008. Developmen t of microsatellite markers for Fusarium oxysporum f. sp. lycopersici and Fusarium oxysporum f. sp radicis lycopersici (Abstr.) Phytopathology 98:S69-S70. 68. Huang, C., Roberts, P. D., and Datnoff, L. E. 2009. Phylogenetic relationships between Fusarium oxysporum f. sp. lycopersici and F oxysporum f. sp. radicis lycopersici inferred from IGS, EF1, and a SSR locus. (Abstr.) Phytopathology 99:S56. 69. Huang, C., Roberts, P. D., Gale, L. R., Elme r, W. H., and Datnoff, L. E. 2009. Population structure of Fusarium oxysporum f. sp. radicis lycopersici in Florida inferred from vegetative compatibility groups and microsat ellites. (Abstr.) Phytopathology 99:S55-56. 70. Huelsenbeck, J. P., and Rannala, B. 1997. P hylogenetic methods come of age: testing hypotheses in an evolutionary context. Science 276:227-232. 71. Jarvis, W. R., and Thorpe, H. J. 1976. Susceptibility of Lycopersicon species and hybrids to the foot and root rot pathogen Fusarium oxysporum Plant Dis. Rep. 60:1027-1031. 72. Jarvis, W. R., and Shoemaker, R. A. 1978. Taxonomic status of Fusarium oxysporum causing foot and root rot of tomato. Phytopathology 68:1679-1680. 73. Jarvis, W. R., and Thorpe, H. J. 1981. Control of fusarium foot and root rot of tomato by soil amendment with lettuce residues. Can. J. Plant Pathol. 3:159-162. 74. Jarvis, W. R. 1988. Fusarium crown and root rot of tomatoes. P hytoprotection 69:49-64. 75. Jones, J. B., Jones, J. P., Stall, J. P ., and Zitter, T. A. 1991. Compendium of tomato diseases. American Phytopathol ogical Society: St. Paul, MN. 76. Jones, J. P., Jones, J. B., Woltz, S. S., and Scott, J. W. 1993. Influence of soil pH, nitrogen source, and transplant drenches on development of crown rot of tomato. Proc. Fla. State Hort. Soc. 106:170-172. 77. Jorge, I., Navas-Cortes, J. A., Jimen ez-Diaz, R. M., and Tena, M. 2006. Cell wall degrading enzymes in Fusarium wilt of ch ickpea: correlation between pectinase and xylanase activities and disease development in plants infected with two pathogenic races of Fusarium oxysporum f. sp. ciceris Can. J. Bot. 84:1395-1404. 78. Karaoglu, H., Lee, C. M. Y., and Meyer, W. 2005. Survey of simple sequence repeats in completed fungal genomes. Mol. Biol. Evol. 22:639-649. 79. Katan, T., Hadar, E., and Katan, J. 1989. Vegetative compatibility of Fusarium oxysporum f. sp. dianthi from carnation in Israel Plant Pathol. 38:376-381. 80. Katan, T., Zamir, D., Sarfatti, M., and Katan, J. 1991. Vegetative compatibility groups and subgroups in Fusarium oxysporum f. sp. radicis lycopersici Phytopathology 81:255262.

