Fitness, Molecular Characterization and Management of Bacterial Leaf Spot in Tomatoes

FITNESS, MOLECULAR CHARACTERIZATION AND MANAGEMENT OF BACTERIAL LEAF SPOT IN TOMATOES

By PETER ABRAHAMIAN

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA 2017

ACKNOWLEDGEMENTS

I would like to thank my major advisor Dr. Gary Vallad for his extensive guidance, advising and help throughout my PhD studies. Dr. Vallad was always encouraging and supportive of pursuing new ideas. I also enjoyed all the times we discusses science. You are an awesome advisor! I also thank Dr. Jeffrey Jones my co-advisor for his extensive support, help and openness throughout my studies. I want to thank my committee members Drs. Erica Goss,

Mathews Paret and Samuel Hutton for their constructive and insightful comments for improving the quality of my research.

I would like to thank all the members of the Vallad Lab at GCREC. Thank you Ai-vy

Riniker for being there for me when I first started working at Dr. Vallad’s lab. I also thank

Rebecca Willis, Scott Hughes and Steve Kalb for their excellent technical help in securing, maintaining and harvesting tomato plants throughout the work conducted at GCREC. I also thank Dr. Aimen Wen, Heather Adkison and Late Wilson and my colleagues, Tyler Jacoby,

Caroline Land, and Andy Shirley, for all the fun times and stimulating talks at the lab. Special thanks to Sujan Timilsina for all helping me with the bioinformatics work and to Sushmita KC for helping in collecting and processing samples in the field and greenhouse experiments.

I want to thank also Dr. Neha Potnis and Jerry Minsavage at the Jones lab in Gainesville for training me in the techniques required for this study.

Finally, I would like to thank my parents and family who accompanied me throughout this journey that started a long time ago. Thank you for your patience and continuous support! I also thank my brother, Roy, for his encouragement.

TABLE OF CONTENTS page ACKNOWLEDGEMENTS ...... 3

LIST OF TABLES ...... 6

LIST OF FIGURES ...... 8

ABSTRACT ...... 10

CHAPTER 1 INTRODUCTION ...... 12 Hosts and Symptomology ...... 13 Distribution and Epidemiology ...... 14 Phenotypic and Genetic Diversity ...... 17 Pathogenicity and Effectors ...... 22 Disease Control ...... 27

2 THE TYPE III EFFECTOR AVRBST ENHANCES XANTHOMONAS PERFORANS FITNESS IN FIELD-GROWN TOMATO ...... 32

Materials and Methods ...... 36 Bacterial Strains, Plasmids and Plants ...... 36 Mutant Construction ...... 37 In planta Growth and Competition Assays ...... 38 Field Trials ...... 39 Colony Testing ...... 40 Statistical Analyses ...... 41 Results ...... 41 In planta Activity and Leaf Infiltration Assays ...... 41 Disease Severity in Field Plants ...... 42 Bacterial Populations in the Field ...... 43 Effect of avrBsT on Movement...... 44 Effect of avrBsT on Strain Recovery ...... 45 Discussion...... 46

3 TRACING XANTHOMONAS PERFORANS POPULATIONS ON TOMATO FROM GREENHOUSE- TO FIELD BY WHOLE-GENOME SEQUENCING ...... 59

Materials and Methods ...... 62 Bacterial Strains and Race Characterization ...... 62 Genome Sequencing and de novo Assembly ...... 63 Characterization with Multi-Locus Sequence Analyses ...... 63

Core-Genome Multi-Locus Sequence Typing ...... 64 Single-Nucleotide Variations Calling ...... 64 Phylogenetic Analyses ...... 65 Effector Analyses ...... 66 Results ...... 66 Genomes and Average Nucleotide Identities ...... 66 Multi-Locus Sequence Analyses...... 66 Core Genome Comparison ...... 67 Genome-Wide SNP Analyses ...... 68 Race Characterization and Effector Profiles ...... 70 Discussion...... 71

4 EFFICACY OF COPPER ALTERNATIVES FOR MANAGEMENT OF BACTERIAL SPOT ON TOMATOES VARIES IN TRANSPLANT AND FIELD PRODUCTION ...... 90

Materials and Methods ...... 93 Bacterial Strains, Inoculation and Plants ...... 93 Greenhouse Trials ...... 94 Field Trials ...... 95 Statistical Analyses ...... 96 Results ...... 97 Efficacy of BST Reduction in Tomato Seedlings ...... 97 BST Disease Level in Field-Grown Tomato ...... 99 Discussion...... 101

5 SUMMARY AND DISCUSSION ...... 114

APPENDIX

A ENVIRONMENT AND STATISTICS TABLES ...... 119

B ADDITIONAL STRAIN CHARACTERIZATION DETAILS ...... 128

LIST OF REFERENCES ...... 144

BIOGRAPHICAL SKETCH ...... 160

LIST OF TABLES

Table page

2-1 List of bacterial strains and plasmids ...... 51

3-1 List of Xanthomonas strains characterized and used in this study ...... 78

3-2 Average number of SNPs, coverage and effects on coding domain sequence for each phylogenetic group ...... 80

3-3 Range of genetic distances between transplant and field in group 3 ...... 81

3-4 Range of genetic distances between transplant and field grower A strains in group 2. ....82

3-5 Range of genetic distances between transplant and field grower B strains in group 2...... 83

4-1 List of chemical product names, active ingredients, manufacturers used in this study ...107

4-2 Tomato yield, average yield and percent of fruit during fall 2016 ‘A’ ...... 108

4-3 Tomato yield, average yield and percent of fruit during fall 2016 ‘B’ ...... 109

A-1 Average daily rainfall, relative humidity and temperature for each trial ...... 119

A-2 Type III tests of fixed effects for total bacterial populations levels of Xanthomonas perforans from in planta infiltrations in greenhouse trials ...... 120

A-3 Type III tests of fixed effects for colony recovery of Xanthomonas perforans GEV1001 from co-infiltration leaf assays in greenhouse trials ...... 121

A-4 Type III tests of fixed effects for total bacterial populations levels recovered from field plants in spring 2016 ...... 122

A-5 Type III tests of fixed effects for total bacterial populations recovered from field plants in spring 2017...... 123

A-6 Type III tests of fixed effects for movement of Xanthomonas perforans GEV872 in field trials...... 124

A-7 Type III tests of fixed effects for movement of Xanthomonas perforans GEV1001 in field trials...... 125

A-8 Type III tests of fixed effects for colony recovery of Xanthomonas perforans GEV872 in field trials ...... 126

A-9 Type III tests of fixed effects for colony recovery of Xanthomonas perforans GEV1001 in field trials ...... 127

B-1 Additional Xanthomonas strains used in this study ...... 128

B-2 Genome details of sequenced strains collected from grower A ...... 129

B-3 Genome details of sequenced strains collected from grower B ...... 130

B-4 Average nucleotide identity showing a one-way comparison of Xanthomonas perforans transplant and field grower A strains ...... 132

B-5 Average nucleotide identity showing a one-way comparison of Xanthomonas perforans transplant and field grower B strains ...... 133

B-6 Singe nucleotide polymorphism details for individually sequenced strains ...... 135

B-7 The range of average genetic distance of transplant strains compared to grower A field strains recovered from various cultivars ...... 138

B-8 The range of average genetic distance of transplant strains compared to grower B field strains recovered from various cultivars based on ...... 139

B-9 Race profile of characterized strains collected from grower A ...... 140

B-10 Race profile of characterized strains collected from grower B ...... 141

LIST OF FIGURES

Figure page 2-1 Infiltration of pepper cv. Early Calwonder and field plots ...... 52

2-2 Growth curve showing population dynamics of Xanthomonas perforans wild-type and mutant strains ...... 53

2-3 Infiltration of Xanthomonas perforans wild-type + mutant co-mixtures for GEV872 and GEV1001 in tomato ...... 54

2-4 Disease severity levels of bacterial spot caused by inoculated Xanthomonas perforans genotypes GEV872 and GEV1001 ...... 55

2-5 Total populations of Xanthomonas perforans GEV872 and GEV1001 recovered at different distances across the plot ...... 56

2-6 Recovery of Xanthomonas perforans GEV872 and GEV1001 wild-type and mutant strains across plots ...... 57

2-7 Average number of colonies recovered for Xanthomonas perforans wild-type and mutant strains ...... 58

3-1 Maximum-likelihood phylogenetic tree of Xanthomonas perforans strains based on housekeeping genes ...... 84

3-2 Maximum-likelihood phylogenetic tree of 73 Xanthomonas perforans strains based on core-genome sequences ...... 85

3-3 Maximum-likelihood phylogenetic tree of 73 Xanthomonas perforans strains based on single-nucleotide polymorphism with a core-genome heat map ...... 86

3-4 Maximum-likelihood phylogenetic tree of group 3 strains of Xanthomonas perforans based on single-nucleotide polymorphism ...... 87

3-5 Maximum-likelihood phylogenetic tree of group 2 strains of Xanthomonas perforans based on single-nucleotide polymorphism ...... 88

3-6 Effector profile of Xanthomonas perforans overlapped with the single-nucleotide polymorphism maximum-likelihood phylogenetic tree ...... 89

4-1 Treatments evaluated for controlling bacterial spot in the greenhouse during fall 2015 of trials 1 and 2 ...... 110

4-2 Treatments evaluated for controlling bacterial spot in the greenhouse during fall 2015 of trials 3 and 4 ...... 111

4-3 Treatments evaluated for controlling bacterial leaf spot in the greenhouse during fall 2016 and spring 2017 ...... 112

4-4 Evaluation of chemical treatments on field tomato for controlling bacterial leaf spot of tomato during fall 2015 ...... 113

B-1 Neighbor-joining phylogenetic tree of the YopJ-like homologs of plant pathogenic bacteria ...... 143

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

FITNESS, MOLECULAR CHARACTERIZATION AND MANAGEMENT OF BACTERIAL LEAF SPOT IN TOMATOES

Peter Abrahamian

December 2017

Chair: Gary E. Vallad Co-Chair: Jeffrey B. Jones Major: Plant Pathology

Bacterial spot of tomatoes (BST) caused by Xanthomonas perforans (Xp), is a ubiquitous disease of tomato in both field and transplant operations in Florida. BST is a significant disease that causes up to 50% yield losses. BLS in Florida is caused by two phylogenomic groups of Xp.

Recent surveys showed an increasing prevalence of the type III secretion effector, avrBsT, among Xp strains. We generated mutant strains with non-functional avrBsT genes in two genotypes representative of the phylogenomic groups found in Florida. The contribution of avrBsT to the fitness of Xp on tomato based on in planta growth, competition, movement and epiphytic survival in comparison to the wild-type strain in tomato under field conditions was evaluated. Findings suggest that avrBsT contributes to the fitness of Xp strains under field conditions, making it an ideal candidate for BLS resistance breeding efforts in tomato. In addition, we conducted a population genetics study by tracking strain movement from the seedling to the field. The draft genomes of 67 strains collected from two major tomato operations were sequenced. Phylogenetic analyses, race testing and effector profiles were established for all the strains. Single-nucleotide polymorphism analyses indicated a very high similarity between

10 strains originating on seedlings in greenhouses and field plants. Disease control of BST relies heavily on using the grower standard, copper or copper-mancozeb sprays. Xp is highly resistant to copper which renders spraying program inefficacious. We evaluated 19 different chemical agents, biological control agents, plant defense activators and novel products for their ability to control bacterial spot on tomato seedlings and in the field. Overall, Actigard provided significant disease reduction as a stand-alone or in a program with other compounds in both the greenhouse and field. However, Cueva alone or as a tank mixture with other compounds are promising alternatives in seedling production. The use of alternative compounds should improve tomato health, reduce disease inoculum introduction into fields, and reduction of residual copper in the environment.

CHAPTER 1 INTRODUCTION

Tomato, Solanum lycopersicon, is an important cash crop worth more than 2.5 billion

U.S. dollars nationwide. Florida is the leader of fresh tomato production with 33,000 planted acres in 2015 at an estimated value of half a billion U.S. dollars (NASS, 2016). From 2005 to

2015, the fresh tomato yield per acre and production value in the U.S. decreased from 15.4 to

14.5 tons and from $1.59 to $1.24 billion, respectively (NASS, 2008; 2016). The cultivated acreage for fresh market tomato production dropped from 131,800 to 97,500 within the same 10- year period (NASS, 2008; 2016).

Tomato is widely susceptible to many fungal, bacterial and viral diseases (Pohronezny et al., 1986). In Florida, bacterial diseases such as bacterial spot, bacterial speck and bacterial wilt can cause losses of marketable yields that may reach up to 50% (Pohronezny and Volin, 1983).

Bacterial spot of tomato (BST) is reported as a challenging factor for crop development during the growing season (Pohronezny and Volin, 1983; Pohronezny et al., 1986). Unlike other parts of the United States, Florida’s subtropical climate consisting of high humidity and prolonged wet periods is highly conducive to bacterial spot, making it a significant disease for sustainable fresh tomato production (Pohronezny et al., 1986). Bacterial spot is caused by the phytopathogenic bacteria Xanthomonas euvesicatoria, X. vesicatoria, X. perforans (Xp) and X. gardneri (Jones et al., 2004). Historically, only two species, X. euvesicatoria and Xp, are reported to cause bacterial spot in Florida (Stall et al., 2009).

Xanthomonas spp. belong to the family Xanthomonadaceae in the order

Xanthomonadales in the class Gammaproteobacteria (Saddler and Bradbury 2005). Xanthomonas spp. are rod-shaped (0.4–0.6 x 0.8–2.0 µm), obligate aerobes and mono-flagellated gram- negative bacteria that produce high amounts of an extracellular polysaccharide (EPS) called

12 xanthan gum (Saddler and Bradbury 2005). Phytopathogenic species of Xanthomonas grow at an optimal temperature of 28°C and produce highly pigmented yellow mucoid and butyrous colonies which differentiates them from non-phytopathogenic species (Saddler and Bradbury

2005).

Hosts and Symptomology

Tomato (Solanum lycopersicum) and pepper (Capsicum annuum) are the two main hosts within the Solanaceae family susceptible to Xanthomonas. Numerous Capsicum spp. such as C. anomallum, C. baccatum, C. chacoense, C. chinense, C. frutescens, C. galapagoense, and C. pubescens are susceptible to Xanthomonas (Sahin and Miller, 1998). Other cultivated plants that are susceptible to Xanthomonas include chili pepper (Capsicum rutescens) and cherry tomato (S. lycopersicum var. cerasiforme) (Jones et al., 1998b; Potnis et al., 2015). However, the pathogenicity of Xanthomonas spp. that cause BST on the aforementioned plants is uncertain and requires further studies (Jones et al., 1998b). Weeds, occurring in close proximity to tomato cultivation, play a minor role in Xanthomonas survival (Jones et al., 1986). Nevertheless,

Xanthomonas was recovered from several solanaceous (Datura stramonium, Hyoscyamus niger,

H. aureus, N. physaloides, S. americanum, S. dulcamara, S. rostratum, Physalis pubescens) and non-solanaceous (Ambrosia artemisifolia, Eclipta alba, Eupatorium capillifolium, Euphorbia heterophylla, and Trifolium repens) weeds (Araújo et al., 2015; Jones et al., 1986; Jones et al.,

1998b).

Symptoms of bacterial spot on both tomato and pepper occur on aboveground plant parts such as foliage, stems and fruits (Jones and Miller, 2014). Foliar and stem lesions start as irregular green water-soaked lesions up to 3 mm in diameter. Water-soaked lesions are easily seen when leaves are wet. Lesions become dry, dark and necrotic and produce a faint halo (Jones

13 and Miller, 2014). At later stages of symptom development, lesions coalesce and plants become blighted and defoliated. Defoliation is less severe on tomato compared to pepper. Xanthomonas perforans can produce a shot-hole appearance on leaves (Jones and Miller 2014). Some strains of

Xp can also induce pith necrosis, vascular discoloration, longitudinal splits and external lesions on tomato plants (Aiello et al., 2013). Not only can blighting and defoliation directly reduce fruit yield, but can indirectly lead to additional losses due to the exposure of fruit to the elements resulting in sun scalding and rain check. BST can also directly infect fruit. Fruit lesions are initially small and as symptoms progress the lesions become dark, scab-like and often with a dark green to yellow halo around the lesion. Xp does not appear to be highly associated with fruit lesions, unlike X. euvesicatoria and X. vesicatoria (Potnis et al., 2015). These lesions typically render the fruit unmarketable and make the fruit prone to infection by opportunistic pathogens that cause postharvest decays.

Distribution and Epidemiology

Bacterial spot is a widely distributed disease (Anonymous, 1992; Potnis et al., 2015).

Historically, X. euvesicatoria and X. vesicatoria occurred more commonly than Xp and X. gardneri (Jones et al., 2005; Potnis et al., 2015). However, recent studies showed an increasing spread and recurrent outbreaks of X. gardneri into the eastern United States (Pennsylvania, Ohio,

Michigan), Canada (Ontario), Europe (Bulgaria, Russia) and Asia (Malaysia) (Abbasi et al.,

2015; Kim et al., 2010b; Kizheva et al., 2011; Kornev et al., 2009; Ma et al., 2011; Rashid et al.,

2016; Timilsina et al., 2015). Likewise, Xp is reported increasingly from different areas worldwide, such as United States (Florida), Canada (Ontario), Brazil, Mexico, Iran, Thailand,

Southwest Indian Ocean (Abbasi et al., 2015; Araújo et al., 2017; Hamza et al., 2010; Jones et al., 2005; Osdaghi et al., 2016). Long-distance movement of Xanthomonas spp. is commonly

14 associated with movement of infested seed and transplants (Gitaitis et al., 1992; Kebede et al.,

2014; Potnis et al., 2015).

BST infections are favored under high relative humidity and optimal temperatures ranging from 25 to 28°C (Obradovic et al., 2008). BST can be primarily introduced into a field through infested seed and diseased tomato seedlings and to a lesser extent by volunteer tomato plants and fields in proximity to infested fields (Jones et al., 1986). Bacteria spreads in the field by wind-driven rain, aerosols and by cultural practices (e.g. clipping, tying, spraying and harvesting) (Bernal and Berger, 1996; McInnes et al., 1988; Pohronezny et al., 1990). Up until the 1990’s tomato transplants were produced in the field and BST outbreaks were common along with other bacterial diseases in the southeast (Gitaitis et al., 1992). Severe disease outbreaks caused the industry to abandon field production and move to growing seedlings under greenhouses in flats (Gitaitis et al., 1992). The dispersal of Xanthomonas euvesicatoria in a pepper field reached up to 32 m after a 2-day rain period (Bernal and Berger, 1996). The distribution pattern initially showed a gradient which quickly plateaued and reached near 100% incidence across the plots (Bernal and Berger, 1996). Also, BST readily spreads inside transplant houses due to high density planting and frequent overhead irrigation practices (Potnis et al.,

2015).

A few studies have elucidated the movement of bacterial pathogens under greenhouse or field conditions, such as Clavibacter michiganensis subsp. michiganensis (Cmm) and

Pseudomonas syringae pv. tomato. Cmm was recovered at a high concentration close to an infected point source and increased over time in tomato pots (Chang et al., 1992). Spraying tomato seedlings in flats with acibenzolar S-methyl, copper hydroxide alone or combined with streptomycin or mancozeb reduced Cmm populations from 108 to 106 CFU/g in plants adjacent to

15 infected plants (Werner et al., 2012). Field populations of Cmm were higher than greenhouse populations and can reach more than 109 CFU/g which is associated with higher disease severity

(Chang et al., 1992; Wener et al., 2012). Further, when workers touching Cmm-infected tomato with guttation, Cmm was able to move a long distance reaching up to 22 plants within the same row (Sharabani et al., 2013). On the other hand, Cuppels et al. (1999) tracked natural populations of P. syringae pv. tomato using a semi-selective media from transplant seedlings in greenhouses to the field. Low bacterial levels and asymptomatic tomato transplant seedlings did not correlate with disease severity in the field plants (Cuppels et al., 1999). P. syringae pv. tomato populations were affected by environmental conditions more than transplanting infested seedlings into a field

(Cuppels et al., 1999).

Xanthomonas spp. can survive as epiphytes on tomato leaves (Jones et al., 1986;

McGuire et al., 1991). Epiphytic bacterial populations colonize leaf surfaces prior to infection under conducive conditions (Hirano and Upper, 1983). Epiphytic populations of X. euvesicatoria and disease severity showed to be positively correlated (McGuire et al., 1991; Jones et al., 1991).

X. euvesicatoria invades the leaf surface through natural openings, such as stomata, lenticels, trichomes and hydathodes, and through wounds (Obradovic et al., 2008; Pohronezny et al.,

1992). Further studies, using a green fluorescent protein (GFP)-labeled X. euvesicatoria strain showed bacterial colonization surrounding the stomata and in depressions between epidermal cells (Zhang et al., 2009). X. euvesicatoria was able to colonize tomato and non-host seeds after inoculating floral parts, indicating possible seed transmission from flowers to fruits and then seeds (Dutta et al., 2014).

Phenotypic and Genetic Diversity

BST was simultaneously observed in the early 1900s in Africa and the United States (Jones et al., 1998b). The causal agent was initially identified as Bacterium vesicatorium and B. exitiosa.

These two different species showed different amylolytic activity but were later combined into one species as B. vesicatorium. The genus name changed several times from Bacterium to

Pseudomonas to Phytomonas and finally to Xanthomonas (Jones et al., 1998b). BST strains of

Pseudomonas gardneri were initially differentiated based on biochemical and pathogenicity assays and not until the 1990’s molecular techniques were used for characterization (Vauterin et al.,

1990).