PAGE 120

120 81. Katan, T. 1999. Current status of ve getative compatibility groups in Fusarium oxysporum Phytoparasitica 27:51-64. 82. Katan, T., and Katan, J. 1999. Vege tative compatibility grouping in Fusarium oxysporum f. sp. radicis lycopersici from the UK, the Netherlands, Belgium and France. Plant Pathol. 48:541-549. 83. Kawabe, M., Kobayashi, Y., Okada, G., Yamaguchi, I., Teraoka, T., and Arie, T. 2005. Three evolutionary lineages of tomato wilt pathogen, Fusarium oxysporum f. sp. lycopersici based on sequences of IGS, MAT1 and pg1, are each composed of isolates of a single mating type and a single or closely related vegetative compatibility group. J. Gen. Plant Pathol. 71:263-272. 84. Kerenyi, Z., Moretti, A., Waalwijk, C ., Olah, B., and Hornok, L. 2004. Mating type sequences in asexually reproducing Fusarium species. Appl. Environ. Microbiol. 70:4419-4423. 85. Kingman, J. F. C. 2000. Origins of th e coalescent: 1974-1982. Genetics 156:1461-1463. 86. Kistler, H. C., Alabouvette, C., Baayen, R. P., Bentley, S., Brayford, D., Coddington, A., Correll, J., Daboussi, M. J., Elias, K., Fernandez, D., Gordon, T. R., Katan, T., Kim, H. G., Leslie, J. F., Martyn, R. D., Migheli, Q., Moore, N. Y., O'Donnell, K., Ploetz, R. C., Rutherford, M. A., Summerell, B., W aalwijk, C., and Woo, S. 1998. Systematic numbering of vegetative compatibility groups in the plant pathogenic fungus Fusarium oxysporum Phytopathology 88:30-32. 87. Klaasen, J. A., and Nelson, P. E. 1996. Identification of a mating population, Gibberella nygamai sp. nov., within the Fusarium nygamai anamorph. Mycologia 88:965-969. 88. Klittich, C. J. R., and Leslie, J. F. 1988. Nitrate reduction mutants of Fusarium moniliforme ( Gibberella fujikuroi ). Genetics 118:417-423. 89. Komada, H. 1975. Development of a selec tive medium for quantitative isolation of Fusarium oxysporum from natural soil. Rev. Plant Prot. Res. 8:114-125. 90. Kuhner, M. K. 2006. LAMARC 2.0: maximum likelihood and Bayesian estimation of population parameters. Bioinformatics 22:768-770. 91. Kuhner, M. K., and Smith, L. P. 2007. Co mparing likelihood and Bayesian coalescent estimation of population para meters. Genetics 175:155-165. 92. Kuniyasu, K., and Yamakawa, K. 1983. Contro l of Fusarium wilt of tomato caused by Fusarium oxysporum f. sp. lycopersici race J3 by grafting to KNVF and KVF rootstocks of interspecific hybrids between Lyco persicon esculentum L hisutum Ann. Phytopathol. Soc. Jpn. 49:581-586.

PAGE 121

121 93. Larkin, M. A., Blackshields, G., Brown, N. P., Chenna, R., McGettigan, P. A., McWilliam, H., Valentin, F., Wallace, I. M ., Wilm, A., Lopez, R., Thompson, J. D., Gibson, T. J., and Higgins, D.G. 2007. Clustal W and Clustal X version 2.0. Bioinformatics 23:2947-2948. 94. Leary, J. V., and Sanchez, L. E. 1976. Mutagenesis of Fusarium oxysporum : analysis of the fungal genome in relation to the expressi on and regulation of pathogenicity. Pages 1215 in: Biochemistry and cytology of Plant-Parasi te Interaction. K. Tomiyama, J. M. Daly, I. Uritani, H. Oku, and S. Ouchi, eds. Kodansha Ltd, Tokyo. 95. Leslie, J. F. 1993. Fungal vegetative comp atibility. Ann. Rev. Phytopathol. 31:127-150. 96. Leslie, J. F. 1996. Fungal vegetative comp atibility promises and prospects. Phytoparasitica 24:3-6. 97. Leslie, J. F., and Summerell, B. A. 2006. The Fusarium laboratory manual. Blackwell Publishing, Ames, IA. 98. Lievens, B., Rep, M., and Thomma, B. 2008. Recent developments in the molecular discrimination of formae speciales of Fusarium oxysporum Pest Manag. Sci. 64:781-788. 99. Ma, J. F., Miyake, Y., and Takahashi, E. 2001. Silicon as a benefi cial element for crop plants. Pages 17-39 in: Silicon in Agriculture: Studies in Plant Science, Vol. 8. L. E. Datnoff, G. H. Snyder, and G. H. Korndorfer, eds. Elsevi er Science, Amsterdam, The Netherlands. 100. Ma, J. F., and Takahashi, E. 2002. Soil, Fert ilizer, and Plant Silicon Research in Japan. Elsevier Science, Amsterdam. 101. Ma, J. F., Tamai, K., Yamaji, N., Mitani, N., Konishi, S., Kats uhara, M., Ishiguro, M., Murata, Y., and Yano, M. 2006. A silicon tr ansporter in rice. Nature 440:688-691. 102. Maddison, W. P. 1997. Gene trees in species trees. Syst. Biol. 46:523-536. 103. Manicom, B. Q., Bar-Joseph, M., and Kotze, J. M. 1990. Molecular methods of potential use in the identification and taxonomy of filament ous fungi, particularly Fusarium oxysporum Phytophylactica 22:233-240. 104. Manicom, B. Q., Barjoseph, M., Kotze, J. M., and Becker, M. M. 1990. A restriction fragment length polymorphism probe rela ting vegetative compatibility groups and pathogenicity in Fusarium oxysporum f. sp. dianthi Phytopathology 80:336-339. 105. Marlatt, M. L., Correll, J. C., Kaufmann, P., and Cooper, P. E. 1996. Two genetically distinct populations of Fusarium oxysporum f. sp. lycopersici race 3 in the United States. Plant Dis. 80:1336-1342.