BST strains were later renamed and classified into a single species and pathovar and designated as X. campestris pv. vesicatoria (Xcv) (Dye et al., 1980). Xcv was later classified into two groups (A and B) based on amylolytic and pectolytic activity, serology, carbon utilization, protein profile, fatty-acid profile, race differentials and through DNA-DNA relatedness (Stall et al., 1994). Group A strains were non-pectolytic, non-amylolytic and incited a hypersensitive response on the tomato breeding line ‘Hawaii 7998’ and shared less than 46 % homology based on DNA-DNA hybridization with group B strains (Stall et al., 1994). Further analysis placed some Xcv strains into sub-group A2, within group A, based on amylolytic and pectolytic activity

(Jones et al., 1998b; 2000). However, strains collected elsewhere (e.g. Iran, Grenada, Brazil) did not exhibit similar pectolytic and amylolytic activity indicating a high phenotypic diversity in

Xcv (Areas et al., 2014; Hamza et al., 2010; Osdaghi et al., 2016). Group B strains were pectolytic, amylolytic and did not produce an HR on ‘Hawaii 7998’ (Stall et al., 1994). Groups A and B were reclassified as X. axonopodis pv. vesicatoria and X. vesciatoria, respectively

(Vauterin et al., 1995). In the early 1990’s, several strains collected from Florida and Mexico

17 were amylolytic and pectolytic, similar to group B, but did not react to monoclonal antibodies specific to groups A or B, and were pathogenic on ‘Hawaii 7998’; these strains were later designated as group C (Bouzar et al., 1994, 1996; Jones et al., 1995). In addition, distinct BST strains collected from Yugoslavia and Costa Rica were non-pectolytic, non-amylolytic (similar to group A), and caused similar symptoms to groups A and B, but had a very low re-association value to group A and B strains in DNA-DNA hybridization tests; thus, were placed in group D

(Jones et al., 2000). Due to the wide diversity of strains and confusion from classification based on biochemical, serological and pathogenic assays, a classification was proposed based on DNA-

DNA hybridization that assigned BST strains to four distinct species, Xanthomonas euvesicatoria (syn. X. campestris pv. vesicatoria; X. axonopodis pv. vesicatoria; Group A), X. vesicatoria (Group B), Xp (syn. X. campestris pv. vesicatoria; Group C), and X. gardneri (Group

D) (Jones et al., 2004). Recently, a polyphasic taxonomic study based on whole-genome sequencing, DNA-DNA hybridization, multi-locus sequence analysis (MLSA) and biochemical analysis, suggested combining X. euvesicatoria and Xp into a single species with two pathovars

X. euvesicatoria pv. euvesicatoria and pv. perforans (Constantin et al., 2016). X. euvesicatoria and Xp exhibited an average nucleotide identity (ANI) of 98.5%; higher than the suggested cut- off value of 95% (Constantin et al., 2016; Goris et al., 2007). Furthermore, Barak et al., (2016) suggested the consolidation of X. euvesicatoria, Xp, X. axonopodis pv. allii, X. alfalfa subsp. citrumelonis and X. dieffenbachiae into one species, X. euvesicatoria.

BST strains can also be categorized into four tomato races (T1, T2, T3, T4) and 11 pepper races (P0-P10) based on their reaction to tomato and pepper differentials (Stall et al.,

2009). These differentials express specific resistance genes that interact in a gene-for-gene manner with avirulence proteins produced by BST strains (Astua-Monge et al., 2000; Minsavage

18 et al., 1990). Most X. euvesicatoria strains are identified as race T1 and elicit an HR on ‘Hawaii

7998’ (Jones and Scott, 1986). Both X. vesicatoria and X. gardneri belong to race T2 and do not cause HR on any known tomato differentials to date (Jones et al., 1995). On the other hand, Xp is composed of two races T3 and T4. Races T3 and T4 cause an HR on Lycopersicon pennellii

‘LA716,’ but only T3 causes an HR on ‘Hawaii 7981’ (Astua-Monge et al., 2000; Jones et al.,

1995). Until the early 1990s, X. euvesicatoria was the prevalent xanthomonad associated with

BST in Florida (Jones and Scott, 1986; Jones et al., 1995). However, X. euvesicatoria race T1 was gradually replaced with the introduction of Xp that represented a new race, T3 (Jones et al.,

1995; Jones et al., 1998a). Xp was similarly isolated from tomato growing areas in Brazil and

Iran at a higher rate than X. euvesicatoria (Araújo et al., 2017; Osdaghi et al., 2017). Laboratory and field studies showed the competitive nature of the T3 strains over T1 by a 10-fold reduction in T1 bacterial population (Jones et al., 1998a). The displacement of X. euvesicatoria T1 strains was linked to the strong antagonistic activity of bacteriocins produced by Xp (Tudor-Nelson et al., 2003). Further characterization of bacteriocins in T3 strains showed production of three different bacteriocins encoded by BcnA, BcnB and BcnC genes (Tudor-Nelson et al., 2003).

Xp T3 became the prevalent race present throughout Florida until Xp race T4 emerged through mutations of the avrXv3 gene (Horvath et al., 2012). Recently, avrXv3 was characterized in several T4 strains and three types of mutations were characterized, genes which contained insertions, early stop codons and pseudogenes (Timilsina et al., 2016). Transposable elements of about ~800 bp in the avrXv3 locus have been recovered elsewhere indicative of a possible horizontal gene transfer from X. gardneri (Araújo et al., 2017). Xp race T3 carries the avrXv3 and avrXv4 genes and trigger a hypersensitive response (HR) on tomato plants, H7981 (or

FL216) and LA716 with the Xv3 and Xv4 resistance loci, respectively (Astua-Monge et al., 2000;

Stall et al., 2009; Wang et al., 2011). Furthermore, surveys showed a tendency towards race displacement of T3 by T4 (Horvath et al., 2012; Timilsina et al., 2016). A survey conducted in

2006 showed that 70% of the strains were T4 while the remaining belonged to the T3 race

(Horvath et al., 2012). More recently, another survey of 175 Xp isolates collected from three major tomato production areas in Florida showed that 100% of the isolates were the T4 race

(Timilsina et al., 2013). Further, Xp race T4 have been reported for the first time in Louisiana

(Lewis Ivey et al., 2016). A recent survey in Brazil, showed that 97.5 % of the strains recovered belong to Xp race T3 (Araújo et al., 2017). However, race T4 strains were reported for the first time in Brazil at a very low incidence (Araújo et al., 2017). Also, one race T4 strain was recovered in another recent survey in Ethiopia (Kebede et al., 2014). Of the four Xanthomonas spp., only Xp is incapable of infecting pepper. However, some atypical strains of Xp are able to infect and cause disease in pepper (Schwartz et al., 2015). Currently, Xp race T4 strains are the prevalent cause of BST in Florida; the other three species were not recovered from tomato in recent surveys (Horvath et al., 2012; Timilsina et al., 2013).

Several molecular tools are commonly used to differentiate between bacterial strains

(Miaden et al., 2013). Several phylogenetic and population studies have shown that

Xanthomonas populations can be diverse and are capable of recombination (Araújo et al., 2017;

Huang et al., 2015; Timilsina et al., 2015). Population genetic studies mostly focus on characterizing specific genes such as housekeeping genes, target effectors and pathogenicity genes (Albuquerque et al., 2012; Hamza et al., 2010; Kebede et al., 2014; Osdaghi et al., 2016;

Timilsina et al., 2015, 2016). A MLSA-based study based on using several housekeeping genes such as lacF, lepA, gyrB, fusA, gltA and gapA and virulence hrpB genes showed high diversity between strains globally and indicated potential recombination between X. euvesicatoria and Xp

(Timilsina et al., 2015). Based on this MLSA, Xp, X. euvesicatoria, and X. vesicatoria were subdivided into two, two and three phylogroups, respectively (Timilsina et al., 2015). However, a MLSA study showed very limited genetic diversity among strains in Ethiopia (Kebede et al.,

2014). Whereas, a global collection of X. gardneri strains showed a uniform haplotype. Xp group

1 is closely related to the strain Xp ‘91-118’. Xp ‘91-118’ is the first race T3 strain isolated in

1991, whereas the second group is closely related to X. euvesicatoria (Timilsina et al., 2015).

Some atypical Xp strains have been isolated which are phylogenetically and phenotypically unique (Kebede et al., 2014; Timilsina et al., 2015). These strains are capable of infecting pepper and have somewhat divergent genomes, yet they are identified as Xp based on phylogenetic analysis (Schwartz et al., 2015). A comprehensive phylogenomic study based on draft genome sequences confirmed MLSA-based phylogenetic relationships (Schwartz et al., 2015). The study sequenced 67 strains of X. euvesicatoria, Xp and X. gardneri and identified 1,152 core genes that were common among all strains (Schwartz et al., 2015). Phylogenetic analysis based on a core gene concatenation provided higher resolution between clades and separated group 1 into two subgroups ‘A’ and ‘B’. Furthermore, a phylogenetic tree of all strains based on 225,284 single nucleotide polymorphisms (SNPs) compared to a reference strain, Xanthomonas axonopodis pv. citri ‘306’, confirmed core genome-based phylogenetic analysis (Schwartz et al., 2015). SNPs are favored for genome-wide analysis of organisms due to their low mutation rates (10-8-10-9), evolutionary stability and their unbiased analysis of coding and non-coding sequence in large genomic regions (Brumfield et al., 2003; Pearson et al., 2004, 2009). SNPs can also be used to track evolutionary change in a population (Brumfield et al., 2003). However, the drawback in conducting SNP-based analysis is the need of an ancestral strain for an unbiased representation of clonal populations (Pearson et al., 2009). Effector-based phylogenetic analysis and typing for

21 comparing bacterial strains are incongruent with whole-genome based analysis (Barak et al.,

2016; Schwartz et al., 2015). More recently, another approach is gaining interest for analyzing bacterial populations which is by analyzing the pan-genome. The pan-genome consists of a repertoire of genes in a group of organisms, and can be open- or closed- pan-genomes (Rouli et al., 2015; Tettelin et al., 2005). Closed pan-genomes usually do not change when new sequenced strains are added. However, in open pan-genomes, the number of genes is increasing in correlation with an increasing number of sequenced strains (Rouli et al., 2015). In contrast to the core-genome, the pan-genome increases by an average of 33 genes for every additional sequenced strain in Streptococcus agalactiae (Tettelin et al., 2005).

Pathogenicity and Effectors

Most phytopathogenic bacteria use the type III secretion system (T3SS) for delivery of effector proteins into plant cells (Galán and Collmer, 1999). The T3SS apparatus is encoded by a

25+ kb cluster of conserved hypersensitive response and pathogenicity (hrp) genes and is particularly important for many Xanthomonas spp. (Boch and Bonas 2010; Galán and Collmer,

1999; White et al., 2009). The structure of the T3SS Hrp gene expression is essential for assembly of the T3SS apparatus and is controlled by a two-component regulatory system, mainly by hrpG and hrpX genes located outside the hrp gene cluster (Boch and Bonas 2010; Galán and

Collmer, 1999, Wengelnik et al., 1996; Wengelnik and Bonas, 1996). These effectors are chaperoned by binding to the hpaB protein and are injected into the cell by the injectisome (Boch and Bonas 2010). T3SS effectors elicit an effector-triggered immunity (ETI) that typically leads to HR in plants with a corresponding resistance (R) gene (Bent and Mackey, 2007). However, in the absence of a corresponding R gene, bacterial colonization and infection occurs (Gabriel,

1999). Xanthomonas spp. contain at least 45 characterized and candidate effectors named

Xanthomonas outer protein (Xop), which are typically common across all Xanthomonas spp.

(Potnis et al., 2011; White et al., 2009). Many effectors are identified through classical gene-for- gene experiments and are eventually renamed as Avr genes when avirulence is detected in the presence of an R gene (Bent and Mackey, 2007; Gabriel, 1999). However, the recent use of high- throughput genome sequencing facilitated the identification of a large number of effectors

(White et al., 2009; Potnis et al., 2011; Schwartz et al., 2015). Nevertheless, several of the newly identified effectors are not extensively characterized (White et al., 2009). Most characterized effectors to date were studied in X. euvesicatoria in tomato or pepper plants by utilizing the avrBs2- or avrBs3-fusion reporter system, cDNA-amplified fragment length polymorphism

(AFLP) analysis or biochemically through the calmodulin-dependent adenylate cyclase domain of Bordetella pertussis (Casper-Lindley et al., 2002; Noël et al., 2001; Roden et al., 2004; Römer et al., 2007).

Xanthomonas effectors belong to one of three described effector families or are unassigned (Büttner and Bonas, 2010). The AvrBs3-effector group comprises the largest effector family of which the genes possess a transcription-activation like (TAL) domain (Boch and

Bonas, 2010; White et al., 2009). XopJ is another major effector subgroup belonging to the

Yersinia outer protein J (YopJ) superfamily (Lewis et al., 2011; Noël et al., 2001). The YopJ superfamily is one of the few conserved effector families that encompasses several plant pathogenic bacteria (e.g. Erwinia, Pseudomonas, Ralstonia), animal pathogenic bacteria (e.g.

Yersinia) and plant symbionts (e.g. Rhizobium) (Lewis et al., 2011; Noël et al., 2001; White et al., 2009). YopJ members share and depend on a conserved cysteine protease catalytic domain for their function (Gürlebeck et al., 2006; Lewis et al., 2011; Orth et al., 2000; White et al.,

2009). The YopJ superfamily belongs to the C55 cysteine protease family (Ma and Ma et al.,

2016; Orth et al., 2000; White et al., 2009). The YopJ superfamily is divided into five phylogenetic clades, of which group I is strictly limited to animal bacterial pathogens, and the remaining four include plant pathogens (Ma and Ma, 2016). The XopJ group contains several characterized effectors such as XopJ, AvrBsT (XopJ2), AvrRxv (XopJ3), AvrXv4 (XopJ4) and

AvrXccB (XopJ5) (Büttner and Bonas, 2010; Lewis et al., 2011; Ma and Ma, 2016; Noël et al.,

2001; Orth et al., 2000; White et al., 2009). The hallmark function of this superfamily is the effector’s ability to suppress host immunity by autoacetylation (AvrBsT), protease activity

(AvrBsT, XopJ), SUMO protease activity (AvrXv4), or ubiquitin protease activity through the cysteine catalytic domain (Lewis et al., 2011; Roden et al., 2004; Szczesny et al., 2010 Üstün and

Börnke, 2015). Xanthomonas YopJ-like effectors belong to group II and avrBsT is closely related to Aave2166 and HopZ2 from Acidovorax citrulli and Pseudomonas syringae, respectively (Lewis et al., 2011; Ma and Ma, 2016). Effectors of this family do not behave similarly even within the same species and have unique functions in planta (Lewis et al., 2011;

Szczesny et al. 2010). In plants, YopJ-like effectors cause localized cell death (Orth et al., 2000).

AvrBsT is a well characterized effector of Xanthomonas (Cheong et al., 2014; Minsavage et al., 1990; Orth et al., 2000; Kim et al., 2010a; Schwartz et al., 2015; Szczesny et al. 2010). avrBsT, a plasmid-borne gene, is not a core effector but is present in most strains of Xp and some strains of X. vesicatoria and X. euvesicatoria (Minsavage et al., 1990; Potnis et al., 2011;

Timilsina et al., 2016; Schwartz et al., 2015; Stall et al., 2009). Loss of the plasmid-borne avrBsT typically broadens pathogenicity from just tomato, to both pepper and tomato (Stall et al., 2009).

Studies showed that HR produced by some strains of X. euvesicatoria is related to the presence of AvrBsT interacting with the Bst resistance gene in Early Cal Wonder (ECW) peppers

(Minsavage et al., 1990). X. euvesicatoria, X. vesicatoria or Xp carrying avrBsT can cause an HR

24 in non-hosts such as Arabidopsis thaliana Pi-0 and Nicotiana benthamiana (Cunnac et al., 2007;

Escolar et al., 2001; Kim et al., 2010a; Orth et al., 2000; Schwartz et al., 2015). avrBsT deletion mutants in X. vesicatoria (strain BV5-4a) produced mild symptoms in pepper when infiltrated at

5x104 CFU.ml-1 (Kim et al., 2010a). However, deletion mutants in Xp did not consistently cause disease in pepper (Schwartz et al., 2015). Some Xp strains (race T4; group 2) were virulent in pepper when inoculated with a non-functional avrBsT (Schwartz et al., 2015). On the other hand, other strains (race T4; group 1) were not virulent, indicating host-range restriction is controlled by more than one factor in some strains (Schwartz et al., 2015). Furthermore, site-directed mutational analysis of the conserved catalytic triad (His, Glu, and Cys) in the avrBsT gene abolished cell death but not effector delivery into plant cells (Cunnac et al., 2007; Orth et al.,

2000). The conserved catalytic triad in AvrBsT is essential for HR activity and a single amino acid mutation renders the pathogen virulent on N. benthamiana and A. thaliana Pi-0 (Orth et al.,

2000). Kim et al. (2010) demonstrated the requirement of a full-length copy of avrBsT for HR- induction in pepper. avrBsT is constitutively expressed independently of the hrp regulatory genes unlike other effector genes (Escolar et al., 2001). AvrBsT interacts with other effectors and host proteins in planta (Kim and Hwang, 2015; Szczesny et al., 2010). Unlike other XopJ-like effectors, AvrBsT is capable of interfering with AvrBs1-induced HR, indicating a direct interference with effector-triggered immunity (ETI) through a possible interaction with cell signaling factors such as SNF1-related kinase 1 (Szczesny et al., 2010). AvrBsT has auto- and trans- acetyltransferase activity and acetylates the ACETYLATED INTERACTING PROTEIN1

(ACIP1) protein which is involved in ETI- and pathogen-associated molecular pattern –triggered

(PTI) immunity responses (Cheong et al., 2014). AvrBsT downregulates defense-related genes

(Senu4, Cevi16 and Tgas118) in tomato (Kim et al., 2010a). Whereas, HR in pepper is

25 accompanied with induction of defense-related genes (CaBPR1, CaPO2, CaSAR82A, and

CaDEF1), electrolyte leakage and hydrogen peroxide accumulation (Kim et al., 2010a).

It was previously thought that avrBsT is absent in Xp race T3 and in X. euvesicatoria.

However, a recent study showed that avrBsT is present in several T3 strains but not in the T3 reference strain (‘91-118’) and other T3 strains isolated in the early 1990’s (Potnis et al., 2011;

Timilsina et al., 2016). Gene sequencing showed that avrBsT between Xp strains and X. vesicatoria was identical (Timilsina et al., 2016). Furthermore, comparative genomic studies between the four reference genomes of Xanthomonas spp. revealed a diversity of effectors that are present in some but absent in others (Potnis et al., 2011; Schwartz et al., 2015). The hrp cluster was almost identical between Xp and X. euvesicatoria, but slightly differed from X. vesicatoria and X. gardneri (Potnis et al., 2011). Only 11 core effectors (AvrBs2, XopD, XopF1,

XopK, XopL, XopN, XopQ, XopR, XopX, XopZ1, XopAD) were common between the tomato- and pepper-infecting species (Potnis et al., 2011). However, most core effectors are highly conserved (e.g. AvrBs2, XopF1, XopL, XopK, and XopX) whereas some have multiple haplotypes (e.g. XopD and XopQ) within each species (Schwartz et al., 2015). Some effectors are only specific to one species or shared by two or three species, and some effectors exhibit host-specificity (Potnis et al., 2011). For instance, double deletion of avrBsT and xopQ resulted in an expanded host range of Xp group 2 strains from tomato to tomato and pepper but not in group 1 strains. This indicates the involvement of more than one factor in determining host range

(Schwartz et al., 2015). XopQ has been identified as a suppressor of ETI in tomato and pepper in

X. euvesicatoria (Teper et al., 2014). XopQ directly interacts with the tomato 14-3-3 isoform

(TFT4), a key player involved in signaling within the ETI pathway (Teper et al., 2014). Also,

AvrRxv, a homolog of AvrBsT, was shown to interact with 14-3-3 proteins in tomato (Whalen et

26 al., 2008). A large number of core and non-core effectors are characterized to date (Gürlebeck et al., 2006; White et al., 2009). For example, effectors XopF2, XopE2, XopAP, XopAE, XopH, and XopAJ interfered with PTI-triggered immunity induced by the bacterial peptide (flg22) in tomato plants (Popov et al., 2016).

Disease Control

BST control mainly relies on a combination of cultural practices and chemical sprays

(Gitaitis et al., 1992; Potnis et al., 2015). Due to the absence of a “silver bullet” management approach for bacterial diseases, sanitation plays an important role in reducing inoculum load in the field. Practices such as using pathogen-free seed and transplants, removing volunteer crops and weeds and the use of resistant cultivars are effective in reducing disease pressure (Bashan et al., 1982; Gitaitis et al., 1992; Jones et al., 1986). Resistant cultivars are critical for effective disease management and several R genes have been introgressed into tomato cultivars (Stall et al., 2009). Several sources of resistance are identified in tomato against Xanthomonas spp. such as Rx1, Rx2, Rx3, Xv3, RXv4, RXopJ4 and Bs4 (Stall et al., 2009). Nevertheless, none of these R genes are deployed in current commercial tomato cultivars, due to the continuous shifts in

Xanthomonas races and strains that have hampered resistance breeding efforts (Jones et al., 1995,

1998a; Kousik, C., & Ritchie, 1996). However, a genetically-modified tomato cultivar, which incorporated Bs2 from pepper, showed very high resistance against BST without protective chemical sprays (Horvath et al., 2012). A recent study which characterized several strains for candidate targets in tomato breeding programs suggested pyramiding three R genes, Bst, Bs2 and

Xv4, for targeting conserved bacterial genes, avrBsT, avrBs2 and avrXv4, respectively (Timilsina et al., 2016).

Antibiotics were initially used for controlling BST (Stall and Thayer, 1962).

Nevertheless, streptomycin resistance was observed earlier than copper products due to the higher selective pressure on bacterial population (Stall and Thayer, 1962). Although antibiotic resistance to streptomycin is an issue in Florida, it is not a problem in other parts of the world

(Bouzar et al., 1999; Osdaghi et al., 2017; Kebede et al., 2014). This is likely due to the fact that antibiotics were never incorporated into field spray programs in other parts of the world. Later, copper-compounds replaced streptomycin in Florida for disease management of BST. For many decades the tomato and pepper industry relied on the use of copper and copper-mancozeb tank mixes (Conover and Gerhold, 1981; Marco and Stall, 1983). The heavy use of copper and copper-based compounds resulted in an increasing level of copper tolerance among worldwide bacterial populations (Abbasi et al., 2015; Araújo et al., 2012; Marco and Stall, 1983; Martin et al., 2004; Mirik et al., 2007; Ritchie and Dittapongpitch, 1991; Stall et al., 1986). Recent surveys in Ontario showed a highly copper resistant Xp population linked to a race shift to T4 strains

(Abbasi et al., 2015). In vivo studies comparing the aggressiveness of copper tolerant and sensitive strains showed a higher virulence for the latter suggesting a fitness cost of copper tolerance or resistance (Araújo et al., 2012). Strains recovered elsewhere were copper sensitive, which is likely related to fewer copper sprays in those areas (Osdaghi et al., 2017; Kebede et al.,

2014). Therefore, copper and copper-mancozeb sprays are effective when the bacterial population is sensitive. Resistance to copper was associated with a plasmid in X. euvesicatoria

(Stall et al., 1986). Further studies showed the presence of five chromosome-encoded copper- associated gene clusters (Basim et al., 2005; Voloudakis et al., 2005). Chromosomal copper genes are present in Xp, X. euvesicatoria, X. vesicatoria and X. gardneri based on comparative genomic studies and are homologous to plasmid-borne genes (Potnis et al., 2011). Plasmid-borne

28 copper-resistance genes (cop) are present in X. vesicatoria and were necessary for copper resistance, whereas chromosome copper homeostasis genes (coh) were associated with copper tolerance (Potnis et al., 2011). copL, was identified upstream of the gene cluster and is a constitutively expressed copper operon regulator (Voloudakis et al., 2005). Three genes, copL, copA and copB, are major regulators of copper resistance (Basim et al., 2005; Behlau et al.,

2011; Voloudakis et al., 2005). The mechanism of copper resistance is likely due to binding of copper ions by CopA, CopB and other Cop-proteins in the periplasmic space (Cha and Cooksey,

1991). In 2006 and 2012 two separate surveys were carried out and showed that Xp strains collected from Florida exhibited 100% tolerance to copper compounds (Horvath et al., 2012;

Timilsina et al., 2013). Moreover, streptomycin resistance was only present in 5% of the strains in 2006 but increased to 32% in 2011 among the collected strains (Horvath et al., 2012;

Timilsina et al., 2013; Vallad et al., 2013). Unexpectedly, the frequency of resistance to streptomycin was higher for strains collected from transplant houses than strains from the field at

86% vs. 14%, respectively (Vallad et al., 2013). In Brazil, more than 90% of Xp strains were sensitive to streptomycin based on in vitro assays (Araújo et al., 2012). However, X. gardneri strains were highly resistant to streptomycin (Araújo et al., 2012).

Environmentally friendly alternatives to copper and streptomycin such as acibenzolar-S- methyl (ASM; Actigard®; Syngenta Crop Protection, Inc., Greensboro, NC) and bacteriophages are reported to reduce disease severity (Jones et al., 2007, 2012; Louws et al., 2001). ASM, a synthetic compound, induces systemic acquired resistance (SAR) against a broad range of pathogens (Vallad and Goodman, 2004). SAR is accompanied with an increase in salicylic acid and upregulation of pathogenesis-related genes (Durrant and Dong, 2004). ASM has been extensively tested as an alternative for copper-mancozeb sprays in tomato production (Huang et

29 al., 2012; Itako et al., 2014; Louws et al., 2001; Roberts et al., 2008; Vallad and Goodman,

2004). ASM showed significant disease reduction compared to copper-mancozeb sprays (Louws et al., 2001). Furthermore, weekly application of ASM was significantly better at reducing disease compared to biweekly applications (Huang et al., 2012). ASM is an effective alternative for copper where copper-tolerant strains are prevalent (Louws et al., 2001). Nevertheless, several studies showed that ASM did not improve yield (Huang et al., 2012; Louws et al., 2001;

Obradovic et al., 2004). Bacteriophages showed to be efficient in reducing BST (Obradovic et al., 2004). Single phage applications or in combination with ASM significantly reduced disease severity (Obradovic et al., 2004, 2005). However, formulation and harsh environmental conditions (ultra-violet rays, high temperatures) play a major role in phage survival on the leaf surface (Jones et al., 2007, 2012). Phage survival rapidly drops in correlation with the application time where the highest efficacy was observed at evening applications (Jones et al. 2007;

Obradovic et al., 2005). More recently, fungicides cymoxanil and famoxadone (Tanos®, E.I. du

Pont de Nemours and Company, Wilmington, DE) were evaluated for control of a copper- tolerant Xp strain (Fayette et al., 2012; Robert et al., 2008). Tank mixes of cymoxanil and famoxadone did not reduce disease or bacterial populations in comparison to non-treated tomato plants under greenhouse conditions (Fayette et al., 2012). However, combination of cymoxanil and famoxadone with copper hydroxide significantly reduced disease (Fayette et al., 2012).

This study was carried out to better understand whether population shifts of Xp occur and how this enhances or affects pathogen fitness and management strategies under different environmental conditions. Therefore, the main objectives of this study were: (1) to evaluate the biological significance of a recently introduced effector (AvrBsT) into the Xp population in

Florida brought by population shifts in the past decade, (2) to sequence whole-genomes and

30 characterize selected strains throughout the tomato production line in order to better understand bacterial outbreaks in the field and (3) to evaluate novel and alternative compounds for managing

BST during transplant and field production due to the highly prevalent copper-tolerant strains.

CHAPTER 2 THE TYPE III EFFECTOR AVRBST ENHANCES XANTHOMONAS PERFORANS FITNESS IN FIELD-GROWN TOMATO

Plant pathogenic bacteria have evolved diverse strategies to suppress plant defense responses (Galán and Collmer, 1999). As a result bacterial pathogenicity depends on the type III secretion system (TTS) (Gürlebeck et al., 2006). The TTS is encoded by a large consereved gene cluster termed hrp (Alfano and Collmer, 2004). The hrp gene cluster is essential for maintain pathogenicity (Alfano and Collmer, 2004). The TTS secretes effector proteins into plant cells that are mainly responsible for suppressing or mimicking host proteins (Gürlebeck et al., 2006).

In addition, plant pathogenic bacteria can acquire novel effectors mainly through horizontal gene transfer (Alfano and Collmer, 2004; Barak et al., 2016). At least 39 characterized or putative effectors are present in Xanthmonas spp. (White et al., 2009). The majority of effectors remain uncharacterized and a large number of effectors have redundant activity (Gürlebeck et al., 2006).

Bacterial spot of tomato (BST) is caused primarily by Xp (Potnis et al., 2015). BST leads to high yield losses in the field due to defoliation (Pohronezny et al., 1986). X. euvesicatoria was the causal organism of BST in Florida through the early 1990s, but was gradually replaced by the introduction of Xp race T3 strains (Jones et al., 1998). However, Xp, race T4, is the prevalent cause of BST and the remaining species were not recovered from tomato fields in Florida

(Horvath et al., 2012; Schwartz et al., 2015; Stall et al., 2009). Phylogenomic analysis of a number of strains from recent surveys revealed the presence of two groups (1 and 2) within the

T4 race structure (Timilsina et al., 2015; Schwartz et al., 2015). Based on a comparative genomic study, the majority of the strains acquired a new effector, avrBsT (Schwartz et al., 2015;

Timilsina et al., 2016). avrBsT was initially characterized from X. euvesicatoria in a gene-for- gene manner in Early Calwonder (ECW) pepper plants (Minsavage et al., 1990). These X.

32 euvesicatoria strains were pathogenic on tomato only and elicited a hypersensitive response on

ECW pepper plants carrying the Bst resistance locus (Minsavage et al., 1990). avrBsT was initially identified in the first race T4 strain isolated in 1998 from Florida (Timilsina et al., 2016).

Race T3 strains collected prior to 1998, did not carry avrBsT, but T3 strains isolated later did

(Timilsina et al., 2016). The presence of avrBsT in Xp limits its host range to tomato only

(Minsavage et al., 1990). In recent studies the majority of Xp race T4 strains were determined to carry avrBsT (Schwartz et al., 2015; Timilsina et al., 2016). Furthermore, avrBsT is present in X. vesicatoria strains but not in X. gardneri (Kim et al., 2010; Schwartz et al., 2015).

AvrBsT (syn. XopJ2) is a plasmid borne type III effector belonging to the XopJ clade within the YopJ superfamily of effectors that encompasses human and plant bacterial pathogens

(Ma and Ma, 2016). The hallmark feature of this effector group is the presence of a conserved cysteine protease catalytic triad domain for their function (Lewis et al., 2011; Orth et al., 2000).

The XopJ effector group contains several characterized effectors such as XopJ1, AvrBsT

(XopJ2), AvrRxv (XopJ3), AvrXv4 (XopJ4), AvrXccB (XopJ5) (Büttner and Bonas, 2010;

Lewis et al., 2011; Ma and Ma, 2016; Orth et al., 2000; White et al., 2009). Effectors of this family have diverse and unique cellular functions in planta even within the same species (Lewis et al., 2011; Szczesny et al. 2010). AvrBsT is a well characterized effector of Xanthomonas

(Cheong et al., 2014; Minsavage et al., 1990; Orth et al., 2000; Kim et al., 2010; Schwartz et al.,

2015; Szczesny et al. 2010). AvrBsT is not a core effector but has a conserved sequence and is commonly found in Xp and some strains of X. vesicatoria and X. euvesicatoria (Minsavage et al.,

1990; Potnis et al., 2011; Timilsina et al., 2016; Schwartz et al., 2015; Stall et al., 2009). X. euvesicatoria or X. vesicatoria or Xp carrying avrBsT can cause an HR in non-hosts such as

Arabidopsis thaliana Pi-0 and Nicotiana benthamiana (Cunnac et al., 2007; Escolar et al., 2001;

Kim et al., 2010; Orth et al., 2000; Schwartz et al., 2015). Deletio mutants of avrBsT in X. vesicatoria (strain BV5-4a) produced mild symptoms in pepper when infiltrated at 5x104

CFU/ml (Kim et al., 2010). However, avrBsT deletion mutants in Xp did not consistently cause disease in pepper (Schwartz et al., 2015). Xp race T4-group 2 strains were virulent on pepper when inoculated with a non-functional avrBsT (Schwartz et al., 2015). However, Xp race T4- group 1 strains were not virulent on pepper, indicating additional factors restricting host-range in these strains (Schwartz et al., 2015). Furthermore, site-directed mutational analysis of the conserved catalytic triad (His, Glu, and Cys) in the avrBsT gene abolished cell death but not effector delivery into plant cells (Cunnac et al., 2007; Orth et al., 2000). The conserved catalytic triad in AvrBsT is essential for HR and cellular function and a single amino acid mutation in the conserved catalytic triad renders the pathogen non-virulent on N. benthamiana and A. thaliana

Pi-0 (Orth et al., 2000). Kim et al. (2010) demonstrated the requirement of a full-length copy of avrBsT for inducing HR in pepper. avrBsT is a constitutively expressed effector independent of the hrp regulatory genes unlike most effector genes (Escolar et al., 2001).

AvrBsT interacts with a number of effectors and host proteins in planta (Han and Hwang,

2017; Szczesny et al., 2010). Unlike other XopJ-like effectors, AvrBsT is capable of interfering with AvrBs1-induced HR, indicating a direct interference with ETI through a possible interaction with cell signaling factors such as SNF1-related kinase 1 (Szczesny et al., 2010). AvrBsT has acetyltransferase activity and acetylates the acetylated interacting protein1 (ACIP1) protein which is involved in ETI- and PTI-triggered responses (Cheong et al., 2014). AvrBsT downregulates defense-related genes (Senu4, Cevi16 and Tgas118) in tomato (Kim et al., 2010), whereas, the HR in pepper is accompanied with induction of defense-related genes (CaBPR1,

CaPO2, CaSAR82A, and CaDEF1), electrolyte leakage and hydrogen peroxide accumulation

(Kim et al., 2010). AvrXccb, an AvrBsT-homolog, suppressed immunity and interacted with cellular proteins in Brassica plants (Liu et al., 2016).

Pathogen fitness is ideally measured as the ability of a pathogen to survive and reproduce on a particular host (Laine and Barrès, 2013; Leach et al., 2001). According to Vanderplank

(1968) pathogens carrying unnecessary virulence genes have a fitness cost or penalty.

Furthermore, mutations in avirulence genes occur to increase fitness in the presence of an R gene, but can incur a fitness cost if an R gene is not present (Vanderplank, 1968). Virulence genes can contribute in different ways to host-bacterial interaction, they can affect host range, colonization, and pathogenicity. Fitness in plant pathogenic bacteria can be measured by examining several factors such as the rate of aggressiveness, infection efficiency, population growth, and survival under adverse environmental conditions (Leach et al., 2001). Several avirulence genes have been studied for their contribution to pathogen fitness in Xanthmonas sp. causing bacterial spot (Kearney and Staskawicz, 1990; Leach et al., 2001). The highly conserved avirulence gene avrBs2 contributes to pathogenicity and its deletion negatively affects fitness by reducing in planta growth of X. euvesicatoria (Kearney and Staskawicz, 1990). On the contrary, avrBs3 does not reduce bacterial populations and does not reduce aggressiveness, but still contributes to pathogen fitness in the field (Wichmann and Bergelson, 2004). Deletion of more than one effector gene in an avrBs2 mutant strain resulted in an additive fitness cost (Wichmann and Bergelson, 2004).

Most studies focus on evaluating the effect of type III effectors on pathogen virulence under controlled environmental conditions. Therefore, a knowledge gap exists in corroborating greenhouse findings with actual field conditions. Furthermore, fitness studies examine pathogen fitness within the context of gene-by-gene interaction, yet the rate of discovering novel effector

35 genes resulted in a shortage of discovering corresponding R genes. As a result using such R-gene based approaches are inadequate or unavailable for evaluating specific and recently introduced effectors into a population. Nevertheless, the significance of possessing such effectors for pathogen virulence under field conditions remains crucial for disease management purposes.

Therefore, in this study we evaluated the effect of a type III effector, AvrBsT, under field conditions. We hypothesized that the ubiquitous presence of AvrBsT confers a selective advantage in the Florida Xp population and the loss of avrBst should lead to a reduced pathogen fitness. For this purpose we generated avrBsT deletion mutants in Xp strains representative of phylogenomic groups 1 and 2 and studied the effect of the avrBsT mutant strains on in planta growth, competition, movement and epiphytic survival in comparison to the wild-type strain in tomato under field conditions.

Materials and Methods Bacterial Strains, Plasmids and Plants Xp strains GEV872 and GEV1001 were collected during a survey in 2012 from commercial grower fields (Schwartz et al., 2015). Xp GEV872 and GEV1001 strains, were selected as representative strains of Xp phylogenomic groups 1 and 2 (Schwartz et al., 2015). Xp

GEV872 and GEV1001 were induced for rifampicin resistance by plating a 108 CFU/ml bacterial culture of each strain on tryptone soy agar media amended with 200 mg/ml of rifampicin. All Xp

GEV872 and Xp GEV1001 strains referred to hereafter are resistant to rifampicin. Plates were monitored over a period of ten days and resistant colonies were selected and verified for rifampicin resistance. Antibiotics were added to media at the following concentrations: ampicillin: 100 μg/ml; rifampicin: 100 μg/ml; kanamycin: 100 μg/ml; spectinomycin: 100 μg/ml; cycloheximide: 100 mg/ml. Bacterial strains, vectors and plasmids used and constructed in this study are described in Table 2-1. Xp strains were cultured on nutrient agar or yeast glycerol agar

36 at 28°C with or without amended antibiotics. Bacterial strains were cultured on nutrient agar at

28°C and 37°C for Xp and E. coli, respectively. Bacterial cultures used for inoculating plants were suspended in sterile tap water and the concentration was adjusted to 108 CFU/ml

(OD600=0.3) using a spectrophotometer.

Tomato cultivar ‘HM1823’ (HM Clause, Davis, CA) was used in field trials conducted at the Gulf Coast Research and Education Center. Tomato cultivars (cv.) ‘HM1823’ and ‘FL47’

(Seminis, St. Louis, MO) were used for in planta growth curve and co-infiltration assays. Both tomato cultivars do not have any tolerance to bacterial spot. Pepper (Capsicum annuum L.) cv.

Early Calwonder (ECW), carrying the Bst resistance locus, was used for determining hypersensitive reaction (HR).

Mutant Construction Strain Xp GEV872 was used as a template for amplifying the full length avrBsT with primers (872F: CAAACATGCCTTCGGTGA/872R: ATGTGCATAGTGCGGTGCAT). An amplicon of 1,858 bp, including 0.5kb flanking regions, was cloned into pGEM®-T Easy vector and sequenced using F20/R24 primers for sequence confirmation. A 44-nt internal gene deletion was introduced into p-GEM-T:avrBsT using internal primers with a HindIII unique restriction site (FBstDel: tatataagcttGGAGACATGGCGGTCATCCAAC; RBstDel: tatataagcttAGCCGGCCATAACCTTAATTTCG) to produce p-GEM-T:ΔavrBsT resulting in a

12 amino acid deletion and frameshift in the avrBsT ORF. The insert was sequenced, excised from the clone by digestion using restriction enzymes (ApaI/SpeI) and sub-cloned into the suicide vector pOK1 restricted with ApaI/XbaI to produce pOK1:ΔavrBsT. Transformants were made in E. coli λpir. pOK1:ΔavrBsT was introduced into Xp by triparental mating using the helper plasmid, pRK2013. The avrBsT deletion was introduced into Xp GEV872 and GEV1001

37 by two-step homologous recombination and confirmation of mutants was made by PCR and sequencing (Huguet et al. 1998). All strains were stored as a 20% glycerol stock at -80°C.

In planta Growth and Competition Assays Strains were grown in nutrient broth (NB; Difco Laboratories, Detroit, MI) for 18 h, harvested by centrifugation, and re-suspended in sterile tap water or 0.01 M MgSO4. Wild-type, mutant strains and water were infiltrated into pepper cv. ECW leaves at a concentration of 5x106 and 5x108 CFU/ml. Four pepper leaves were infiltrated with each strain and water control at both concentrations and symptoms of HR were visually evaluated one to three days after inoculation.

Pepper infiltration was conducted twice.

Bacterial suspensions of each strain were infiltrated separately in 4 or 6-week-old tomato

‘FL.47’ or ‘HM1823’ seedlings at a concentration of 5x106 CFU/ml. Three leaves in multiple plants were infiltrated with each strain (Xp GEV872:WT; Xp GEV872:ΔavrBsT; Xp

GEV1001:WT; Xp GEV1001:ΔavrBsT). Following inoculation, plants were incubated at 24°C to

28°C. Three tomato leaflets were collected at a 24 h interval starting at 0 h to 96 h after inoculation. Three leaf discs from inoculated leaves were collected for each infiltrated strain at each time point. Leaf discs (1 cm2) were individually macerated in 1 ml sterile tap water and a

10-fold serial dilution was plated on NA media amended with rifampicin. Plates were incubated at 28°C, and colonies were counted after 48 to 72 h. CFU values were log-transformed and calculated as CFU/cm2. In planta growth assays were repeated twice.

Furthermore, a co-mixture of Xp GEV872:WT+GEV872:ΔavrBsT or Xp

GEV1001:WT+GEV1001:ΔavrBsT was infiltrated into 4-week old tomato ‘HM1823’ seedlings at a 1:1 ratio and adjusted to 5x106 CFU/ml. Leaf discs (1 cm2) were collected at 1, 3, 5 and 1, 3,

5 and 7 days post inoculation (dpi) for the first and second assay, respectively. Two and three leaf discs were collected in the first and second assay, respectively. Leaf discs were processed as

38 described above and diluted as necessary. Plates were incubated for 48 h at 28°C. Eight single colonies from each leaf disc representative of the co-mixture were isolated, sub-streaked and stored for further testing (see colony testing below).

Field Trials

Three field trials were conducted during the 2015 (fall), 2016 (spring) and 2017 (spring) growing seasons at the Gulf Coast Research and Education Center in Balm, Florida. Beds were fumigated, fertilized, sprayed with pre-plant herbicide and covered with a white plastic mulch according to standard grower practices (Freeman et al., 2017). Gloves were changed between plots whenever touching plants was required to reduce cross contamination between plots. Plots were on 7.6 m long beds and at least 15 to 21 m buffer zones between plots. Plants were obtained from a commercial grower. Tomato ‘FL47’ and ‘HM1823’ were used in the fall and springs seasons, respectively. Seedlings were transplanted at a 0.6 m distance within plots. Each plot contained 15 plants and treatments were setup in a randomized complete block design. The center plant of each plot was inoculated at a concentration of 5x106 CFU/ml. Each plot was either inoculated with Xp GEV872 or GEV1001. During fall 2015, four leaves of the center plant in each plot were individually infiltrated with Xp GEV872:WT and GEV872:ΔavrBsT, or with

Xp GEV1001:WT and GEV1001:ΔavrBsT. In the spring 2016 and 2017 trials, center plants were dip-inoculated into a suspension containing a mixture of Xp GEV872:WT+GEV872:ΔavrBsT or

Xp GEV1001:WT+GEV1001:ΔavrBsT each adjusted to 5x106 CFU/ml. The suspension also contained 0.05% (v/v) Silwett® L77 (Helena, Collierville, TN). All trials included non-inoculated control plots.

Disease severity was evaluated weekly and simultaneously with sample collection according to the Horsfall-Barratt scale (Horsfall and Barratt, 1945). In fall 2015, only

39 symptomatic leaves were sampled every 7 days starting at 7 to 49 dpi. In spring 2016 and 2017, all leaves were sampled every 7 days starting at 7 to 70 dpi. Furthermore, non-inoculated control plants were also sampled randomly throughout the trials. Four leaflets (~2 g) were sampled from the new growth to avoid redundant sampling. Leaves were washed in an extraction buffer (0.01

® M MgSO4 + 0.2% Tween 20 (Sigma-Aldrich, St. Louis, MO), and a 1 ml aliquot was sampled and diluted 10- to 200-fold. Only during the fall 2015 trial, the suspension was streaked using a sterile loop on un-amended NA and rifampicin amended NA. During the spring 2016 and 2017 trials, a 50 µl aliquot was plated using the exponential slow deposition mode in an Autoplate

4000® Spiroplater (Spiral Biotech, Norwood, MA) on un-amended NA and rifampicin amended

NA. Plates were incubated for two days at 28°C. Bacterial concentrations were determined by counting colonies in the outer plate sectors in a grid and then dividing the number of colonies by the volume deposited in the sector and multiplied by the dilution factor for conversion into

CFU/ml. The minimum detection threshold in our field trials was at 50 CFU/ml (~log 1.6).

Plants with no CFU recovered were assigned the detection threshold for statistical analyses. Up to four single colonies, representative of each plant, were picked, sub-streaked and stored for further colony testing.

Colony Testing Colonies collected from rifampicin-amended plates of leaf washings were tested for avrBsT wild-type or mutant genotypes. DNA was extracted using the boiling method. Briefly, bacterial stocks of each single colony (total of ~6000) were freshly streaked on NA amended with rifampicin. Using the tip of a loop, a small amount of pure bacterial growth was transferred into a 1.5 ml microcentrifuge tube and re-suspended in 400 ml of sterile water. Tubes were boiled at 100°C for 15 min and then immediately placed on ice for 5 min followed by centrifugation at 10,000 rpm for 5 min. A 2 µl aliquot of the supernatant was used in PCR

40 reactions. Primers 872F/R were used for amplifying the whole gene with flanking regions (1,858 bp). PCR products were then analyzed using agarose gel electrophoresis to confirm amplicon presence. Amplicons were then digested overnight using HindIII restriction enzyme and analyzed by electrophoresis on a 1% agarose gel. Xp colonies with a disrupted avrBsT ORF yielded two equal sized bands (~900bp).