PAGE 122

122 106. Marois, J. J., and Mitchell, D. J. 1981. E ffects of fumigation and fungal antagonists on the relationships of inoculum density to in fection incidence and disease severity in Fusarium crown rot of tomato. Phytopathology 71:167-170. 107. McDonald, B. A., and Linde, C. 2002. Pa thogen population genetics, evolutionary potential, and durable resistance. Ann. Rev. Phytopathol. 40:349-379. 108. McDonald, B. A., and Linde, C. 2002. The popul ation genetics of plant pathogens and breeding strategies for durable resistance. Euphytica 124:163-180. 109. McGovern, R. J., Vavrina, C. S., Noling, J. W., Datnoff, L. D., and Yonce, H. D. 1998. Evaluation of different application methods of metam sodium for management of Fusarium crown and root rot in tomato in southwest Florida. Plant Dis. 82:919-923. 110. Menzies, J. G., Koch, C., and Seywerd, F. 1990. Additions to the host ranges of Fusarium oxysporum f. sp. radicis lycopersici Plant Dis. 74:569-572. 111. Menzies, J. G., and Jarvis, W. R. 1994. The infestation of tomato seed by Fusarium oxysporum f. sp. radicis lycopersici Plant Pathol. 43:378-386. 112. Menzies, J. G., Ehret, D. L., Chrif, M., and Blanger, R. R. 2001. Plant-related silicon research in Canada. Pages 323-341 in: Silicon in Agriculture: Studies in Plant Science, Vol. 8. L. E. Datnoff, G. H. Snyder, and G. H. Krndorfer, eds. Elsevier Science, Amsterdam, The Netherlands. 113. Milgroom, M. G. 1996. Recombination and the multilocus structure of fungal populations. Ann. Rev. Phytopathol. 34:457-477. 114. Montarry, J., Corbiere, R., Lesueur, S., Gl ais, I., and Andrivon, D. 2006. Does selection by resistant hosts trigger local adaptation in plant-pathoge n systems? J. Evol. Biol. 19:522-531. 115. Nei, M. 1987. Molecular Evolutionary Genetics. Columbia University Press, New York, NY. 116. Nelson, P. E., Tousson, T. A., and Marasas, W. F. O. 1983. Fusarium species: An illustrated Manual for Identification. Pennsylvania State University Press, PA. 117. Nirenberg, H. I., and O'Donnell, K. 1998. New Fusarium species and combinations within the Gibberella fujikuroi species complex. Mycologia 90:434-458. 118. O'Donnell, K., and Cigelnik, E. 1997. Two divergent intragenomic rDNA ITS2 types within a monophyletic lineage of the fungus Fusarium are nonorthologous. Mol. Phyl. Evol. 7:103-116. 119. O'Donnell, K., Kistler, H. C., Cigelnik, E., and Ploetz, R. C. 1998. Multiple evolutionary origins of the fungus causing Panama diseas e of banana: Concordant evidence from nuclear and mitochondrial gene genealogies. PNAS 95:2044-2049.