Statistical Analyses All statistical analysis were performed in SAS v. 9.2 (SAS Institute, Gary, NC).

Generalized linear models were employed using PROC GLIMMIX in SAS to determine the effect of treatments on response variables. Data were log transformed into a normal distribution prior to statistical analysis. Data were analyzed using a repeated measure model including a

SPLICE function to compare treatments at individual intervals and to identify any treatment, time or time*treatment effect. Data was back-transformed and standard errors of the means were determined.

Results In planta Activity and Leaf Infiltration Assays Wild-type strains of Xp GEV872 and GEV1001 elicited an HR starting at 36 to 48 h after inoculation (Figure 2-1). Infiltrated areas were completely necrotic by 72 h. Xp

GEV872:ΔavrBsT and Xp GEV1001:ΔavrBsT did not elicit an HR response in ECW pepper plants carrying the cognate Bst resistance gene.

In leaf-infiltration assays on tomato, wild-type strains and mutant strains of Xp GEV872 or Xp GEV1001 did not show significant differences between populations (Figure 2-2).

Populations initially started at around 3.4x104 CFU/cm2 and then increased by 96 h to 2.43x108 and 4.6x107 CFU/cm2 for Xp GEV872 and Xp GEV1001, respectively. The type III tests for fixed effects are included in the appendix (Table A-2).

Furthermore, the co-infiltration assays showed no significant differences between wild- type strains and mutant strains of Xp GEV872 in both trials and Xp GEV1001 in the first trial

(Figure 2-3). In the second trial, recovery of Xp GEV1001:WT was two- to three- fold higher

(P=0.0227) than Xp GEV1001:ΔavrBsT (Figure 2-3). The type III tests for fixed effects for Xp

GEV1001 are included in the appendix (Table A-3).

Disease Severity in Field Plants Large differences were observed in disease severity between field trials due to different environmental conditions. Rainfall was higher in fall than in spring trials in the early plant stages at a monthly average of 4 and 2.3 mm, respectively (Figure 2-4). In spring, rainfall dropped sharply and coincided with lower disease severity in both trials (Figure 2-4), whereas, in the fall trials, rainfall slightly dropped and then decreased sharply in the last two weeks of the trial with some sporadic rainfall. Also, average relative humidity and temperatures were higher in the fall season compared to spring at 87% and 75% and 25 and 23°C, respectively. Additional environmental conditions are supplemented in the appendix (Table A-1).

Center inoculated plants were heavily diseased and were slightly stunted compared to non-inoculated plants. Disease severity was not significantly different between Xp GEV872 and

Xp GEV1001 in all three field trials (Figure 2-4).However, there was a significant difference in disease severity at each time interval throughout the three trials. During fall 2015, disease severity levels across the 49-day duration ranged between 0.7 to 25 % and 0.5 to 27% for Xp

GEV872 and Xp GEV1001, respectively (Figure 2-4). Disease severity beyond the inoculated center plant was nearly absent during the first four weeks and slightly increased in week five to 6 and 8%. At 42 dpi, disease severity reached 19 and 17 % and increased to 25 and 27 % one week later for Xp GEV872 and Xp GEV1001, respectively (Figure 2-4). In spring 2016, disease severity levels across the 63-day duration ranged between 3 to 84% and 3 to 86% for Xp

GEV872 and Xp GEV1001, respectively (Figure 2-4). No disease was observed in the plots for the first three weeks except on inoculated plants. Severity levels dropped at 42 to 49 dpi due to dry weather and then sharply increased at 56 dpi after a heavy rain event and conducive weather conditions resumed. In spring 2017, disease severity initially increased to around 10% for both strains in the first two weeks and then dropped to 2 to 4% for the next six weeks. At 63 dpi, weather conditions became more conducive and disease severity rapidly increased to 55%

(Figure 2-4).

Bacterial Populations in the Field

Total bacterial populations of Xp GEV872 or GEV1001 across trial plots initially exhibited a slightly bell-shaped distribution early in the first few weeks, with the highest recovery of Xanthomonas occurring on plants within the center of the plot (Figure 2-5). During spring 2016, total CFU in the center inoculated plant reached 3.8x105 and 7.9x105 CFU/g tissue at 21 dpi, while neighboring plants had a high CFU concentration at 105 to 5.2x105 CFU/g tissue and 3.8x105 to 8.5x105 CFU/g tissue for Xp GEV872 and Xp GEV1001, respectively (Figure 2-5 and Figure 2-5). At 28 dpi, populations across plots reached an average of 106 CFU/g tissue for

Xp GEV872 and Xp GEV1001, however populations on center plants were higher at 1.7x106 and

1.2x106 CFU/g tissue, respectively. There were no significant differences at 28, 35, 49, 56, 63 dpi for Xp GEV872 when CFU levels reached a plateau. Also, there were no significant differences at 28, 35, 49, 56 dpi for Xp GEV1001. At 63 dpi, Xp GEV1001 CFU levels slightly dropped and were not significantly different than 21 dpi. However, CFU levels reached a peak at

49 dpi at an average of 2.2 x106 and 2.5x106 CFU/g tissue for Xp GEV872 and Xp GEV1001

(Figure 2-5 and Figure 2-6).

In spring 2017, CFU levels of the center plants were initially at 4.1x106 and 3.7x106

CFU/g tissue but decreased within 14 days to 6.8x103 and 50 CFU/g tissue for Xp GEV872 and

Xp GEV1001, respectively. Between 14 to 42 dpi, several plants had very low CFU levels and

CFU values below the limit of detection except for the center inoculated plants. At 21 and 28 dpi, Xp GEV1001 was below the limit of detection and Xp GEV872 was recovered at very low levels of 5x103 CFU/g tissue (Figure 2-5 and Figure 2-5). The recovery of Xp GEV1001 and Xp

GEV872 increased by 49 dpi but was not significantly different from the levels during the first two weeks. At 56 dpi CFU levels reached an average of 5.6x105 and 9.5x105 CFU/g tissue for Xp

GEV1001 and Xp GEV872, respectively. The recovery of both strains eventually plateaued after

56 dpi and were not significantly different from each other (Figure 2-5 and Figure 2-5). There was no significant interaction for bacterial populations between plant distance from center inoculated and time in spring 2016 or 2017 (Table A-4 and Table A-5). However, bacterial populations of Xp GEV872 and GEV1001 were significantly different in terms of distance from the inoculated center plant within plots.

Effect of avrBsT on Movement Xp GEV872 and Xp GEV1001 showed similar movement trends in each trial (Figure 2-

6). Movement was recorded as incidence of the WT or ΔavrBsT strain across the plot to either side of the center inoculated plant. The distance Xp GEV872:WT moved across the plot was not significantly different from Xp GEV872:ΔavrBsT. In fall 2015, Xp GEV872:WT and

GEV872:ΔavrBsT were initially similar (Figure 2-6). At 21 to 28 dpi, Xp GEV872:ΔavrBsT surpassed GEV872:WT in distance but was not statistically different. At 43 dpi, Xp GEV872:WT spread only 1.2 times farther than Xp GEV872:ΔavrBsT from the center inoculated plant. Xp

GEV1001:WT significantly (P=0.0004) spread further than Xp GEV1001:ΔavrBsT across all time points (Figure 2-6). Between 7 and 28 dpi, Xp GEV1001:ΔavrBsT did not readily spread

44 across the plots. At 36 dpi, Xp GEV1001:ΔavrBsT strain was almost half-way across the plot from the center plant, whereas GEV1001:WT was 1.7 times further along the plot (Figure 2-6).

In spring 2016, Xp GEV1001:WT moved further across the plot compared to Xp

GEV1001:ΔavrBsT with significant differences at 21 and 42 dpi. Xp GEV872:WT spread more than GEV872:ΔavrBsT at each time period except 21 dpi, but there were no significant differences between strains over time. In spring 2017, both Xp GEV872:WT and GEV1001:WT spread significantly (P=0.0048 and P=0.0085, respectively) further than both ΔavrBsT strains.

At 7 dpi, GEV872:WT and GEV1001:WT spread at 4 and 1.1 times more than

GEV872:ΔavrBsT and GEV1001: ΔavrBsT, respectively. After 42 dpi, Xp GEV872:WT and

GEV1001:WT were more widespread than the ΔavrBsT strains. The ΔavrBsT strains were not able to surpass the spread of the wild-type strains at any time period in the spring 2016 and 2017 trials but only at one time period during fall 2015 trial. There was a significant difference across time but there was no interaction between time and treatments in all trials. The type III tests for fixed effects are included in the appendix (Table A-6 and Table A-7).

Effect of avrBsT on Strain Recovery More than 6,000 rifampicin resistant colonies were tested across all three field trials.

Rifampicin-resistant colonies were recovered from the leaf surface of symptomatic or non- symptomatic leaves. In all three trials, wild-type strains were generally recovered at a higher frequency compared to ΔavrBsT strains. In fall 2015, Xp GEV872:WT treatment was significantly different from Xp GEV872:ΔavrBsT at 42 (P=0.0052) and 49 (P=0.0219) dpi but not at 14 to 35 dpi (Figure 2-7). Xp GEV1001:WT treatment was significantly (P=0.0004) different from Xp GEV1001:ΔavrBsT.

In spring 2016, there was a significant difference (P=0.0001) between WT and ΔavrBsT strains of each Xp GEV872 and Xp GEV1001. At 21 dpi, there was only a 1.1 fold difference in

Xp GEV872:WT compared to Xp GEV872:ΔavrBsT (Figure 2-7). However after 28 dpi the difference ranged between 2 and 4.4 across all time points. A low number of colonies was recovered at 42 dpi coinciding with low total bacterial CFU (Figure 5). However at 21 dpi Xp

GEV1001:WT was 2.8-fold higher than Xp GEV1001:ΔavrBsT. Xp GEV1001:WT decreased slightly to a 1.5-fold difference and then increased to 3.1-fold difference by 63 dpi (Figure 2-7).

In spring 2017, during the first 7 weeks there was a low recovery rate of bacterial colonies (Figure 2-5) resulting in a low number of tested colonies for either wild-type or mutant strains. At 7 dpi there was no significant difference between treatments. However, at 14 dpi Xp

GEV1001:WT was about two-fold higher than the corresponding Xp GEV1001: ΔavrBsT. At 21 and 28 dpi, Xp GEV1001 was not recovered whereas Xp GEV872 colony recovery was low

(Figure 2-7). By 42 and 49 dpi, recovery of WT strains of each genotype was higher but not statistically different than the corresponding ΔavrBsT strains. WT strains of Xp GEV827 and Xp

GEV1001 significantly peaked at 56 dpi to about 5- and 7-fold compared to ΔavrBsT strains, respectively. A slight decrease followed at 63 and 70 dpi in recovering WT colonies however the difference with ΔavrBsT remained significant. In spring 2017, we recovered a very low number

(N=39) of rifampicin-resistant colonies from the five non-inoculated control plots at 70 dpi. The recovered colonies from non-inoculated plots were WT (N=37) and the remaining colonies were

ΔavrBsT strains. There was no interaction between time and treatments in all trials except for Xp

GEV1001 in spring 2017. The type III tests for fixed effects are included in the appendix (Table

A-8 and Table A-9).

Discussion Effector genes are an important component of host-pathogen interaction which variably contribute to pathogen fitness and virulence levels. Effector genes are secreted by the type III secretion system which is regulated by hrp genes. It is well documented that that deletion of hrp

46 genes of the type III secretion system in plant pathogenic gram-negative bacteria leads to a dramatic decrease or abolishment of infection (Gürlebeck et al., 2006). Further, several effectors were shown to contribute to pathogen fitness in planta or epiphytically under greenhouse conditions and very few under field conditions (reviewed in Leach et al., 2001; Hirano et al.,

1999; Vera Cruz et al., 2000; Wichmann and Bergelson, 2004). In this study, we examined for the first time the effect of avrBsT, an increasingly prevalent type III effector in the Xp population, on pathogen fitness on a large scale.

Xp mutant strains in this study were disrupted at the acetyltransferase catalytic domain of the gene resulting in the abolishment of two essential amino acids responsible for cell death, auto- and trans- acetylation (Han and Hwang, 2017). Therefore, mutant strains did not produce an HR in pepper plants indicating gene inactivation and disruption of cellular catalytic activity.

Population growth curves showed no differences between wild-type and mutant strains indicating that avrBsT appears to be a non-essential virulence factor and does not have a fitness effect on in planta population. Likewise, other studies showed that knocking out avrBs1 and avrBs3 in X. euvesicatoria do not alter in planta population but do have a fitness cost in the field

(Wichmann and Bergelson, 2004). However, this is in contrast to another study in which avrBsT deletion mutant strains of X. vesicatoria showed a lower population than the wild-type strain

(Kim et al., 2010). The effect of AvrBsT on fitness might be species or strain-specific but further studies are needed for confirmation. However, our results are in line with those described elsewhere for other XopJ-like effectors (Ciesiolka et al., 1999; Jiang et al., 2009; Noel et al.,

2003). Deletion mutants of avrRxv (Ciesiolka et al., 1999) and avrXccb (Jiang et al., 2009) in X. euvesicatoria and X. campestris pv. campestris and a frameshift deletion mutant of XopJ in X. euvesicatoria (Noel et al., 2003; Üstün et al., 2013) showed slight or no difference in bacterial

47 populations compared to wild-type strains. Furthermore, co-infiltration assay confirmed that there is no in planta competition between wild-type and mutant avrBsT strains. Evidence based on previous studies and those within this study suggest that members of the XopJ family do not appear to have a significant effect on bacterial in planta growth.

Environmental conditions play a role in influencing pathogen population dynamics in the field. Xanthomonas spp. can spread between plant beds and throughout production fields by wind-driven rain, irrigation water that is contaminated with the bacteria and by handling plants

(Pohronezny et al., 1990). Throughout the three trials prolonged dry periods reduced rates of pathogen movement throughout the field, however after periods of wetness disease severity increased. Since plants were co-inoculated simultaneously disease severity was not distinguishable between wild-type and mutant strains. Disease severity levels were also similar, across different time periods, between the group 1 and group 2 strains indicating a uniform level of disease development in the field. Although disease severity was low at specific time periods, bacterial population levels remained somewhat high. When epiphytic populations were examined in the field plants we observed a strong trend in which mutant strains of either genotype had lower recovery levels than wild-type strains. The reduction in epiphytic populations of strains lacking certain effectors is not unusual. For example, X. campestris pv. malvacearum mutant strains with a non-functional avrb6 did not reduce pathogen population in planta however pathogen egress to the leaf surface was reduced (Yang et al., 1994). Similarly, avrBsT mutant strains appear to be significantly less prevalent as an epiphyte and thus disperse shorter distances than wild-type strains. As a result of higher epiphytic wild-type populations and conducive environmental conditions wild-type strains were more readily disseminated across plants for both genotypes. Since, we did not find any differences in planta but only through epiphytic survival

48 we propose that environmental conditions might play a role in enhancing pathogen fitness. We also observed that throughout the growing season wild-type strains appeared to be more readily disseminated and recoverable than strains lacking avrBsT.

Currently, there is no commercially deployed R genes against avrBsT and therefore the presence of avrBsT should incur a fitness cost with no selective advantage in Xp. Nevertheless, evidence from this study showed that avrBsT provides a benefit to Xp with no associated cost.

We showed that deploying avrBsT mutant strains in the field incurred a fitness cost apart from the presence of an R gene. In this study we used two genotypes representative of two phylogenomic groups and observed that avrBsT disruption in both genotypes is detrimental to fitness. Our data suggest that avrBsT might have a strain-dependent fitness cost however this needs to be confirmed in more strains of the same group. Strain-dependent virulence has been shown for avrPto, an effector in Pseudomonas syringae pv. tomato (Shan et al., 2000). Further, it is worth noting that Xp group 2 strains might have a higher basal virulence compared to group 1 which might explain the higher prevalence of group 2 strains in Florida (P. Abrahamian, unpublished data). avrBsT is highly conserved and has no polymorphism based on recent comparative genomic studies in more than 100 strains (P. Abrahamian, unpublished data;

Timilsina et al., 2016; Schwartz et al., 2015). Although pepper has a Bst resistance locus which limits Xp host range from a fitness cost perspective the presence avrBsT should be dispensable but this is not the case for avrBsT. The presence of this effector for almost 20 years in the Xp population makes this locus a good target in resistance breeding efforts. Sources of resistance to avrBsT were found in almost all pepper lines and in Solanum lycopersicoides and a line of S. lycopersicum (Wang, 1992). Future work should focus on mapping the Bst locus in tomato and introgress the R gene into commercially grown tomato cultivars. Pyramiding resistance genes

49 against other conserved bacterial loci such as avrBs2 and avrXv4 is essential to reduce resistance breakdown based on single-locus resistance.

Table 2-1. List of bacterial strains and plasmids. Strain or plasmid Relevant characteristic(s) Source or referencea Xanthomonas perforans GEV872 Wild-type; RifR Schwartz et al., 2015; This study GEV872ΔavrBsT AvrBsT deletion mutant; RifR This study GEV1001 Wild-type ; RifR Schwartz et al., 2015; This study GEV1001 ΔavrBsT AvrBsT deletion mutant; RifR This study

E. coli DH5α F−recA BRL

λpir Host for pOK1; Spr UB

Plasmids pGEM-T AmpR Promega Inc. pOK1 Suicide vector; SacB Huguet et al., 1998 pRK2013 Helper plasmid; KanR Figurski and Helinski, 1979 a BRL: Bethesda Research Laboratories; UB: University of Berkely.

Figure 2-1. Infiltration of pepper cv. Early Calwonder with wild-type (WT) and mutant (Δ) avrBsT Xanthomonas perforans of two genotype GEV872 and GEV1001 for evaluating hypersensitive response (HR) (A). Field plots with at least 18 m distance between treatments, arrow shows inoculated center plant with either a mixture of GEV872:WT+GEV872:ΔavrBsT or GEV1001:WT+GEV1001:ΔavrBsT (B). (Photo courtesy of author)

Figure 2-2. Growth curve showing population dynamics after infiltrating tomato cv. FL. 47 with representative isolates of Xanthomonas perforans wild-type (WT) and mutant (ΔavrBsT) strains in the greenhouse. X. perforans GEV872:WT, GEV872:ΔavrBsT, GEV1001:WT, and GEV1001:ΔavrBsT were each infiltrated into separate leaves. Values at each point are average of three replicates.

Figure 2-3. Infiltration of Xanthomonas perforans wild-type (WT) + mutant (ΔavrBsT) co- mixtures for GEV872 and GEV1001 in tomato ‘HM1823’ leaves in the greenhouse. Average number of WT or ΔavrBsT colonies for two and three replicates in trials 1 (A) and 2 (B), respectively. Each replicate contained eight colonies. Standard error of the mean are indicated on bars.

Figure 2-4. Disease severity levels of bacterial spot caused by inoculated Xanthomonas perforans genotypes GEV872 and GEV1001 during fall 2015 (A), spring 2016 (B) and 2017 (C). Rainfall (mm) is presented across all dates across each trial. Rainfall peaks coincide with high disease severity. Standard error of the means are indicated on bars.

Figure 2-5. Total bacterial populations of rifampicin-resistant Xanthomonas perforans (Xp) GEV872 (A and C) and GEV1001 (B and D) recovered at different distances across the plot. Center plants are indicated as 0 cm and distance flanking center plants extend up to 4.3 m from the bed. Colony forming units (CFU) of spring 2016 (upper panel) and spring 2017 (lower panel).

Figure 2-6. Recovery of Xanthomonas perforans GEV872 and GEV1001 wild-type (WT) and mutant (ΔavrBsT) strains across plots during fall 2015 (A), spring 2016 (B) and spring 2017 (C). Standard error of means are indicated on bars.

Figure 2-7. Average number of colonies recovered for each Xanthomonas perforans wild-type (WT) and mutant (ΔavrBsT) strains during fall 2015 (A), spring 2016 (B) and spring 2017 (C). Standard error of means are indicated on bars.

CHAPTER 3 TRACING XANTHOMONAS PERFORANS POPULATIONS ON TOMATO FROM GREENHOUSE TO FIELD BY WHOLE-GENOME SEQUENCING

Bacterial spot of tomato (BST) is caused by Xanthomonas perforans (Jones et al., 2004).

In Florida, X. euvesicatoria (race T1) was the predominant species up until the mid-1990 (Jones and Scott, 1986; Jones et al., 1995). X. euvesicatoria was eventually displaced by Xp (race T3)

(Jones et al., 1995; Jones et al., 1998). In 1998, the first Xp race T4 was recovered and by 2006 more than 70% of the strains belonged to T4, whereas the remaining were T3 strains (Horvath et al., 2012; Timilsina et al., 2016). Race T3 strains carry the avrXv3 and avrXv4 genes that elicit a hypersensitive response on plants with the cognate R-genes (Stall et al., 2009). However, avrXv3 activity in race T4 strains is disrupted through the introduction of early stop codons, transposon insertions and pseudogenes (Stall et al., 2009; Timilsina et al., 2016). Recent field surveys showed a complete race displacement of T3 by T4 (Horvath et al., 2012; Potnis et al., 2015;

Timilsina et al., 2013). A survey conducted in 2006 showed that 77% of the strains were race T4, while the remaining were race T3 (Horvath et al., 2012). In 2011, all 175 Xp isolates collected from three major tomato production areas in Florida were race T4 strains (Timilsina et al., 2013).

Of the four Xanthomonas spp., Xp is typically incapable of infecting pepper. However, some atypical strains of Xp are able to infect and cause disease in pepper (Schwartz et al., 2015).

Currently, Xp race T4 is the only causal agent of BST in Florida; the other three species were not recovered from tomato in recent surveys (Horvath et al., 2012; Timilsina et al., 2013).

Several phylogenetic and population studies have shown that Xanthomonas populations can be diverse and capable of recombination (Araújo et al., 2017; Huang et al., 2015; Timilsina et al., 2015). Nevertheless, most population genetics studies focus on characterizing specific genes such as housekeeping genes, effectors and pathogenicity genes (Albuquerque et al., 2012;

Hamza et al., 2010; Kebede et al., 2014; Osdaghi et al., 2016; Timilsina et al., 2015, 2016).

Tools such as MLSA based on using several housekeeping genes such as lacF, lepA, gyrB, fusA, gltA and gapA and virulence hrpB genes are standard procedures in order to group microorganisms at the species level (Almeida et al., 2010; Constantin et al., 2016). Timilsina et al. (2015) showed high diversity between strains globally and indicated potential recombination between X. euvesicatoria and Xp. Based on MLSA, Xp, X. euvesicatoria, and X. vesicatoria were subdivided into two, two and three phylogroups, respectively (Timilsina et al., 2015). Xp group 1 is closely related to the reference strain Xp strain 91-118 which is a race T3 strain, initially isolated in 1991, whereas the second group contains some X. euvesicatoria alleles throughout their genome (Timilsina et al., 2015). However, another MLSA study showed very limited genetic diversity among strains in Ethiopia (Kebede et al., 2014). Some atypical Xp strains were found, phylogenetically and phenotypically unique (Kebede et al., 2014; Timilsina et al., 2015).

These strains are capable of infecting pepper and have somewhat divergent genomes, yet they are identified as Xp (Schwartz et al., 2015). More recently, a comprehensive phylogenomic study based on draft genome sequences confirmed MLSA-based phylogenetic obvervations (Schwartz et al., 2015). The study analyzed 67 strains of X. euvesicatoria, Xp and X. gardneri and identified

1152 core genes that were common among all strains (Schwartz et al., 2015). Phylogenetic analysis based on a core protein-coding gene concatenation provided higher resolution between clades and separated group 1 into two subgroups ‘A’ and ‘B’. Furthermore, a phylogenetic tree of all strains based on 225,284 single nucleotide polymorphisms (SNPs) compared to a reference strain, Xanthomonas axonopodis pv. citri ‘306’, confirmed core genome-based phylogenetic analysis (Schwartz et al., 2015). SNPs are favored for genome-wide analysis of organisms due to their low mutation rates (10-8-10-9), evolutionary stability and their unbiased analysis of coding

60 and non-coding sequence in large genomic regions (Brumfield et al., 2003; Pearson et al., 2004,

2009). SNPs can also be used to track evolutionary change in a population (Brumfield et al.,

2003). However, the drawback in conducting SNP-based analysis is the need of an ancestral strain for an unbiased representation of clonal populations (Pearson et al., 2009). Several studies showed that SNPs are a better discriminatory tool for detecting changes in bacterial populations

(Aritua et al., 2015; Schwartz et al., 2015). Effector-based typing and phylogenetic analysis are also used to show pathogen diversity; however, these tools are often incongruent with whole- genome based analysis (Barak et al., 2016; Schwartz et al., 2015). Nevertheless, mining genome sequences for effectors are still important for resistance breeding efforts (Timilsina et al., 2016).

In Florida, tomato seedlings are commonly used for establishing field plants. Seedlings are typically grown in Speedling flats in rudimentary high tunnel structures with adjustable open sides. Flats are laid out on steel rails and irrigated using an overhead irrigation system for water and chemigation. This system will be referred to as a transplant house and plants grown within it as tomato seedlings. Seedlings can become contaminated with bacterial spot or other diseases prior to transplanting, especially after heavy rains and high humidity (Gitaitis et al., 1992).

However, seedling manufacturers often rogue diseased seedlings, whether infected with bacterial spot or other diseases, to deliver healthy appearing plants to growers. BST outbreaks ultimately occur in the field throughout the growing season. However, the source of such outbreaks in the field is unknown. While most studies showed that BST is introduced mainly by wind-driven rain from neighboring fields, we hypothesized that BST outbreaks in the field are in fact primarily caused by bacterial populations originating from seedling production. Therefore, we utilized whole-genome sequencing data to analyze and track bacterial strains of Xp using MLSA, core-

61 genome and genome-wide SNP-based analysis, on seedlings from transplant houses into production fields.

Material and Methods Bacterial Strains and Race Characterization Tomato leaf samples showing bacterial spot symptoms were sampled during the 2015 and

2016 growing seasons. Samples were collected from two major transplant and field operations in central (Polk and Manatee County) and south Florida (Collier County) designated in this study as grower A and B, respectively. Initially, tomato seedlings were sampled during an outbreak in the transplant house. The seedling lot was marked, tracked and eventually transplanted into the field.

Later, these full-grown tomato plants were sampled for BST during the growing season, just prior to harvest. A number of strains were recovered from various cultivars in transplant houses from December 2015 and March 2016 from grower B. Field plants, originating from transplants sampled in March 2016, remained healthy in the field and we were not able to recover any field strains. Bacterial strains were isolated from single lesions on tomato leaves. Distance between transplant and field operations were about 30 miles and less than 5 miles for growers A and B, respectively. Bacteria were isolated on nutrient agar amended with cycloheximide (50 mg/ml) and single colonies were stored in 20% glycerol nutrient broth at -80°C. More than 300 strains were collected, from which 67 strains were randomly selected from both sites for further characterization. A list and description of all sequenced bacterial strains in this study are listed in

Table 3-1.

Bacterial strains were tested on tomato differentials to confirm pathogenicity and

8 determine race (Stall et al., 2009). Briefly, a 5x10 CFU/ml (A600 = 0.3) suspension of each strain was inoculated into tomato leaves of Bonnie Best, Hawaii 7998, FL216 and pepper leaves of

Early Calwonder (ECW. X. euvesicatoria 85-10 (race 1) and Xp 91-118 (race 3) were used as

62 positive control and water as a negative control. A hypersensitive reaction (HR) was recorded 24 to 48 hours after inoculation. Strains that produced a questionable HR reaction were re-tested.

Genome Sequencing and de novo Assembly Bacterial genome sequencing was carried out according to the Illumina manual (Illumina

Inc., San Diego, CA). Briefly, bacterial genomic DNA was isolated using a CTAB protocol and libraries were constructed using a Nextera library preparation kit (Illumina Inc., San Diego, CA).

Genomes were sequenced in two separate runs. In the first and second run, 20 and 47 samples were pooled into a single lane for sequencing, respectively. The genomes were sequenced using the Illumina MiSeq platform at the Interdisciplinary Center for Biotechnology Research,

University of Florida. Draft genomes were de novo assembled using CLC Genomics Workbench v5 (Qiagen, Hilden, Germany) with a minimum contig size of 500 bp. The assembled sequences were annotated using the IMG/JGI platform (Markowitz et al. 2012). The pairwise average nucleotide identity (ANI) based on BLAST was calculated using jSpecies v1.2.1 for all 67 strains

(Richter and Rosselló-Móra 2009).

Characterization with Multi-Locus Sequence Analysis A preliminary MLSA based on six housekeeping genes (fusA, gapA, gltA, gyrB, lacF, lepA) was conducted with the strains (Almeida et al., 2010). Sequences of each housekeeping gene were extracted from the 67 whole-genome sequences obtained in this study and from an additional 35 genomes sequenced in previous studies (Schwartz et al., 2015). X. euvesicatoria

85-10, Xp 91-118, X. gardneri ATCC19865 and X. vesicatoria ATCC35937 and other previously sequenced strains were included as a reference (see table B-1 for strain details). Sequences were concatenated and aligned using MUSCLE within MEGA 7 (Kumar et al., 2016).

Core-Genome Multi-Locus Sequence Typing The bacterial draft genome assemblies were annotated using the IMG-JGI annotation pipeline (Markowitz et al. 2012). The annotated sequences were used to identify the core genes among the 67 genomes using get_homologues v2.0.19 (Contreras-Moreira and Vinuesa, 2013).

Genes present in at least 95% of the input genomes with at least 70% pairwise alignment coverage were considered as core genes. The output core gene sequences were parsed using a python script to identify core genes that were present in 100% of the genomes that contained a start codon (Timilsina, 2016). Multiple copies of the same gene within a single genome were removed from the analysis. The sequences were then aligned by MAFFT alignment using a python script (Cock et al. 2009). A total of 886 genes were identified as core genes from all Xp strains that were concatenated to create a 0.8 Mb length sequence for each strain using Sequence matrix v.1.8 (Vaidya et al. 2011). Sequence matrix v.1.8 was used to calculate uncorrected pairwise distance for each gene with Xp91-118 as the reference. The pairwise distance was then used for sequence typing of all the core genes. Allele types of genes identical with genes from

Xp91-118 were labeled ‘0’ and the allele type for genes was assigned gradually based on the pairwise distance from Xp91-118. A color-coded heat map of all the polymorphic core genes within Xp strains was generated in R (https://www.rstudio.com/) based on the allele types.

Single-Nucleotide Variations Calling SNPs across all 67 genomes were identified by mapping the raw reads to the complete genome of Xp 91-118 (GenBank ID. CP019725.1) as a reference. Illumina raw reads were converted to Sanger format using FASTQ groomer (version 1.0.4). Paired-end raw sequence reads were aligned against the reference genome using Burrows-Wheeler Aligner (BWA)

(version 1.2.3) using the default parameters. From the alignment files we converted files into

BAM format in SAMTools as an input for Realigner Target Creator (v.0.0.4) for further local

64 realignment. In order to reduce SNP calling errors due to indel misalignments in our reads against the reference genome we used IndelRealigner (v.0.0.6) using the default settings. SNPs were called using UnifiedGenotyper (v.0.0.6). VCF files for each strain were merged into one file using VCFtools. Analysis were performed in Genome Analysis Toolkit (GATK) on the

Galaxy platform. SNP positions with a Phred score lower than 30 (99.9% accuracy) were discarded. Furthermore, we compared the output from GATK to SNPs called using Geneious v.

10.1.3 (https://www.geneious.com, Kearse et al., 2012). Illumina raw sequence reads were imported into Sanger format, trimmed and mapped to the complete reference genome of Xp 91-

118 using the Geneious mapper tool. SNPs caused by insertions and deletions were discarded from the consensus. SNPs were called based on a minimum 8x coverage, a minimum variant frequency occurring in 80% of mapped reads and a Phred score greater than 50 (99.999% accuracy) per position. After manually stripping gaps, indels and ambiguous bases a concatenated nucleotide sequence was created for downstream analysis.

Phylogenetic Analyses

Nucleotide substitution models that best fit the aligned sequences were selected using the

Akaike Information Criterion corrected (AICc) within MEGA 7 (Kumar et al., 2016). The

Tamura 3-parameter with a discrete GAMMA distribution (T92+G) model was selected and used for construction of a neighbor-joining phylogenetic tree for MLSA (Kumar et al., 2016). A 1000 bootstrap samples was set for creating the tree. Trees were annotated with the source of strain location (transplants or field) and to the Xp group they belong. Phylogenetic trees for the core- genome sequences were constructed in IQTree v1.5.5 using the GTR+G+I substitution model with 1000 bootstrap samples. A multiple sequence alignment was constructed using MUSCLE and the best fit model was determined in MEGA6 for SNP outputs from Geneious and Galaxy.

GTR+G was the best fit model used in the phylogenetic analysis. SNPs maximum-likelihood trees were constructed using RAxML v8.2.7 using the GTR substitution model and the GAMMA model rate heterogeneity. A 1000 bootstrap samples was set for building trees.

Effector Analyses Based on previous studies (Schwartz et al., 2015; Potnis et al., 2011) we examined the presence or absence of the following effectors: AvrXv3, XopJ4, AvrBsT, XopQ, XopF2 and

XopAD. Complete gene sequences from previously sequenced draft and reference genomes were extracted and were compared to the draft genomes from this study using the BLAST tool in

NCBI or JGI. Query sequences were extracted and allelic profiles were assigned for each effector allele. Xp strains 91-118, GEV872, GEV993, GEV839, GEV1001 and Xp17-12 were included for comparison.

Results Genomes and Average Nucleotide Identities A total of 67 novel draft genomes were sequenced and assembled into contigs with an average depth coverage of at least 31.7x. Genome GC% ranged from 64.61 to 64.81% across all strains. Genome sizes ranged from 4.9 to 5.2 Mb. See appendix for more detailed genome information for individual strains (Table B-2 and Table B3). The ANI values ranged between

99.3 to 100% among all 67 strains. A pairwise comparison between strains collected within the transplant and field showed a 99.4 to 100% and 99.3 to 100% identity, respectively.

Furthermore, transplant strains from grower A and B were 99.6 to 100% and 99.4 to 99.9% identical, respectively, to corresponding field strains for each grower (Table B-4 and Table B-5).

Multi-Locus Sequence Analysis An MLSA tree was constructed based on six concatenated housekeeping genes from the draft genomes of all 67 strains and reference genomes for a total of 98 strains (Figure 3-1). All

66 strains grouped within the Xp clade into two different groups, 1 and 2, with corresponding strains collected in previous studies (Schwartz et al., 2015; Timilsina et al., 2015). Eight strains

(GEV1920, GEV2067, GEV2087, GEV2088, GEV2108, GEV2109, GEV2121, and GEV2125) had at least one missing or partial housekeeping gene and were omitted from the analysis. Three transplant strains and eight field strains grouped with group 1 which includes the T3 Xp 91-118 strain. The remaining 56 strains grouped within group 2 of T4 Xp. There was no variation between strains within each group and between grower A and B strains.

Core-Genome Comparison Core-genome analysis revealed three major clades within our population. The allelic profiles of the core genes illustrated as a heat map of the core-genome showed variation across only 208 from a total of 886 loci in the population (Figure 3-3). Allelic profiles also corroborated the phylogenetic groupings. Three group 1, reference and representative strains (91-118,

GEV872 and GEV993) did not cluster with any sequenced strain from this study. The majority

(N=53) of strains grouped within group 2 and the remaining strains (N=14) clustered into a sub- clade with the representative strain Xp17-12. The clade consisting of strains clustering with

Xp17-12 was called group 3. Within group 3, only Grower B strains and only one grower A strain (GEV2010) formed group 3. Only field strains GEV2112 and GEV2108 were genetically similar to the three transplant strains based on genetic distance and high bootstrap support.

However, the remaining field strains (GEV2121, GEV2122, GEV2124, GEV2125, GEV2127,

GEV2128, GEV2130, GEV2134) were more distantly related to transplant strains based on genetic distances and separate branch clustering. In group 2, most grower A strains clustered together but three strains (GEV2009, GEV2011 and GEV2013) were highly similar to grower B strains from transplants. Strains from grower A showed less genetic diversity as shown by the branch length and formed a separate clade with high bootstrap support (Figure 3-2). All field

67 strains were genetically similar to transplant strains collected from grower A. Strains collected from grower B showed more genetic diversity and less genetic relatedness between transplant and field strains. The majority of grower B strains clustered distinctly based on either transplant or field recovery location (Figure 3-2). Only nine field strains (GEV2110, GEV2118, GEV2113,

GEV2117, GEV2111, GEV2114, GEV2126, GEV2116 and GEV2109) were genetically similar to transplant strains. The remaining eight field strains were more distantly related to transplant strains based on genetic distances.

Genome-Wide SNP Analyses A total of 401,857 polymorphic nucleotides were distributed across 16,092 positions for

72 strains when compared to the reference genome Xp 91-118. Galaxy workflow produced

16,074 polymorphic positions and a total of 391,996 nucleotides. The outputs from Galaxy and

Geneious overlapped for most polymorphic sites. For the sake of simplicity, the SNP output produced in Geneious will be presented in this study. Most SNPs (>77%) were in coding DNA sequences (CDS) and more than 73% of these CDS SNPs were synonymous substitutions. A summary of the average number of SNPs, coverage and their effect on CDS compared to the reference genome of Xp 91-118 are summarized in Table 3-2. For SNP details of each strain see appendix (Table B-6).

Field outbreaks were linked to transplant outbreaks based on three different criteria: genetic distance, cultivar and grower location. The SNP-based tree showed group 1 forming a distinctive clade which excluded all the strains used in this study. Furthermore, two distinct clades were formed; group 2 which included the representative strains GEV1001 and GEV839 and another group including Xp17-12 (Figure 3-3). Strains recovered from grower A did not cluster with grower B strains and vice versa. In group 3, linking field outbreaks to diseased transplants was more difficult than in group 2, since we did not recover transplant strains from

68 the same field cultivar (Figure 3-4). The group 3 SNP-based tree showed high bootstrap support.

Nevertheless, genetic distances between transplant strains recovered from different cultivars ranged from 0.028 to 0.03 (Table 3-3). The genetic distances of field strains were compared to the transplant population in this group. Field strains with genetic distances outside the transplant pairwise distance range (0.028 to 0.03) were considered genetically distant from transplant strains. Eight grower B field strains were closely related to the three transplant strains within group 3. Nevertheless, these transplant strains were recovered at the same time with the field strains. Two field strains (GEV2125 and GEV2127) were distantly related to any of the three transplant strains within group 3 with a genetic distance ranging from 0.037 to 0.103 (Figure 3-

4).

In group 2, two major sub-clades were formed. The first sub-clade consisted of 16 and three grower A strains clustered within the second sub-clade, respectively. The latter three field strains (GEV2009, GEV2011 and GEV2013) from grower A formed a small clade with high bootstrap support that did not consist of any transplant strains. Pairwise comparison between grower A transplant strains in group 2 showed a genetic distance range of 0.046 to 0.1.

GEV2004 was not closely related to any transplant strains from grower A with a genetic distance of 0.117 to 0.142 (Table 3-4). Strains collected from tomato cultivar H clustered together, however some strains clustered with strains recovered from cultivar M (Figure 3-5). The range of genetic distances between field and transplant strains recovered from cultivars H and M was

0.047 to 0.1 and 0.094 to 0.108, respectively. The range of genetic distances between transplant strains of cultivars H and M was 0.05 to 0.1 and 0.46 to 0.081, respectively. Field strains recovered from H were more genetically similar to transplant strain recovered from the same cultivar. However, strains recovered from field-grown M were genetically distant compared to

69 strains recovered from M transplants based on genetic distances. See appendix for field-cultivar pairwise comparisons for grower A strains (Table B-7). The second sub-clade within group 2 contained grower B strains (Figure 3-5). Transplant strains recovered from different cultivars did not cluster separately. Transplant strains appeared to be closely related regardless of cultivar and collection date. When cultivar-specific strains were compared we found genetic relatedness. The range of genetic distances between transplant strains recovered from cultivar P8 was 0.073 to

0.188. However, the pairwise genetic distance between cultivar P8 transplant strains was 0.077 to 0.131. Based on genetic distances, we found that 11 grower B field strains appeared to be closely related to strains recovered from cultivar P8 in transplant houses. Further, we observed that only 8 field strains closely related to strains recovered from B7 transplants from December

2015. The strains recovered from different transplant cultivars showed high genetic relatedness to field strains of cultivar P0 during March 2016. Six (GEV2110, GEV2111, GEV2117,

GEV2126, GEV2132 and GEV2135) field strains from grower B that were distantly related to strains recovered from transplants. However, these six strains were more genetically similar to transplant strains recovered from the same time period. Based on genetic distances, these field strains were closely related to strains recovered from cultivar P0. Transplant strains recovered from cultivars P0 and P9 were closely related to 15 and 9 field strains, respectively. See appendix for field-cultivar pairwise comparisons (Table B-8).

Race Characterization and Effector Profiles Strains were tested on race differentials carrying different R genes. For the complete list of strain race differential testing see appendix (Table B-9 and Table B-10). All strains were pathogenic and gave a susceptible reaction on tomato cvs. Bonnie Best and Hawaii 7998, except for GEV2112 elicited a weak HR on Hawaii7998. However, 67 strains were pathogenic on

FL216 except for six strains elicited very weak HRs 36 hours post inoculation (Table B-10).

Based on the phenotypic data, all strains belong to Xp race T4. Three strains from grower A did not elicit an HR on pepper cv. ECW, whereas the remaining strains elicited an HR on pepper.

Seven strains from grower B did not elicit an HR on pepper cv. ECW, while the remaining strains did.

Phenotypic data were compared to the effector profiles of each strain. The avrXv3 locus involved in a gene-for-gene interaction with Xv3 in tomato FL216 was examined for all strains. avrXv3 gene sequences in all the draft genomes were either truncated or disrupted by an early stop codon or were located on split contigs (Figure 3-6). avrXv3 was only present in the reference genome 91-118. The core effector, xopQ (Potnis et al., 2011; Schwartz et al., 2015) was present in all strains in groups 2 and 3. Two alleles (1 and 2) were found for the xopQ gene sequences. Allele 2 of xopQ was prevalent in group 2 strains, whereas the other allele was found only in group 1 and 3. Furthermore, avrXv4, another core effector in T4 strains, was present in all the strains with no genetic diversity. In pepper ECW, avrBsT, is involved in a gene-for-gene interaction and was present only in group 1 and 2 strains but not in the reference genome or group 3 strains. Six group 2 strains lacked avrBsT. Although group 3 strains lacked avrBsT, 7 of the 14 strains contained an avrBsT-homolog designated XopJ6 (Figure 3-6). XopJ6, shares around 71% nt identity with avrBsT and belongs to the YopJ-like effector family (Figure B-1). A number of strains in group 2 which lacked avrBsT produced a weak HR, on pepper plants (Table

B-10). On the other hand a number of strains within group 2 that carried avrBsT did not elicit an

HR (Table B-10).

Discussion In this study we sequenced Xp strains by tracking seedlings from the transplant source to commercial grower fields at two different grower sites. We showed that strains recovered from diseased seedlings are readily introduced and can cause outbreaks in tomato fields. Apart from

71 this study no one has studied the genetic diversity of Xanthomonas in the tomato production line.

We utilized three different methods, MLSA, core-genome MLST and SNPs, to examine and measure when possible the genetic relatedness between species. It should be noted that identifying the exact source of a disease outbreak remains a difficult task (Schürch and Siezen,

2010). Several external factors can limit the tracing of epidemics to a single strain such as sampling bias, secondary infections caused by introduced inoculum, plant movement, and horizontal gene transfer in the field. Such complex situations are common in medical bacterial epidemics in which showing a connection to a single progenitor strain is often difficult (Schürch and Siezen, 2010). Nevertheless, an association with a source strain can be demonstrated through utilizing molecular biology tools, specifically by whole-genome sequencing and SNP typing

(Croucher et al., 2013; Maiden et al., 2013; Schürch and Siezen, 2010).