PAGE 123

123 120. Olivain, C., and Alabouvette, C. 1997. Coloni zation of tomato root by a non-pathogenic strain of Fusarium oxysporum New Phytol. 137:481-494. 121. Olivain, C., and Alabouvette, C. 1999. Process of tomato root colonization by a pathogenic strain of Fusarium oxysporum f. sp. lycopersici in comparison with a nonpathogenic strain. New Phytol. 141:497-510. 122. Peaslee, D. E., and Frink, C. R. 1969. Influe nce of silicic acid on uptake of Mn, Al, Zn, and Cu by tomatoes ( Lycopersicum esculentum ) grown an an acid soil. Soil Sci. Soc. Am. Proc. 33:569-571. 123. Peralta, I. E., and Spooner, D. M. 2007. History, origin and early cu ltivation of tomato (Solanaceae). Pages 1-24 in: Genetic impr ovement of solanaceous crops. Volume 2: tomato. M. K. Razdan and A. K. Mattoo, eds. Science Publishers, Enfield, NH. 124. Posada, D., and Crandall, K. A. 1998. MODELTEST: testing the model of DNA substitution. Bioinformatics 14:817-818. 125. Primo, P. D., Cartia, G., a nd Katan, T. 2001. Vegetative co mpatibility and heterokaryon stability in Fusarium oxysporum f. sp. radicis lycopersici from Italy. Plant Pathol. 50:371-382. 126. Pritchard, J. K., Stephens, M., and Donnell y, P. 2000. Inference of population structure using multilocus genotype data. Genetics 155:945-959. 127. Prospero, S., Grunwald, N. J., Winton, L. M., and Hansen, E. M. 2009. Migration Patterns of the Emerging Plant Pathogen Phytophthora ramorum on the West Coast of the United States of America. Phytopathology 99:739-749. 128. Puhalla, J. E. 1985. Classification of strains of Fusarium oxysporum on the basis of vegetative compatibility. Can. J. Bot. 63:179-183. 129. Pyhjrvi, T., Garca-Gil, M. R., Kn urr, T., Mikkonen, M ., Wachowiak, W., and Savolainen, O. 2007. Demographi c history has influenced nucleotide diversity in European Pinus sylvestris populations. Genetics 177:1713-1724. 130. Raaijmakers, J. M., Paulitz, T. C., Steinber g, C., Alabouvette, C., and Moenne-Loccoz, Y. 2009. The rhizosphere: a playground and batt lefield for soilborne pathogens and beneficial microorganism s. Plant Soil 321:341-361. 131. Raymond, M., and Rousset, F. 1995. GE NEPOP Version 1.2: population genetics software for exact tests and ecumenicism. J. Hered. 86:248-249. 132. Recorbet, G., Steinberg, C ., Olivain, C., Edel, V., Trouvelot, S., Dumas-Gaudot, E., Gianinazzi, S., and Alabouvette, C. 2003. Wanted: pathogenesis-related marker molecules for Fusarium oxysporum New Phytol. 159:73-92.