MLSA results confirmed that Xp remains the causal agent of BST in Florida. More than

80% of the strains were clustered in group 2, the X. euvesicatoria-like recombinant group

(Timilsina et al., 2015), whereas less than half were in group 1 which is more similar to the Xp

91-118 reference strain. Although MLSA analysis placed strains within Xp, it lacked sufficient resolution to discriminate between strains within individual groups. The use of MLSA for discriminating strains of low diversity or variants of a single-clone is insufficient and requires more sensitive tools (Maiden et al., 2013). Core-genome MLST (cgMLST) was conducted to achieve a higher resolving power. cgMLST showed that several transplant and field strains did not cluster together. Nevertheless, grower A strains tightly clustered between transplant and field strains. On the other hand, grower B strains showed a distinct split between strains based on either transplant or field location. Grower B strains showed higher diversity than grower A strains. We did observe similar phylogenetic grouping patterns to Schwartz et al. (2015).

However, within our cgMLST a distinct clade with high branch support was observed for strains clustering with Xp17-12 in which we proposed a new group. Previously, Xp17-12 was placed in subgroup 1B within group 1 (Schwartz et al., 2015). Interestingly, Xp17-12 was isolated in 2006 from tomato in Florida and shares two effectors (XopF2 and XopAD) that are identical to X. euvesicatoria 85-10, but not to other Xp strains. AvrBsT is missing in Xp17-12 and as a result this strain causes disease in pepper plants (Schwartz et al., 2015). Strains of grower B within group 3 were not very related based on cgMLST. cgMLST failed to provide a high resolution for identifying any relatedness between transplant and field strains when compared to the SNP-based approach. Although cgMLST is considered a highly sensitive tool in speciation and for other taxonomic purposes this method has its limitations. The lower resolving power of cgMLST can be attributed to less genome coverage compared to a SNP-based method that extends over the whole-genome and not only core-genes.

SNP analysis was shown to be suitable for resolving differences within populations with low genetic diversity when a suitable reference genome is available (Croucher et al., 2013;

Maiden et al., 2013). A previous study analyzed the diversity of Xanthomonas sp. causing BST using a different reference strain, X. axonopodis pv. citri ‘306’, which could be a source of error in creating a suitable SNP-based phylogeny (Schwartz et al., 2015). The SNPs used in this study were based on the complete genome of Xp strain 91-118. SNP-based analysis provided a higher resolution than cgMLST. The SNP-based analysis showed three groups similar to cgMLST.

Interestingly, group 1 had the least amount of SNPs compared to the strains in the other groups based on the reference genome 91-118. This can be explained by the higher genetic relatedness of group 1 strains to the reference strain previously observed (Schwartz et al., 2015; Timilsina et al., 2015). Interestingly, based on cgMLST and SNP analysis, typical group 1 strains were not

73 recovered in our collection, indicating a possible phylogenomic shift and drift from the ancestral

91-118-like strains. This rapid shift in only two years since the last known survey indicates a continuous genetic migration of Xp strains. When examining strains in the SNP tree, we found that several transplant strains appear to be very closely related to field strains from both grower sites. All field strains from grower A appear to resemble those from contaminated transplants.

Similarly, the set of field strains from grower B showed a similar trend. Although some field strains appear to be more closely related to transplant strains and vice versa. The overall trend indicates strain movement from transplants to the field. When we examined field and transplant strains from the same grower but from different cultivars and collection dates we found high genetic similarity. The lack of strain clustering based on cultivar indicates that strains are circulating in the transplant facility. This is mainly due to the proximity of flats in the same house. We have seen that strains can easily move inside a transplant facility over a short period of time (P. Abrahamian, unpublished data). As a result, we also saw that several transplant strains isolated from different cultivars are similar to field strains. The high level of similarity between strains from different environments and time periods was unexpected. Nevertheless, a few number of strains showed no genetic relatedness between transplant and field strains. We propose two explanations for this occurrence. First, some field strains from both grower sites that do not have any counterpart transplant strain are possibly exotic strains that did not originate from the transplants and were most likely introduced throughout the season. However, more importantly the sample size of transplant strains was very small due to the continuous rogueing of diseased plants in the transplant facilitiy and we only collected once during the production cycle.

The majority of the strains from this survey belong to race T4, consistent with prior surveys showing the increasing prevalence of race T4 over the last decade. Timilsina et al.

(2016) showed that race T4 strains displaced the T3 strains by 2012. Race testing in this study did not strictly corroborate the effector profiles of the strains. However, these strains gave inconsistent reactions corresponding to their genotype. Therefore, we characterized the sequence of four different effector genes that are of biological relevance for Xp. These four effectors are potentially involved in pathogen fitness, pathogenicity and host range (Chapter 2; Schwartz et al., 2015; Potnis et al., 2011, 2015). The race T3 determinant, avrXv3, was absent but avrXv4 was present in all the strains similar to other characterized strains that belong to race T4

(Timilsina et al., 2016). The strains that gave a very weak HR had a partially or completely missing avrXv3. It is possible that the Xv3 resistance locus might be interacting with either a partially expressed avrXv3 or with other unknown Xp effectors. We also found the erosion and acquisition of type III effectors such as the absence of AvrBsT and acquisition of XopJ6; both are plasmid-borne. Similarly, Barak et al. (2016) observed that several previously identified core effectors, such as xopAD and xopC2, underwent mutations rendering these genes inactive. On the other hand, several genes were also acquired compensating for this loss in X. euvesicatoria

(Barak e al., 2016). The effector, XopQ, which is ubiquitous in Xathomonas sp. (Potnis et al.,

2011) was present in our population confirming it as a core effector. Although avrBsT was missing in several strains in both groups, a novel gene, xopJ6, which shares high homology with avrBsT was only present in strains belonging to group 3. This gene also appears to interact with the Bst locus in pepper since strains carrying xopJ6 elicited an HR in pepper ECW. It would be interesting in the future to sample more strains and examine whether this effector gene replacement is indeed specific to group 3 only. Furthermore, the presence of xopJ6 was not only

75 limited to field strains. This is further evidence indicating field strains carrying xopJ6 originated from the movement of contaminated transplants.

The introduction of strains to the field is a serious issue that poses a continuous risk in sustainable tomato production. Based on results in this study, we demonstrated that strains in the field genetically resemble strains recovered from the same plants in transplant houses. This is most likely caused by the planting of transplant seedlings in the field. This suggests that the primary source of disease outbreaks in the field may originate from non-symptomatic tomato transplants. This study also reinforces the classical observation that movement of plant material results in long-distance pathogen dispersal. This raises a big challenge for sustainable vegetable production in which exotic or local strains are in a constant influx and threatens plant resistance breeding efforts. The implications of this study are manifold at the ecological, disease management and scientific level. At the ecological level, Xp populations are primarily shaped by the strains that contaminate transplant growers. This results in a low genetic diversity in which strains are constantly circulated from the transplant house to the open field. Seeds are another potential source of BST inoculum that can carry pathogens for long-distances. Further studies should reveal the genetic population of Xp in contaminated tomato seed lots and how this relates to the global movement of the pathogen. Maintaining healthy transplants and excluding Xp infections is crucial to reduce BST field outbreaks. Transplants are most likely the primary source of inoculum into a production field. While transplant growers typically manage BST on seedlings through chemical sprays, such strategies are not absolutely effective in preventing disease outbreaks. Furthermore, BST management in transplant houses is extremely difficult due to the inability of current structures to mitigate conducive environmental factors. In the future,

76 the strategy developed in this study can be applied for characterizing and tracing bacterial disease outbreaks in other agricultural systems.

Table 3-1. List of Xanthomonas strains characterized and used in this study. Grower Strain Date (mo-yr) Tomato cultivar Locationa Sourceb GEV1989 10-2015 H Manatee Co. F GEV1991 10-2015 H Manatee Co. F GEV1992 10-2015 H Manatee Co. F GEV1993 10-2015 H Manatee Co. F GEV2004 10-2015 H Manatee Co. F GEV2009 10-2015 M Manatee Co. F GEV2010 10-2015 M Manatee Co. F GEV2011 10-2015 M Manatee Co. F GEV2013 10-2015 M Manatee Co. F GEV2015 10-2015 M Manatee Co. F A GEV2047 9-2015 M Polk Co. GH GEV2048 9-2015 M Polk Co. GH GEV2049 9-2015 M Polk Co. GH GEV2050 9-2015 M Polk Co. GH GEV2052 9-2015 M Polk Co. GH GEV2055 9-2015 H Polk Co. GH GEV2058 9-2015 H Polk Co. GH GEV2059 9-2015 H Polk Co. GH GEV2060 9-2015 H Polk Co. GH GEV2063 9-2015 H Polk Co. GH GEV1911 12-2015 H Collier Co. GH GEV1912 12-2015 H Collier Co. GH GEV1913 12-2015 H Collier Co. GH GEV1914 12-2015 P8 Collier Co. GH GEV1915 12-2015 P8 Collier Co. GH GEV1916 12-2015 P8 Collier Co. GH GEV1917 12-2015 P8 Collier Co. GH GEV1918 12-2015 P8 Collier Co. GH GEV1919 12-2015 B7 Collier Co. GH GEV1920 12-2015 B7 Collier Co. GH GEV1921 12-2015 P9 Collier Co. GH GEV2065 3-2016 P9 Collier Co. GH GEV2067 3-2016 P9 Collier Co. GH B GEV2072 3-2016 P9 Collier Co. GH GEV2087 3-2016 P0 Collier Co. GH GEV2088 3-2016 P0 Collier Co. GH GEV2089 3-2016 P0 Collier Co. GH GEV2097 3-2016 P3 Collier Co. GH GEV2098 3-2016 P3 Collier Co. GH GEV2099 3-2016 P3 Collier Co. GH GEV2108 3-2016 P8 Collier Co. F GEV2109 3-2016 P8 Collier Co. F GEV2110 3-2016 P8 Collier Co. F GEV2111 3-2016 P8 Collier Co. F GEV2112 3-2016 P8 Collier Co. F GEV2113 3-2016 P8 Collier Co. F GEV2114 3-2016 P8 Collier Co. F

Table 3-1. Continued. GEV2115 3-2016 P8 Collier Co. F GEV2116 3-2016 P8 Collier Co. F GEV2117 3-2016 P8 Collier Co. F GEV2118 3-2016 P8 Collier Co. F GEV2119 3-2016 P8 Collier Co. F GEV2120 3-2016 P8 Collier Co. F GEV2121 3-2016 P8 Collier Co. F GEV2122 3-2016 P8 Collier Co. F GEV2123 3-2016 P8 Collier Co. F GEV2124 3-2016 P8 Collier Co. F B GEV2125 3-2016 P8 Collier Co. F GEV2126 3-2016 P8 Collier Co. F GEV2127 3-2016 P8 Collier Co. F GEV2128 3-2016 P8 Collier Co. F GEV2129 3-2016 P8 Collier Co. F GEV2130 3-2016 P8 Collier Co. F GEV2132 3-2016 P8 Collier Co. F GEV2133 3-2016 P8 Collier Co. F GEV2134 3-2016 P8 Collier Co. F GEV2135 3-2016 P8 Collier Co. F a Strains were collected from Florida, US. b F: field; T: transplant.

Table 3-2. Average number of single nucleotide polymorphisms (SNPs), coverage and effects of SNPs on CDS for each phylogenetic group compared to reference strain Xp 91-118. SNP effect on CDSb

SNPs SNPs Genetic Average Average Non- in in Synonymous Truncation Extension Group SNPs Coverage synonymous CDSa non- CDS 1 1568 49 1255 313 335 917 2 1 2 5642 34 4362 1280 1047 3303 8 7 3 5895 58 4601 1294 1103 3489 7 2 a CDS: coding DNA sequence b Effect of SNP on amino acid sequence. Non-synonymous: amino acid change; synonymous: no amino acid change; truncation: early stop codon in CDS; extension: amino acid addition.

Table 3-3. Range of genetic distances between transplant (upper panel) and field (left panel) of grower B strains in group 3 based on single-nucleotide polymorphisms. GEV2065 GEV2097 GEV2099 Range GEV2112 0.024 0.03 0.031 0.024-0.031 GEV2128 0.018 0.029 0.03 0.018-0.03 GEV2108 0.024 0.019 0.027 0.019-0.027 GEV2124 0.027 0.026 0.03 0.026-0.03 GEV2134 0.028 0.027 0.03 0.027-0.03 GEV2125 0.102 0.103 0.102 0.102-0.103 GEV2121 0.03 0.031 0.029 0.029-0.031 GEV2130 0.031 0.032 0.03 0.03-0.032 GEV2127 0.04 0.041 0.037 0.037-0.041 GEV2010a 0.03 0.031 0.023 0.023-0.031 GEV2122 0.027 0.033 0.035 0.027-0.035 Overall distance between transplant strains 0.028-0.03 a GEV2010 is the only grower A strain in group 3.

Table 3-4. Range of genetic distances between transplant (upper panel) and field (left panel) grower A strains in group 2 based on single-nucleotide polymorphisms.

GEV2060 GEV2055 GEV2058 GEV2048 GEV2050 GEV2052 GEV2049 GEV2047 GEV2059 GEV2063 Range GEV2004 0.142 0.134 0.127 0.124 0.123 0.122 0.117 0.128 0.125 0.124 0.117-0.142 GEV1993 0.074 0.066 0.059 0.056 0.068 0.067 0.062 0.074 0.07 0.069 0.056-0.074 GEV1989 0.089 0.081 0.073 0.07 0.045 0.059 0.053 0.065 0.065 0.064 0.045-0.089 GEV1992 0.1 0.092 0.084 0.081 0.071 0.066 0.061 0.051 0.076 0.075 0.051-0.1 GEV2015 0.088 0.08 0.072 0.07 0.059 0.055 0.049 0.051 0.064 0.063 0.049-0.088 GEV1991 0.094 0.086 0.079 0.076 0.068 0.067 0.062 0.074 0.047 0.063 0.047-0.094 GEV2011 0.12 0.112 0.104 0.101 0.1 0.099 0.094 0.106 0.102 0.101 0.094-0.12 GEV2013 0.121 0.113 0.106 0.103 0.102 0.101 0.096 0.107 0.103 0.103 0.096-0.121 GEV2009 0.122 0.114 0.106 0.104 0.102 0.102 0.096 0.108 0.104 0.104 0.096-0.122

Overall distance between transplant strains 0.046-0.1

8 2

Table 3-5. Range of genetic distances between transplant (upper panel) and field (left panel) grower B strains in group 2 based on single-nucleotide polymorphisms. GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV Range 1912 2072 1918 2089 1921 2088 2087 2098 1920 2067 1914 1915 1916 1911 1913 1917 1919 GEV 0.065- 0.079 0.065 0.086 0.083 0.093 0.119 0.13 0.119 0.12 0.115 0.137 0.12 0.09 0.099 0.078 0.073 0.069 2109 0.137 GEV 0.069- 0.096 0.069 0.103 0.1 0.11 0.136 0.147 0.136 0.137 0.132 0.154 0.137 0.107 0.116 0.095 0.09 0.086 2118 0.154 GEV 0.07- 0.089 0.091 0.08 0.078 0.07 0.105 0.116 0.105 0.106 0.101 0.123 0.106 0.081 0.089 0.077 0.079 0.079 2129 0.123 GEV 0.12- 0.144 0.147 0.136 0.133 0.135 0.12 0.131 0.121 0.139 0.134 0.157 0.14 0.137 0.145 0.133 0.135 0.135 2135 0.157 GEV 0.118- 0.142 0.145 0.134 0.131 0.132 0.118 0.129 0.118 0.137 0.132 0.154 0.137 0.134 0.143 0.131 0.132 0.132 2117 0.154 GEV 0.119- 0.143 0.145 0.134 0.132 0.133 0.119 0.13 0.119 0.137 0.133 0.155 0.138 0.135 0.143 0.131 0.133 0.133 2132 0.155 GEV 0.116- 0.14 0.143 0.132 0.129 0.13 0.116 0.127 0.116 0.135 0.13 0.152 0.135 0.132 0.141 0.129 0.13 0.131 2126 0.152 GEV 0.12- 0.144 0.147 0.136 0.133 0.134 0.12 0.131 0.12 0.139 0.134 0.156 0.139 0.136 0.145 0.133 0.134 0.134

8 2111 0.156 3

GEV 0.125- 0.149 0.151 0.141 0.138 0.139 0.125 0.136 0.125 0.144 0.139 0.161 0.144 0.141 0.15 0.138 0.139 0.139 2110 0.161 GEV 0.092- 0.126 0.129 0.118 0.116 0.117 0.092 0.103 0.092 0.121 0.117 0.139 0.122 0.119 0.127 0.115 0.117 0.117 2119 0.139 GEV 0.105- 0.134 0.137 0.126 0.123 0.124 0.105 0.116 0.105 0.129 0.124 0.146 0.129 0.126 0.135 0.123 0.124 0.124 2133 0.146 GEV 0.095- 0.108 0.111 0.1 0.098 0.099 0.103 0.113 0.103 0.099 0.095 0.117 0.1 0.101 0.109 0.097 0.099 0.099 2113 0.117 GEV 0.068- 0.116 0.118 0.107 0.105 0.106 0.11 0.121 0.11 0.072 0.068 0.097 0.08 0.108 0.116 0.104 0.106 0.106 2114 0.121 GEV 0.081- 0.121 0.124 0.113 0.11 0.112 0.115 0.126 0.116 0.085 0.081 0.086 0.082 0.114 0.122 0.11 0.112 0.112 2120 0.126 GEV 0.096- 0.114 0.117 0.106 0.104 0.105 0.108 0.119 0.109 0.101 0.096 0.119 0.102 0.107 0.115 0.103 0.105 0.105 2116 0.119 GEV 0.086- 0.104 0.107 0.096 0.093 0.094 0.098 0.109 0.099 0.091 0.086 0.108 0.091 0.097 0.105 0.093 0.094 0.095 2115 0.109 GEV 0.136- 0.149 0.151 0.14 0.138 0.143 0.17 0.18 0.17 0.17 0.166 0.188 0.171 0.136 0.144 0.137 0.139 0.139 2123 0.188 0.060- Overall distance between transplant strains 0.143

Figure 3-1. Maximum-likelihood phylogenetic tree of Xanthomonas perforans strains based on the concatenated sequences of housekeeping genes (lacF, gyrB, fusA, gapA, lepA, gltA). Green squares and purple circles refer to field and transplant isolates, respectively. Numbers on branches indicate bootstrap support.

Figure 3-2. Maximum-likelihood phylogenetic tree of 73 Xanthomonas perforans strains based on the concatenated core-genome sequences of 886 genes (0.8 Mb). Green squares and purple circles refer to field and transplant strains, respectively. Tree was rooted with X. perforans 91-118 (red triangle). Group 1 consist of representative strains GEV993 and GEV872 (blue triangles) only. Group 2 representative strains are GEV1001 and GEV839 (blue triangles). Numbers on branches indicate bootstrap support.

Figure 3-3. Maximum-likelihood phylogenetic tree of 73 Xanthomonas perforans strains based on 16,092 concatenated SNPs with a heat map based on the allelic type of 208 out of 886 core genes that were not identical among all X. perforans strains. Colors in the heat map indicate the allelic type. Green squares, purple circles, red and blue triangles refer to field, transplant, reference and representative strains, respectively. Letters near strain names indicate grower A or B. Numbers on branches indicate bootstrap support.

Figure 3-4. Unrooted maximum-likelihood phylogenetic tree of group 3 strains of Xanthomonas perforans strains based on 16,092 concatenated SNPs. Strains are annotated with location and cultivar. Numbers on branches indicate bootstrap support.

Figure 3-5. Unrooted maximum-likelihood phylogenetic tree of group 2 strains of Xanthomonas perforans strains based on 16,092 concatenated SNPs. Strains are annotated with location and cultivar. Numbers on branches indicate bootstrap support.

Figure 3-6. Effector profile based on xopJ2, xopJ6, avrXv3, xopJ4 and xopQ overlapping with the maximum-likelihood phylogenetic tree of 73 Xanthomonas perforans strains based on 16,092 concatenated SNPs. Black boxes indicate presence of gene and high similarity. Empty boxes indicate either complete absence of gene, the presence of an early stop codon disrupting the open reading frame or a contig break. XopQ has two allele types 1 and 2. Green squares, purple circles, red and blue triangles refer to field, transplant, reference and representative strains, respectively. Numbers on branches indicate bootstrap support.

CHAPTER 4 EFFICACY OF COPPER ALTERNATIVES FOR MANAGEMENT OF BACTERIAL SPOT ON TOMATOES VARIES IN TRANSPLANT AND FIELD PRODUCTION

Tomato, Solanum lycopersicum L., is an important cash crop in the United States worth more than 1 billion dollars of which Florida accounts for one-third of total production (NASS,

2016). Bacterial spot of tomatoes (BST) is an important disease of tomato seedlings and in field production and can cause up to 50% yield reduction in tomato (Gitaitis et al., 1992; Pohronezny and Volin, 1983). Bacterial leaf spot is caused by four different species of Xathomonas; Xp, X. euvesicatoria, X. gardneri and X vesicatoria (Stall et al., 2009; Potnis et al., 2015). Four races

(T1, T2, T3, T4) are described for the four Xanthomonas species (Jones et al., 2005). Xp is the only prevalent species and causal agent of BST in Florida, and is comprised of two races, T3 and

T4 (Stall et al., 2009). Race T3 was the prevalent race in Florida in the 1990’s and early 2000’s.

However, surveys showed high prevalence of race T4 by 2006 and complete displacement of T3 by 2012 (Horvath et al., 2012; Jones et al., 1995, 1998; Potnis et al., 2015; Timilsina et al.,

2016). BST is favored by high relative humidity and optimal temperatures ranging from 25 to

28°C (Obradovic et al., 2008). BST foliar symptoms consist of irregular water-soaked lesions that later become necrotic, coalesce and lead to defoliation (Jones and Miller, 2014). Fruit symptoms consist of upraised, scabby lesions that render fruit unmarketable (Jones and Miller,

2014).

The southern US states, especially Georgia and Florida, are major suppliers of tomato transplant seedlings for the northeastern and central US growers for the fresh tomato market

(Gitaitis et al., 1992). Until the 1990’s, tomato seedlings for fresh market were produced directly in the field (Gitaitis et al., 1992). However, BST outbreaks in transplant fields along with other foliar fungal diseases and the introduction of mechanized planting required a shift to greenhouse-

90 grown seedlings (Gitaitis et al., 1992). Currently, tomato transplant seedlings are produced in greenhouses. Environmental conditions within these structures favor BST epidemics, due to frequent overhead watering, high humidity and high plant densities (Potnis et al., 2015). Under field conditions, BST can be primarily introduced into a field through diseased tomato seedlings and to a lesser extent by volunteer tomato plants and neighboring infested fields, whereas, bacteria spreads in the field by wind-driven rain, aerosols and by cultural practices (e.g. pruning, tying, spraying and harvesting) (Bernal and Berger, 1996; Jones et al., 1986; McInnes et al.,

1988; Pohronezny et al., 1990). Furthermore, Xanthomonas spp. can survive as an epiphyte on tomato leaves (Jones et al., 1986). Epiphytic bacterial populations can be problematic due to prolonged surface colonization prior to infection. As a result, under conducive conditions, symptom development can rapidly occur (Hirano and Upper, 1983).

BST is best controlled through an integrated disease management approach. Cultural practices are commonly used, such as roguing diseased plants, removal of weed and volunteer plants in and around fields, crop rotation and transplanting healthy seedlings (Potnis et al., 2015).

Host resistance against bacterial spot has been identified in tomato (Scott et al., 2006).

Nevertheless, none of these resistance (R) genes are currently deployed in commercial tomato cultivars, due to the continuous shifts in Xanthomonas races and strains that have hampered resistance breeding efforts (Jones et al., 1995, 1998). Bactericides have played a major role in managing BST in the field. Copper and copper tank mixes of copper-mancozeb have been primarily used as the grower standard (Ritchie, and Dittapongpitch, 1991). However, in recent decades an increasing number of Xanthomonas strains were copper-tolerant (Timilsina et al.,

2013). In 2006 and 2012 two separate surveys were carried out and showed that Xp strains collected from Florida exhibited 100% tolerance to copper compounds (Horvath et al., 2012;

Timilsina et al., 2013). Copper tolerance and resistance is associated with plasmid-borne or chromosomal genes (Basim et al., 2005; Stall et al., 1986). Moreover, in the same surveys, streptomycin resistance was only present in 5% of the strains in 2006 but increased to 32% of the strains in 2012 (Horvath et al., 2012; Timilsina et al., 2013; Vallad et al., 2013). Also, the frequency of resistance to streptomycin was higher for strains collected from transplant houses than strains from the field at 86% and 14%, respectively (Vallad et al., 2013).

Alternatives to copper, mancozeb and streptomycin such as acibenzolar-S-methyl (ASM;

Actigard®; Syngenta Crop Protection, Inc., Greensboro, NC), bacteriophages, Bacillus subtilis

QST 713 (Serande® Opti; Bayer Crop Science, Leverkusen, Germany), cymoxanil+famoxadone

(Tanos®, E.I. du Pont de Nemours and Company, Wilmington, DE), nanoparticles were reported to reduce disease severity (Abbasi and Weselowski, 2015; Jones et al., 2012; Louws et al., 2001;

Robert et al., 2008; Strayer et al., 2015). ASM, a synthetic compound, induces systemic acquired resistance (SAR) against a broad range of pathogens (Vallad and Goodman, 2004). ASM showed significant disease reduction compared to copper-mancozeb sprays (Louws et al., 2001). ASM is an effective alternative for copper where copper-tolerant strains are prevalent (Louws et al.,

2001). Foliar- or drip- applications of ASM showed statistically similar levels of disease control under field conditions (Huang and Vallad, 2011). Nevertheless, several studies showed that ASM did not improve yield (Huang et al., 2012; Louws et al., 2001; Obradovic et al., 2004).

Bacteriophages were shown to be efficient in reducing BST, however the persistence on leaf surfaces is not very long and requires special application (Jones et al., 2012; Obradovic et al.,

2004). Single phage applications or tank-mixes with ASM significantly reduced disease severity

(Obradovic et al., 2004, 2005). Cymoxanil+famoxadone (Tanos®) were evaluated for control of a copper-tolerant Xp strain (Fayette et al., 2012; Robert et al., 2008). Disease management in

92 transplant houses, relies on a combination of cultural practices and chemical sprays prior to seedling delivery. Greenhouse cultural practices such as the removal of diseased plants, reduced plant densities and limiting periods of plant wetness can reduce BST. Further, the management of diseased transplants is crucial in reducing BST field outbreaks.

BST management relies heavily on chemical applications. However, the limited number of efficacious chemical alternatives to copper, the widespread prevalence of copper tolerant strains and recurrent BST outbreaks requires the development of new integrated approaches for sustainable disease management. The efficacy and suitability of current field-labeled products for managing BST on tomato seedlings under greenhouse conditions is unknown. Therefore, in this study we evaluated the efficacy of utilizing novel alternatives to copper such as biological control agents (BCAs), fungicides, plant defense activators and other miscellaneous products for

BST control and compared the effectiveness of such products on transplants and field-grown tomato plants.

Materials and Methods Bacterial Strains, Inoculation and Plants Four different strains of Xp (GEV599, GEV904, GEV917, and GEV1063) were used for the field trials. Strains GEV599 and GEV872, GEV904, GEV917, GEV1063 were isolated in

2007 and 2012, respectively. All strains used in this study are tolerant to copper. Xp strains

GEV872, GEV904 and GEV917 are also resistant to streptomycin. Xp strains GEV599 and

GEV1063 are sensitive to streptomycin. Bacterial isolates were stored in a 30% glycerol-nutrient broth stock at -80°C and were freshly streaked on nutrient agar for preparing fresh inoculum from single colonies. Bacteria were grown at 28°C for 48 h. Bacteria were harvested and suspended in 0.01 M MgSO4 with a surfactant, 0.5% Tween or 0.05 % Silwett L-77, and

8 -1 6 adjusted to 5x10 CFU.ml (A600=0.3). Bacterial concentrations were further adjusted to 10 and

107 CFU.ml-1 for the first (Fall 2015) and second (Fall 2016 A and B) field trials, respectively.

For inoculation in the field, two plants at the opposing ends of each field plot were spray- inoculated. For inoculations in greenhouse experiments Xp strain GEV872 was adjusted to 104

(trial 1) or 105 CFU/ml (trials 2 to 6) with 0.05% of Silwet L-77. Bacterial suspensions were spray-inoculated using a hand-held mechanical mister one day after applying greenhouse treatments.

Tomato plants at the second true leaf stage were obtained from a commercial transplant producer and maintained in the greenhouse at the University of Florida’s Gulf Coast Research and Education Center in Balm, FL. Tomato ‘Tycoon’ was used in field trials. Tomato ‘HM1823’ and ‘FL.47’ were used in the first and remaining greenhouse trials, respectively. Tomato plants were irrigated and fertilized according to standard grower practices.

Greenhouse Trials Greenhouse trials were repeated six times during September to December 2015, November

2016 and February to March 2017 in a temperature controlled greenhouse. Tomato (‘HM1823’ and ‘FL.47’) seedlings produced in 128-cell Speedling trays were laid out in a randomized

5 2 complete block design. Chemicals were applied using a CO2 back-pack sprayer at 4.1x10 N/m with a single hollow cone tip. Each Speedling tray received about 250 ml of chemical product.

Initial treatments were one day before inoculation, with the exception of one of the ASM treatments that was applied to plants three weeks after bacterial inoculation during fall 2016 and spring 2017 trials. Chemical application frequency varied between one to three times depending on treatments. Each treatment was repeated at least in two trials, some treatments such as copper hydroxide and non-treated control were included in all six trials for comparing efficacies. For each trial, treatments were replicated four times and each treatment consisted of four 128-cell

Speedling trays (one tray per replicate) within a single trial. Each trial spanned a period of 3 to 4

94 weeks from inoculation date until final rating. Refer to Table 4-1 for a list of chemical names and their corresponding active ingredient and manufacturer used. The spray rates used in the greenhouse were: ammonium chloride 0.45 ml/L; ASM 0.07 g/L; Bacillus amyloliquefaciens

‘D747’ 2.5 ml/L; Bacillus subtilis ‘QST 713’ 1.04 g/L; bacteriophage 5 ml/L; copper hydroxide

1.2 g/L; copper octanoate 5 ml/L; famoxodone+cymoxanil 0.6 g/L; oxytretracycline 1.17 g/L; oxysilver nitrate 1 g/L; pentasilver hexaoxoiodate 1.4 g/L; potassium bicarbonate 2.4 g/L; potassium silicate 10 ml/L; quinoxyfen 0.4 ml/L, 0.23 ml/L, 0.08 ml/L; streptomycin 1.17 g/L; sodium phosphoric acid 7.5 ml/L; USF2018A 0.07 ml/L, 0.13 ml/L, 0.27 ml/L. Quinoxyfen was applied at three different rates corresponding to three different treatments. USF2018A was applied at an increasing spray rate within the same treatment. Plants were rated for disease severity, based on percentage of symptomatic foliage showing brown necrotic lesions per tray at two or five days after applying the last treatment. Plants were visually evaluated for phytotoxicity at each rating period.

Field Trials Field trials were conducted during fall 2015 and 2016 to test bacterial spot management in field-grown tomatoes. Each tomato bed consisted of three 97.5 m long rows with 7.6 m plots on 1.8 m centers with plants grown on 60 cm spacing. Each plot had 10 tomato plants. Beds were fumigated, fertilized and covered with white plastic mulch according to grower practices.

Fungicides and insecticides were applied to manage foliar fungal diseases such as target spot and early blight and insect infestations. Trials included a grower standard treatment consisting either of copper oxychloride or copper hydroxide+ mancozeb or copper sulfate+ mancozeb and several non-copper alternatives. Each trial included a non-treated control. Treatments were applied using a high-clearance tractor sprayer equipped with 8 hollow cone spray nozzles for each plant row.

Spray nozzles were calibrated to deliver 0.04, 0.07, and 0.09 L per m2 at 1.448x106 N/m2

95 according to different tomato growth stages. Each treatment was replicated four times in a randomized complete block design. The center 10 plants in the center bed of each 3-bed plot were individually rated every 7 to 14 days for the severity of bacterial spot based on the Horsfall-

Barratt scale until first harvest. The center 10 plants in each plot were hand harvested once, graded for market size, and the incidence of fruit with BST symptoms during fall 2016 trials.

The number of spray applications varied between treatments. During fall 2015, plants were treated over a period of eight weeks. All seedlings were treated with one spray prior to transplanting. All plants were treated weekly except for the following treatments: cymoxanil+famoxadone and oxytetracycline were applied four times during every other week.

Streptomycin was applied once before transplanting. The spray rates for fall 2015 were similar to treatments listed in tables 2 and 3. Spray rates for ammonium chloride and sodium phosphoric acid were at 7.5 ml/L.

During fall 2016, plants in trials ‘A’ and ‘B’ were treated over a period of five and eight weeks, respectively. The number of applications and spray rates used for each compound are listed in tables 2 and 3 for trials ‘A’ and ‘B’ conducted in fall 2016.

Statistical Analyses All statistical analyses were performed using SAS® 9.3 (SAS Institute Inc., Cary, NC,

USA). Homogeneity of variances was tested using the covariance test and residuals plotted to check for normality. Horsfall-Barratt values for field ratings were converted to mid-point percentages according to Bock et al. (2009). Area under the disease progress curve (AUDPC) was determined for the field data. AUDPC was calculated using the trapezoidal method (Jeger,

2004). Disease severity percentages and AUDPC values were log-transformed prior to analyses.

Log-transformed values were analyzed using a generalized linear mixed model (PROC

GLIMMIX) within SAS® 9.3 (SAS Institute Inc., Cary, NC, USA) at alpha=0.05. Treatment

96 means were separated using Fisher’s protected least significant difference (LSD) test. Disease severity percentages and AUDPC values were back-transformed for data presentation.

Results Efficacy of BST Reduction in Tomato Seedlings

Untreated tomato seedlings showed a very high disease incidence that ranged from 45 to

88% across all greenhouse trials. In only the first trial, disease pressure was lower than the remaining five trials (Figure 4-1). The grower standard copper oxychloride was significantly different from the control in three out of six trials and the disease severity ranged from 3.5 to

45% across the repeated trials. The addition of cymoxanil+famoxadone improved copper oxychloride performance in only one trial when the disease pressure was low (Figure 4-1).

Antibiotics, streptomycin and oxytetracycline, and the fungicide potassium silicate treatments were significantly different from the control plants and had equal or less efficacy compared to copper oxychloride in only the first trial. Single treatments of compounds such as ammonium chloride, phosphoric acid and bacteriophage were statistically ineffective against BST compared to the untreated control treatment in three different trials (Figure 4-1 and Figure 4-2).

Quinoxyfen caused phytotoxicity at application rates greater than 0.4 ml/L. Plants treated with quinoxyfen exhibited stunted growth, leaf epinasty and thickened stems. However, quinoxyfen at rates of 0.4 ml/L significantly reduced BST severity by 40 to 51 % compared to the untreated control treatment. Furthermore, the three experimental compounds, USF2018A, oxysilver nitrate

(OSN) and pentasilver hexaoxoiodate (Ag5IO6) exhibited variable results. USF2018A was the least effective and similar to the untreated control in two trials (Figure 4-1). However, OSN and

Ag5IO6 was significantly different than the control in three out of four and two out of four trials, respectively (Figure 4-1 and Figure 4-3). OSN application resulted in a brown discoloration of leaves and pitting of the leaf surface when viewed under the microscope in Trials 5 and 6. In the

97 first two greenhouse trials, we used a different batch of OSN supplied by the manufacturer than in the latter two and did not observe any residue on the leaves. ASM applications significantly reduced BST severity by 43 to 62% compared to non-treated plants in five out of six trials. Both,

ASM and copper octanoate applications were significantly different than copper oxychloride treatment in three out of six and one out of two trials, respectively. Further, a significant disease reduction of 9 to 38% and 25% was observed for ASM and copper octanoate, respectively, compared to copper oxychloride. Treatments in which ASM was applied after inoculation were not significantly different than non-treated plants in two trials or copper oxychloride in one trial

(Figure 4-3).

Tank mixtures of Bacillus amyloliquefaciens ‘D747’+ copper octanoate and phosphoric acid + bacteriophage significantly improved disease control as compared to untreated seedlings in only one trial (Figure 1). The combination of bacteriophage or copper octanoate to cymoxanil+famoxadone+phosphoric acid did not significantly reduce disease level compared to cymoxanil+famoxadone+phosphoric acid (Figure 4-2). Applications of cymoxanil+famoxadone+ASM+bacteriophage significantly reduced disease in all four trials compared to the untreated control but not when compared to single treatments of ASM (Figure

4-2 and Figure 4-3). Cymoxanil+famoxadone+ copper octanoate mixtures with either potassium bicarbonate+B. subtilis ‘QST 713’, B. amyloliquefaciens ‘D747’or bacteriophage did not significantly improve disease reduction compared to cymoxanil+famoxadone+copper octanoate in two trials. Whereas, cymoxanil+famoxadone+copper octanoate with either phosphoric acid + bacteriophage or ASM significantly improved disease control in one out of two trials (Figure 4-

3). Cymoxanil+famoxadone+ASM provided a somewhat similar control level to ASM ‘2’ alone or other tank mixture treatments containing ASM.

BST Disease Level in Field-Grown Tomato

During the fall 2015 season, the following treatments: phosphoric acid, oxytetracycline, ammonium chloride, streptomycin+ oxytetracycline, B. amyloliquefaciens ‘D747’+copper octanoate, phosphoric acid + bacteriophage, and cymoxanil+famoxadone+phosphoric acid+copper octanoate did not show any significant improvement in disease control compared to the untreated control. Furthermore, ASM, potassium silicate, OSN, Ag5IO6, cymoxanil+famoxadone+copper oxychloride, cymoxanil+famoxadone+ASM+bacteriophage, cymoxanil+famoxadone+ phosphoric acid + bacteriophage were significantly similar to the grower standard copper oxychloride+mancozeb (Figure 4-4). USF2018A was significantly different from all treatments during during fall 2015 (Figure 4-4). However, USF2018A showed similar control levels to the grower standard copper oxychloride but not to copper oxychloride+mancozeb treatments in fall 2016 trials (Table 4-2).

Disease pressure was slightly lower during fall 2015 compared to fall 2016. Disease severity in untreated plots averaged 35%, 66% and 32% in fall 2015, fall 2016 ‘A’ and ‘B’, respectively (Figure 4-4) (Table 4-2 and 4-3). During fall 2016 all treatments in trial ‘A’ showed a lower disease severity compared to the non-treated control. Two grower standards, copper oxychloride and copper oxychloride+mancozeb, were tested during trial ‘A’ and both were significantly different from the nontreated control. However, copper oxychloride+mancozeb provided the best disease control across all treatments. OSN and USF2018A showed a relatively similar disease level. ASM treatment were significantly different from the non-treated control but did not provide better disease control compared to copper oxychloride alone. Lone applications of B. subtilis ‘QST were not significantly different than tank mixes with copper oxychloride. Streptomycin was not significantly different from oxytetracycline or the grower

99 standard, copper oxychloride, but both antibiotics were significantly different from the nontreated control.

In field trial ‘B’ treatments showed a significant effect based on AUDPC. Compared to the non-treated control, treatments containing a tank mixture of copper octanoate+cymoxanil+famoxadone+phosphoric acid+ bacteriophage, copper octanoate+cymoxanil+famoxadone+phosphoric acid+ASM, copper octanoate+cymoxanil+famoxadone+ASM or ASM were significantly different than the remaining treatments. Combined treatments of copper octanoate +cymoxanil+famoxadone with chemicals such as or bacteriophage, bacteriophage + ASM or B. subtilis ‘QST 713’+potassium bicarbonate were not significantly different from each other. Copper octanoate+cymoxanil+famoxadone+ B. amyloliquefaciens ‘D747’was the least effective treatment in reducing BST. Treatments containing ASM were the most effective except when combined with copper octanoate+bacteriophage+cymoxanil+famoxadone. Further, reducing copper octanoate from weekly sprays to three sprays or increasing bacteriophage application from two to five times did not alter disease levels (Table 4-2). No phytotoxicity was observed with any of the treatments across field trials.

There was no significant difference between treatments for total tomato yield during fall

2016 trials (Table 4-2 and Table 4-3). The number of diseased fruits was very low but was significantly different between treatments. However, there was a significant difference for diseased fruit in one trial during fall 2016 (Table 4-2). The incidence of diseased fruit was very low and ranged between 0.1 to 1.1% during both fall 2016 trials. Streptomycin, oxytetracycline and B. subtilis ‘QST 713’ treatments had the highest and lowest diseased fruits, respectively.

100

Discussion

In this study, we evaluated a number of different non-copper based products, such as biological agents, plant defense activators, antibiotics, detergents and several experimental compounds for the control of BST in transplants and field-grown tomato. Most of the compounds used in this study are currently not registered or not labeled for controlling BST on tomato seedlings. The efficacy of all the treatments varied within and between greenhouse and field settings. Environmental conditions play a major role in how these products can be efficient in controlling disease. BST levels in the field are more variable than inside the greenhouse due to variable environmental conditions which can favor or curb disease development. BST management relies heavily on copper and copper-mancozeb tank mixes which are not very effective due to increasing copper tolerance among BST populations (Ritchie and

Dittapongpitch, 1991). Copper and copper-mancozeb sprays are considered as the grower standard. However, based on recent surveys bacterial populations throughout Florida are highly tolerant (Horvath et al., 2012; Timilsina et al., 2013). Griffin et al. (2017) showed that around

54% of Xanthomonas strains tested in previous studies were copper tolerant. Copper sprays in our trials showed to be inconsistent under greenhouse and field conditions. Nevertheless, copper applications reduced disease under both conditions even when using copper tolerant strains.

Although in vitro assays show almost complete resistance to copper, in vivo assays show otherwise, which might indicate that Xp tolerance to copper might be more complex than expected. This has been observed in other field trials where BST caused by copper tolerant and sensitive strains showed a 32% and 76% disease reduction (Pernezny et al., 2008; Roberts et al.,

2008).

101

Effective alternatives to copper-based compounds were identified for use on tomato seedlings such as ASM, copper octanoate, OSN and quinoxyfen as stand-alone treatments. ASM and copper octanoate were effective alone as well as when either product was included in tank- mixes. ASM, a plant-defense activator, was one of the most effective chemicals for BST control in our transplant and field studies. ASM has been extensively tested as an alternative for copper- mancozeb sprays in tomato field production (Huang et al., 2012; Itako et al., 2014; Louws et al.,

2001; Roberts et al., 2008; Vallad and Goodman, 2004). ASM gave superior control either alone or in combination with other products in tomato seedlings. Further, a single application of ASM as a stand-alone product showed sufficient control of BST for a period of two to three weeks compared to other products. Similar results showed that ASM applications alone provided superior control of copper-sensitive X. vesicatoria than in combinations with copper, agriphages or Bacillus spp. (Briceno-Montero and Miller, 2004). B. subtillis QST713 and B. amyloliquefaciens ‘D747’ were generally ineffective when applied alone but were effective when combined with well performing products such as copper octanoate. The better performance of tank mixes containing Bacillus sp. is likely due to the superior performance of the added product in the tank mix. Similarly, previous studies showed the inefficacy of B. subtillis QST713 when applied alone compared to tank mixes with copper hydroxide (Abbasi and Weselowski,

2015). Further, weekly applications of ASM alone were similar or more effective than in combination with other products in field trials. Weekly applications of ASM were significantly better at reducing disease compared to biweekly applications (Huang et al., 2012). The consistent efficacy of ASM in reducing disease is likely due to elicitation of a systemic resistance response.

Quinoxyfen, a fungicide used for suppression of BST, showed phytotoxicity at rates lower than the labeled field rate. Although reduced rates of quinoxyfen also improved disease

102 control, multiple applications still caused some plant stunting. Copper octanoate, a copper-based compound, showed good to excellent control in multiple trials. Nevertheless, the formulation of copper octanoate has only 1.8% of metallic copper compared to 30% of metallic copper to that in the grower standard treatment. The control of a copper tolerant strains of Xp might be improved by an enhanced copper formulations, such as with copper octanoate that is labeled as a copper soap. Recently, Strayer-Scherer et al. (2017) showed that enhanced nano-copper formulations using 80% less metallic copper enhanced the control of copper-resistant strains in vitro and in vivo. Reducing the size of copper particle from micrometer to nanometers significantly improves bacterial control (Strayer-Scherer et al. 2017). However, the particle size of copper in the copper octanoate product is unknown. Although, copper octanoate was effective under greenhouse conditions, it did not adequately control BST in the field. The inefficacy of copper octanoate might be due to environmental conditions, or due to a diverse epiphytic bacterial population. The addition of cymoxanil+famoxadone to copper octanoate did not improve control. Prior studies showed that tank mixes of cymoxanil+famoxadone did not reduce disease or bacterial populations in comparison to non-treated tomato plants under greenhouse conditions although combination of cymoxanil+famoxadone with copper hydroxide significantly reduced disease

(Fayette et al., 2012).