PAGE 124

124 133. Rekah, Y., Shtienberg, D., and Katan, J. 1999. Spatial distribution and temporal development of Fusarium crown and root rot of tomato and pathogen dissemination in field soil. Phytopathology 89:831-839. 134. Rekah, Y., Shtienberg, D., and Kata n, J. 2001. Population dynamics of Fusarium oxysporum f. sp. radicis lycopersici in relation to the onset of Fusarium crown and root rot of tomato. Eur. J. Plant Pathol. 107:367-375. 135. Richmond, K. E., and Sussman, M. 2003. Got si licon? The non-essentia l beneficial plant nutrient. Curr. Opin. Plant Biol. 6:268-272. 136. Ronquist, F., and Huelsenbeck, J. P. 2003. Mr Bayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19:1572-1574. 137. Rosenberg, N. A., and Nordborg, M. 2002. Gen ealogical trees, coalescent theory and the analysis of genetic polymorphism s. Nat. Rev. Genet. 3:380-390. 138. Rosenberg, N. A. 2004. DISTRUCT: a program for the graphical display of population structure. Mol. Ecol. Notes 4:137-138. 139. Rosewich, U. L., Pettway, R. E., Katan, T., and Kistler, H. C. 1999. Population genetic analysis corroborates dispersal of Fusarium oxysporum f. sp. radicis lycopersici from Florida to Europe. Phytopathology 89:623-630. 140. Rowe, R. C., Farley, J. D., and Copli n, D. L. 1977. Airborne spore dispersal and recolonization of steamed soil by Fusarium oxysporum in tomato greenhouses. Phytopathology 67:1513-1517. 141. Rowe, R. C. 1980. Comparative pathogenicity and host ranges of Fusarium oxysporum isolates causing crown and root rot of gr eenhouse and field -grown tomatoes in North America and Japan. Phytopathology 70:1143-1148. 142. Rozen, S., and Skaletsky, H. 2000. Primer 3 on the WWW for general users and for biologist programmers. Pages 365-386 in: Bioinformatics methods and protocols: methods in molecular biology. S. Krawetz and S. Misener, eds. Humana Press, Totowa, NJ. 143. Sacristan, S., and Garcia-Arenal, F. 2008. The evolution of vi rulence and pathogenicity in plant pathogen populations. Mol. Plant Pathol. 9:369-384. 144. Samuels, A. L., Glass, A. D. M., Ehret, D. L., and Menzies, J. G. 1991. Mobility and deposition of silicon in cucumber plants. Plant Cell Environ. 14:485-492. 145. Sanchez, L. E., Endo, R. M., and Leary, J. V. 1975. A rapid technique for identifying the clones of Fusarium oxysporum f. sp. lycopersici causing crown-and root-rot of tomato. Phytopathology 65:726-727.

PAGE 125

125 146. Sato, R., and Araki, T. 1974. On the tomato root-rot disease occurr ing under vinyl-house conditions in southern Hokkaido. Annu. Re p. Soc. Plant Prot. North Jpn. 25:5-13. 147. Schweigkofler, W., O'Donnell, K., and Garb elotto, M. 2004. Detecti on and quantification of airborne conidia of Fusarium circinatum the causal agent of pine pitch canker, from two California sites by using a real-time PCR approach combined with a simple spore trapping method. Appl. Environ. Microbiol. 70:3512-3520. 148. Scott, J. W. 1999. University of Florida tomato breeding accomplishments and future directions. Soil Crop Sci. Soc. Fla. Proc. 58:8-11. 149. Scott, J. W., and Jones, J. P. 2000. Fla. 7775 and Fla. 7781: Tomato breeding lines resistant to Fusarium crown and root rot. Hortscience 35:1183-1184. 150. Scott, J. W., and Gardner, R. G. 2007. Breed ing for resistance to fungal pathogens. Pages 421-456 in: Genetic improvement of solanace ous crops. Volume 2: tomato. M. K. Razdan and A. K. Mattoo, eds. Science Publishers, Enfield, NH. 151. Scott, J. W. 2008. Fresh market tomato breed ing in the USA. Acta Horticult. 789:21-25. 152. Sexton, A. C., and Howlett, B. J. 2004. Microsatellite markers reveal genetic differentiation among populations of Sclerotinia sclerotiorum from Australian canola fields. Curr. Genet. 46:357-365. 153. Shackelton, L. A., Rambaut, A., Pybus, O. G., and Holmes, E. C. 2006. JC virus evolution and its association with human populations. J. Virol. 80:9928-9933. 154. Shaner, G., and Finney, R. E. 1977. The effect of nitrogen fertilizer on the expression of slow mildewing resistance in Knox wheat. Phytopathology 67:1051-1056. 155. Sims, W. L. 1980. History of tomato pr oduction for industry around the world. Acta Horticul. 100:25-26. 156. Slatkin, M. 1995. A measure of population subdivision based on microsatellite allele frequencies. Genetics 139:457-462. 157. Smith, J. M., Smith, N. H., Orourke, M., and Spratt, B. G. 1993. How clonal are bacteria? PNAS 90:4384-4388. 158. Sommer, M., Kaczorek, D., Kuzyakov, Y., and Breuer, J. 2006. Silicon pools and fluxes in soils and landscapes a review. J. Plant Nutr. Soil Sci. 169:310-329. 159. Sonoda, R. M. 1976. The occurrence of a Fusarium root rot of tomato in south Florida. Plant Dis. Rep. 60:271-274. 160. Stukenbrock, E. H., Banke, S., and McDona ld, B. A. 2006. Global migration patterns in the fungal wheat pathogen Phaeosphaeria nodorum Mol. Ecol. 15:2895-2904.

PAGE 126

126 161. Stukenbrock, E. H., and McDonald, B. A. 2008. The origins of plant pathogens in agroecosystems. Ann. Rev. Phytopathol. 46:75-100. 162. Swofford, D. L. 2001. PAUP*. Phylogenetic Analysis Using Parasimony (*and Other Methods). v.4 ed. Sinauer Associates, Sunderland, MA. 163. Tajima, F. 1983. Evolutionary relationship of DNA sequences in finite populations. Genetics 105:437-460. 164. Tajima, F. 1989. Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics 123:585-595. 165. Tajima, F. 1993. Measurment of DNA pol ymorphism. Pages 37-59 in: Mechanisms of Molecular Evolution: Introduction to Molecular Paleopopulation Biology. N. Takahata and A. G. Clark, eds. Sinauer Associates, Sunderland, MA. 166. Tamura, K., Dudley, J., Nei, M., and Kuma r, S. 2007. MEGA4: Mo lecular evolutionary genetics analysis (MEGA) software version 4.0. Mol. Biol. Evol. 24:1596-1599. 167. Tautz, D., and Renz, M. 1984. Simple sequences are ubiquitous repe titive components of eukaryotic genomes. Nucl. Acids Res. 12:4127-4138. 168. Taylor, J. W., Jacobson, D. J., and Fisher M. C. 1999. The evolution of asexual fungi: Reproduction, speciation and classifica tion. Ann. Rev. Phytopathol. 37:197-246. 169. Taylor, J. W., Jacobson, D. J., Kroken, S., Ka suga, T., Geiser, D. M., Hibbett, D. S., and Fisher, M. C. 2000. Phylogenetic species r ecognition and species concepts in fungi. Fungal Genet. Biol. 31:21-32. 170. Teunissen, H. A. S., Verkooijen, J., Corne lissen, B. J. C., and Haring, M. A. 2002. Genetic exchange of avirulence determinants and extensive karyotype rearrangements in parasexual recombinants of Fusarium oxysporum Mol. Gen. Genet. 268:298-310. 171. Validov, S. Z., Qil, S., Azarova, T., Kamilova, F., and Lugtenberg, B. J. J. 2006. Heterogeneity of Fusarium oxysporum strains isolated as formae specialis radicis lycopersici Bulletin OILB/SROP 29:151-155. 172. van der Does, H. C., Lievens, B., Claes, L., Houterman, P. M., Cornelissen, B. J. C., and Rep, M. 2008. The presence of a virulence locus discriminates Fusarium oxysporum isolates causing tomato wilt from othe r isolates. Environ. Microbiol. 10:1475-1485. 173. van Oosterhout, C., Hutchinson, W. F., W ills, D. P. M., and Shipley, P. 2004. MICROCHECKER: software for identifying and corr ecting genotyping errors in microsatellite data. Mol. Ecol. Notes 4:535-538. 174. Walker, J. C. 1971. Fusarium wilt of tomato. Monogr. 6. American Phytopathology Society, St. Paul, MN.

PAGE 127

127 175. Ward, J. H. 1963. Hierarchical grouping to opt imize an objective f unction. J. American Stat. Assoc. 58:236-244. 176. Woudt, L. P., Neuvel, A., Sikkema, A., Vangr insven, M., Demilliano, W. A. J., Campbell, C. L., and Leslie, J. F. 1995. Genetic variation in Fusarium oxysporum from cyclamen. Phytopathology 85:1348-1355. 177. Wright, G. K., Pascoc, I., van Heeswijc k, R., Guest, D., and Wimalajeewa, S. 1992. Characterisation of isolates of Fusarium oxysporum from carnations in Victoria. Acta Horticul. 307:65-72. 178. Xu, L., Nonomura, T., Suzuki, S., Kitagawa Y., Tajima, H., Okada, K., Kusakari, S., Matsuda, Y., and Toyoda, H. 2006. Symptomatic evidence for differential root invasion by Fusarium crown and root rot pathogens between common tomato Lycopersicon esculentum and its varieties. J. Phytopathol. 154:577-586. 179. Yoshida, S., Ohnishi, Y., and Kitagish i, K. 1962. Chemical forms, mobility and deposition of silicon in rice plant. Soil Sci. Plant Nutr. 8:107-113. 180. Yu, N., Jensen-Seaman, M. I., Chemnick, L., Ryder, O., and Li, W. H. 2004. Nucleotide diversity in gorilla s. Genetics 166:1375-1383. 181. Yun, S. H., Arie, T., Kaneko, I., Yoder, O. C., and Turgeon, B. G. 2000. Molecular organization of mating type loci in he terothallic, homothallic, and asexual Gibberella / Fusarium species. Fungal Gen. Biol. 31:7-20. 182. Zane, L., Bargelloni, L., and Patarnello, T. 2002. Strategies for micros atellite isolation: a review. Mol. Ecol. 11:1-16. 183. Zheng, Y., Zhang, G., Lin, F., Wang, Z., Jin, G., Yang, L., Wang, Y., Chen, X., Xu, Z., Zhao, X., Wang, H., Lu, J., Lu, G., and W u, W. 2008. Development of microsatellite markers and construction of gene tic map in rice blast pathogen Magnaporthe grisea Fungal Gen. and Biol. 45:1340-1347. 184. Zwickl, D. J. 2006. Genetic algorithm approach es for the phylogenetic analysis of large biological sequence datasets under the maxi mum likelihood criterion. Ph.D. dissertation. The University of Texas, Austin.

PAGE 128

128 BIOGRAPHICAL SKETCH Cheng-Hua Huang was born in Taiwan. He receive d his bachelors and masters degrees in soil science in Taiwan. His thesis is entitled Effects of Nitrogen Forms and Organic Acids on the Infection of Gray Mold on Gloxinia by Botrytis cinerea From 2002 to 2005, he was employed at the MOA Internati onal Foundation of Nature Ecology in Taiwan, where he was an organic inspector interacting with hundreds of organic farmers. Meanwhile, he became aware of the importance of plant pathology, leading him to pursue an advanced degree in the United States. In 2005, his major advisor, Dr. Pamela D. Roberts, kindly provided an assistantship for his work on Fusarium oxysporum f. sp. radicis lycopersici, the causal agent of Fu sarium crown and root rot of tomato. His co-advisor, Dr. Lawrence E. Datnoff, enlightened him on the research of silicon effects on plant disease. Cheng-Hua received his Ph.D. from the Univers ity of Florida in the fall of 2009. He plans to continue studying phylogenetics, population genetics, and bio-base d control strategies of plant pathogens beginning in January 201 0 as a post-doc with Dr. Gary E. Vallad at the Gulf Coast Research and Education Center, UF/IFAS.