Antibiotics were initially used for controlling BST, but resistance developed faster than observed with copper products due to the higher selective pressure on bacterial population (Stall and Thayer, 1962). In our study, antibiotics, such as streptomycin and oxytetracycline, were ineffective. Streptomycin is labeled for use on tomato, but its efficacy in controlling BST varied due to the partial resistance of our Xp population. Antibiotic resistance to streptomycin is an issue in Florida but has not been reported to be problematic elsewhere (Bouzar et al., 1999;

103

Osdaghi et al., 2016; Kebede et al., 2014). On the other hand, oxytetracycline is used for control of bacterial spot of peaches and nectarines, but not vegetables and was ineffective on seedlings or field plants. Bacteriophage applications were not effective except in combination with other treatments that included either copper octanoate or ASM. Similarly, prior studies reported that bacteriophage applications on tomato seedlings did not control X. vesicatoria in tomato seedlings

(Briceno-Montero and Miller, 2004). Bacteriophage performance in the field was much lower than greenhouse conditions and this might be correlated with environmental conditons such as

UV-rays and leaf wetness. However, formulation and harsh environmental conditions (ultra- violet rays, high temperatures) play a major role in phage survival on the leaf surface (Jones et al., 2012). Phage survival rapidly drops in relation to UV intensity with the highest efficacy being observed with evening applications (Jones et al. 2007; Obradovic et al., 2005). The application of phosphoric acid salt was ineffective on tomato seedlings when applied alone, but was effective when combined with other treatments such as ASM, copper octanoate or bacteriophage. Similar observations were reported when phosphoric acid salts were combined with copper-mancozeb applications in seedlings, as well as the varying efficacy associated with phosphoric acid salts in field studies (Wen et al., 2009).

The experimental compounds tested in this study had variable efficacy in controlling BST on tomato seedlings and field plants. However, OSN, reduced disease to very low levels. Spray applications of OSN resulted in a brown discoloration of leaf surfaces which was observed in all greenhouse trials. This phytotoxicity might be attributed to the high oxidation state of silver ions in the formulation. Similar phytotoxic reactions have been obsereved with the use of silver nanoparticles when sprayed at rates greater than 200 µg/ml (Paret et al., 2013). Nevertheless, we did not observe phytotoxicity with our silver-based products under field conditions. OSN activity

104 in the field was not as good as observed in our seedling studies and this might be explained by using a low application rate for field-grown tomatoes. OSN is registered as Agress® as a seed treatment against fungal and bacterial diseases of row crops, but not yet widely marketed. OSN is formulated in a high oxidation state in which silver exists in three valent forms Ag (I), Ag(II) and Ag(III) (Nadworny et al., 2015). The high valency of silver has strong antimicrobial activity compared to low valency silver formulations (Rai et al., 2012). On the other hand, Ag5IO6 contains silver in only one valent state Ag (I), but also contains a highly oxidized iodine (Incani et al., 2015). Ag5IO6 was not effective compared to OSN against BST in tomato seedlings likely due to its lower solubility in water that led to the rapid precipitation of the product in the tank.

Recent studies, have shown that using silver nanocomposite particles can significantly reduce bacterial spot severity (Paret et al., 2013; Strayer et al., 2015). In addition, Ag5IO6 was shown to have antimicrobial activity against biofilms of human microbial pathogens (Incani et al., 2015).

In this study we observed that several products that are efficient on seedlings are not effective on field tomato and vice versa. For instance, copper octanoate and tank mixes of copper octanoate with bacteriophage or phosphoric acid provided adequate disease control in seedling trials but not in field trials. This is important for pesticide manufacturers for evaluating products that are tested only in vitro or under controlled environmental conditions and overlooking actual field conditions. Furthermore, due to the lack of a reliable and consistent product for controlling

BST during field production, we suggest that disease control should focus on transplant production to minimize the introduction of Xp to the field. Seedlings are more sensitive and prone to disease than full grown tomato and are the primary source of introducing Xp inoculum into the field. Transplant seedlings may already have a variable amount of epiphytic

Xanthomonas prior to conducive weather conditions (Pernezny and Collins, 1997; P.

105

Abrahamian, unpublished data). Outbreaks of BST were observed in tomato seedlings in commercial transplant operations (P.Abrahamian and G.E. Vallad, unpublished data). Further, another possibility for the higher disease in treated plots is due to spraying. Handling or spraying plants with copper-mancozeb in the field increased BST spread due to increased aerosolization

(McInnes et al., 1988). Therefore, it is possible that the cause of the higher disease pressure in treated plots may be attributed to the introduction of high epiphytic bacterial populations with transplants and exacerbated Xp movement during spraying. Under field conditions, we observed a very low disease incidence on fruits. Xp is not highly virulent on fruits compared to other bacterial spot species (Potnis et al., 2015).

The lack of a ‘silver bullet’ product for BST control remains an obstacle in tomato production. In this study we identified several effective products applied alone or in combination with other products. We suggest copper octanoate as a supplemental product to ASM applications in spray programs for controlling BST in tomato seedlings and ASM alone for field production. Incorporation of ASM in integrated disease management programs is favorable as there is a little risk of developing resistance in bacterial populations. Also, silver-based products are potential alternatives in the future; however, further testing on tomato and enhanced product formulations is required to reduce phytotoxicity and particle aggregation. Nevertheless the usage of any effective product should be judiciously used to reduce the risk of developing resistance within the pathogen population. Integrated disease management should rely on several strategies such as cultural control, chemical control and the use of tolerant varieties.

106

Table 4-1. List of chemical product names, active ingredients, manufacturers used in this study. Product name Active ingredient Manufacturer Actigard® acibenzolar-s-methyl (ASM) Syngenta Agress® oxysilver nitrate (OSN) Innovotech Inc. ™ AgreGuard -1 pentasilver hexaoxoiodate Innovotech Inc. (Ag5IO6) Agri-mycin® 17 Streptomycin Nufarm ™ AgriPhage bacteriophage OmniLytics, Inc. Cueva™ copper octanoate Certis USA ® Cuprofix Ultra 40D copper sulfate United Phosphorous, Inc. ™ Double Nickel 55 Bacillus amyloliquefaciens Certis USA ‘D747’ ™ KleenGrow ammonium chloride Pace 49, Inc. Kocide® 3000 copper hydroxide DuPont ® K-Phite mono- and di- sodium Plant Food Systems, phosphoric acid Inc. Mycoshield® oxytetracycline Nufarm ® Milstop pottasium bicarbonate BioWorks Inc. (KHCO3) ® Penncozeb 75DF mancozeb United Phosphorous, Inc. ® Quintec quinoxyfen Dow AgroSciences Sil-matrix™ potassium silicate Certis USA ® Serenade Opti Bacillus subtilis ‘QST 713’ Bayer Crop Science Tanos® famoxodone, cymoxanil DuPont a USF2018A - Bayer Crop Science a Proprietary active ingredient.

107

Table 4-2. Tomato yield, average yield and percent of fruit diseased with bacterial spot per hectare for single treatments application during fall 2016 ‘A’. Yield Diseased fruit AUDPC Treatment, rate/1L (appl. no.)z (Kg/Ha) Kg/Ha % 1 Non-treated control 1492 ay 37575 265.9 ab 0.7 2 OSN, 1 g (1-6) 1391 bc 36736 143.2 bc 0.4 3 USF2018A 200SC, 0.26 ml (1-6) 1367 ab 37084 163.6 c 0.4 4 OTC, 1.16 g (1,3,5) ; Cu(OH)2, 2.11 g (2,4,6) 1349 bc 36736 388.6 ab 1.1 5 OTC, 1.16 g (1-6) 1347 bc 33995 245.5 bc 0.7 6 Streptomycin, 1.16 g (1-6) 1315 bcd 40705 81.8 bc 0.2 7 Cu(OH)2, 2.11 g (1-6) 1307 bcd 39457 429.5 a 1.1 8 Bsub713, 1.8 g (1-6) 1283 bcd 38843 143.2 bc 0.4 9 ASM 0.06 g (1-6) 1270 bcd 34425 209.7 bc 0.6 10 Cu(OH)2, 2.11 g (1,3,5); Bsub713, 1.8 g (2,4,6) 1246 cd 38209 184.1 bc 0.5 11 Ag5IO6, 1.4 g (1-6) 1222 d 34895 122.7 bc 0.4 12 Cu(OH)2, 2.11 g (1-6); mancozeb, 1.2 g (1-6) 1044 e 39559 184.1 bc 0.5 P-value < 0.0001 0.9721 0.0407 z Listed treatment rates are on a per 100 gal basis unless noted otherwise; foliar treatments were applied on 28 Sept, 6 Oct, 11 Oct, 18 Oct, 24 Oct, and 31 Oct (corresponding to applications 1- 6). Plants were inoculated with a mixture of streptomycin sensitive and resistant strains and all strains were copper tolerant. Abbreviations: ASM: acibenzolar-s-methyl; Bsub713: Bacillus subtilis ‘QST 713’; Cu(OH)2: copper hydroxide; CuOct: copper octanoate; Fam+Cym: famoxodone, cymoxanil; OSN: Oxysilver nitrate; OTC: Oxytetracycline; Ag5IO6: pentasilver hexaoxoiodate. y Means followed by the same letter are not significantly different according to Fisher’s LSD test (α = 0.05).

108

Table 4-3. Tomato yield, average yield and percent of fruit diseased with bacterial spot per hectare for combined and alternating spray treatment application during fall 2016 ‘B’. Diseased fruit z Treatment, rate/1 L (appl. no.) AUDPC Yield(Kg/Ha) Kg/Ha %

1 Non-treated control 1020 fy 6952.2 78.3 1.1 2 CuOct 5 ml (1, 4, 7); Fam+Cym 0.6 g (1, 4, 7); Bamylo747 0.6 g (2, 3, 5, 6, 8) 1475 a 5863.0 26.1 0.4 3 CuOct 5 ml (1-8) ; Fam+Cym 0.6 g (1, 3, 5, 7) 1322 b 6710.9 39.1 0.6 4 CuOct 5 ml (1, 4, 7); Fam+Cym 0.6 g (1, 4, 7); Bsub713 0.45 g (2, 3, 5, 6, 8); KHCO3 2.4 g (2, 3, 5, 6, 8) 1295 b 5073.9 6.5 0.1 5 CuOct 5 ml (1, 4, 7); Fam+Cym 0.6 g (1, 4, 7); bacteriophage 5 ml (3, 6); ASM 0.02 g (1- 3); ASM 0.03 g (4-6); ASM 0.05 g (7, 8) 1260 bc 5400.0 32.6 0.6 6 CuOct 5 ml (1-8) ; Fam+Cym 0.6 g (1, 4, 7) 1252 bc 7193.5 65.2 0.9 7 CuOct 5 ml (1, 4, 7); Fam+Cym 0.6 g (1, 4, 7); Bacteriophage 5 ml (2, 3, 5, 6, 8) 1232 bcd 6691.3 26.1 0.4 8 CuSO4, 2.4 g (1-8); mancozeb 1.2 g (1-8) 1221 bcd 5289.1 13.0 0.2 9 CuOct 5 ml (1, 4, 7); Fam+Cym 0.6 g (1, 4, 7); SPA 7.5 ml (2,5,8); bacteriophage 5 ml (3, 6) 1133 cde 7500.0 26.1 0.3 10 CuOct 5 ml (1-8); Fam+Cym 0.6 g (1, 4, 7); ASM 0.02 g (1-3); ASM 0.03 g (4-6); ASM 0.05 g (7, 8) 1123 de 6293.5 32.6 0.5 11 CuOct 5 ml (1, 4, 7); Fam+Cym 0.6 g (1, 4, 7); SPA 7.5 ml (2, 5, 8); ASM 0.02 g (1-3); ASM 0.03 g (4-6); ASM 0.05 g (7, 8) 1078 ef 5908.7 26.1 0.4 12 ASM 0.02 g (1-3); ASM 0.03 g (4-6); ASM 0.05 g (7, 8) 1013 f 5126.1 13.0 0.3 P-value < 0.0001 0.3587 0.8026 z Listed treatment rates are on a per 100 gal basis unless noted otherwise; foliar treatments were applied on 26 Aug, 9 Sep, 15 Sep, 23 Sep, 28 Sep, 6 Oct, 12 Oct, and 20 Oct (corresponding to applications 1-8). Plants were inoculated with a mixture of streptomycin sensitive and resistant strains and all strains were copper tolerant. Abbreviations: ASM: acibenzolar-s-methyl; Bamylo747: Bacillus amyloliquefaciens ‘D747’; Bsub713: Bacillus subtilis ‘QST 713’; CuOct: copper octanoate; CuSO4: copper sulfate; Fam+Cym: famoxodone, cymoxanil; KHCO3: pottasium bicarbonate; SPA: sodium phosphoric acid. y Means followed by the same letter are not significantly different according to Fisher’s LSD test (α = 0.05).

109

Figure 4-1. Treatments evaluated for controlling bacterial leaf spot in the greenhouse during fall 2015 (trial 1: A; trial 2: B). ASM was applied once before inoculation with Xanthomonas perforans. Quinoxyfen was tested at three rates (0.4, 0.23 and 0.08 ml/L) and only the lowest rate is shown in B due to phytotoxicity at higher rates. Means followed by the same letter are not significantly different according to Fisher’s LSD test (α = 0.05) within the same trial. Abbreviations used in the figure: acibenzolar-S-methyl, ASM; copper hydroxide, Cu hydroxide; copper octanoate, Cu oct; famoxodone, cymoxanil, Fam,cym; oxysilver nitrate, OSN; pentasilver hexaoxoiodate, Ag5IO6.

110

Figure 4-2. Treatments evaluated for controlling bacterial leaf spot in the greenhouse during fall 2015 (trial 3: A; trial 4: B). Copper octanoate as a single treatment was not tested during trial 4. Acibenzolar-S-methyl (ASM) was applied once before inoculation with Xanthomonas perforans. Quinoxyfen was tested at 0.08 ml/L in both trials. Means followed by the same letter are not significantly different according to Fisher’s LSD test (α = 0.05) within the same trial. Abbreviations used in the figure: acibenzolar-S-methyl, ASM; Bacillus amyloliquefaciens, B. amylo; copper hydroxide, Cu hydroxide; copper octanoate, Cu oct; famoxodone, cymoxanil, Fam,cym; phosphoric acid, P-acid.

111

Figure 4-3. Treatments evaluated for controlling bacterial leaf spot in the greenhouse during fall 2016 (A: trial 5) and spring 2017 (B: trial 6). ASM ‘1’ was applied once after inoculation with Xanthomonas perforans whereas ASM ‘2’ was applied once prior to bacterial inoculation. Means followed by the same letter are not significantly different according to Fisher’s LSD test (α = 0.05) within the same trial. Abbreviations used in the figure: acibenzolar-S-methyl, ASM; Bacillus subtilis, B. sub; Bacillus amyloliquefaciens, B. amylo; copper hydroxide, Cu hydroxide; copper octanoate, Cu oct; famoxodone, cymoxanil, Fam,cym; oxysilver nitrate, OSN; pentasilver hexaoxoiodate, Ag5IO6; phosphoric acid, P-acid; potassium bicarbonate, KHCO3.

112

Figure 4-4. Evaluation of chemical treatments on field tomato for controlling bacterial leaf spot of tomato during fall 2015 using single compounds and tank mixes. Means followed by the same letter are not significantly different according to Fisher’s LSD test (α = 0.05). Abbreviations used in the figure: acibenzolar-S-methyl, ASM; Bacillus amyloliquefaciens, B. amylo; copper hydroxide, Cu hydroxide; copper octanoate, Cu oct; famoxodone, cymoxanil, Fam,cym; oxysilver nitrate, OSN; phosphoric acid, P- acid.

113

CHAPTER 5 SUMMARY AND DISCUSSION

Tomato is a major vegetable commodity in the United States. Florida plays a major role in fresh market tomato production as well as the commercial production of transplants (Gitaitis et al., 1992). However, tomatoes, both transplant and field plants, are affected by many diseases of which those caused by bacteria are the most important due to Florida’s conducive subtropical climate. Outbreaks of BST remains a major hurdle for tomato transplant operations and field production in Florida. Xp appeared in Florida in the early 1990’s. Since then, the Xp population has changed dramatically and resulted in acquisition of effector and antibiotic resistance genes, recombination and race shifts. As a result, the management of BST became increasingly difficult.

Xp is getting more established across the globe. Xp was reported for the first time in Louisiana

(Lewis Ivey et al., 2016). A recent survey in Brazil, showed that 97.5 % of the strains recovered belong to Xp race T3 (Araújo et al., 2017). Race T4 strains were reported for the first time in

Brazil at a very low incidence (Araújo et al., 2017). Race T4 strains were also recovered in another recent survey in Ethiopia (Kebede et al., 2014). The capability of Xp to easily establish itself as a pathogen is worrisome as evident by Xp’s ability to displace other Xanthomonas spp.

(Timilsina et al., 2016; Tudor-Nelson et al., 2003). In this study, we studied several aspects of host-pathogen-environment interaction with regards to Xp fitness, genetics and response to novel chemical alternatives.

T4 was shown to be the dominant race of Xp in Florida (Timilsina et al., 2016). However, when the population structure was examined closely, certain pathogenicity factors (e.g. effectors) were lost and gained simultaneously (Schwartz et al., 2015). This was also shown in the closely related X. euvesicatoria (Barak et al., 2016). The introduction of effectors into a population is generally attributed to horizontal gene transfer and recombination with closely-related species

114

(White et al., 2009). AvrBsT is a highly conserved effector ubiquitously present in many strains

(Timilsina et al., 2016). Although the origin of avrBsT is unknown, its origin is likely attributed to horizontal gene transfer from X. vesicatoria or X. euvesicatoria to Xp. This effector is a member of a unique superfamily that has a conserved cellular function of acetylation (Lewis et al., 2011). Xp strains do not possess an avrBsT-homolog with a similar function within the same bacterial cell. However, avrBsT-homologs across species or strains exist and are well- characterized. Thus, we carried out experiments to evaluate the fitness benefit or penalty of acquiring AvrBsT in the Xp population in planta in greenhouse and field studies. In planta experiments showed that Xp strains lacking avrBsT do not incur a fitness cost with regards to population growth rates. Furthermore, co-infiltration of WT and ΔavrBsT strains to determine in planta competition between strains showed no major differences. While this is not an odd observation, since members of this effector family tend to not influence bacterial populations in planta. Nevertheless, when ΔavrBsT strains were deployed in the field we saw tremendous fitness effects compared to WT strains. This fitness effect was mainly manifested in a reduction of the epiphytic population of mutant strains and ability to spread in the field. The impact of understanding the effect of AvrBsT is important primarily for breeding purposes. Breeding for resistance is a difficult task due to the continuous host-pathogen arms race (Stall et al., 2009).

Based on the results in our study, AvrBsT is a good target to reduce bacterial populations and their dissemination. The targeting of AvrBsT and other conserved effectors, such as AvrBs2 and

AvrXv4, is recommended in a resistance-breeding program to reduce the risk of resistance breakdown. Pyramiding genes is important in order to reduce the risk of selecting for bacterial strains containing mutated effectors that renders plant resistance ineffective.

115

In the second part of our study, we attempted to find a link often overlooked in the movement of Xp during tomato production. Field-grown tomatoes are commonly initiated as transplants. Florida is a major supplier of tomato seedlings for growers throughout the Southeast

U.S. BST outbreaks are also common in transplant operations and in turn these seedlings are a potential source of inoculum for field outbreaks. Therefore, we utilized genomic tools to show a connection in transplant houses and field tomatoes by tracking plants from the former to the latter throughout the growing season. We sampled BST from two major commercial transplant operations and field growers in Florida during 2015 and 2016. Strains were isolated and their draft genome sequenced. We used several methods such as MLSA, cgMLST and SNPs to determine similarities between strains. MLSA was not enough to show an association between transplant and field strains, however two distinct lineages were observed. According to MLSA, the majority of the strains collected belonged to group 2 and the remaining in group 1. We used cgMLST to refine our output. We found 886 core genes similar across all our sequenced, type and reference strains. The cgMLST showed that most strains indeed belonged in group 2, but none belonged in group 1, while a new group similar to Xp17-12 was identified. Xp 17-12 is a strain isolated from tomato in 2006 from Florida and shows some distinct features from the general Xp population. Based on cgMLST, we a found few field strains that were genetically similar to transplant strains. However, a large number of field strains did not cluster with transplant strains and vice versa. Based on cgMLST we could not establish a strong link between transplant and field strains. cgMLST has its limitations. While cgMLST includes a large subset of genes similar across all strains, a genome-wide SNP analysis utilizes more sequencing data and covers coding and non-coding regions. Maiden et al. (2013) suggested that SNP analysis is a suitable method to identify differences between isolates of low genetic diversity. Based on ANI

116 analysis we observed very low genetic diversity within our Xp population. ANI analysis showed that transplant and field strains were 99 to 100% similar within strains collected from each grower and when comparing the different grower strains to each other. Therefore, we utilized a genome-wide SNP-based analysis to determine clonal lineages among strains. This method has been pioneered in the medical field for sourcing outbreaks in healthcare-associated outbreaks, food-borne illnesses and biological weapons (Schürch and Siezen, 2010). Our strain grouping

SNP analysis was in line with the general lineages produced by cgMLST. However, unlike cgMLST, SNP-based analysis was able to discriminate between closely related strains. With a few exceptions, most field strains were highly similar to transplant strains based on genetic distances. Those field strains that did not cluster with any transplant strains and had very high genetic distances were most likely introduced during the growing season. Field plants throughout the growing season are handled continuously (e.g. tying, spraying). Therefore, strains can possibly move from one field to another due to typical grower operations. Furthermore, we sampled once during the growing season towards harvesting and thus we might have sampled strains introduced towards the end of the season. Our sample size was likely not sufficient to include all available transplant and field strains. However such a task is not reasonable and feasible. Nevertheless, our results suggest that a strong link exists between BST field outbreaks and transplant seedlings.

Management of BST relies heavily on chemical sprays due to the lack of commercially resistant tomato cultivars. Although transgenic tomato cultivars expressing Bs2 resistance gene from pepper were developed, public acceptance remains a problem (Horvath et al., 2012). Most bacterial diseases are usually controlled with copper or antibiotics. Copper and copper-mancozeb sprays are the grower standard for managing BST. Nevertheless, the reliance on copper for

117 several decades resulted in a complete bacterial tolerance resulting in ineffective sprays. Also, antibiotics are effective but not ideal as resistance development is rapid and inevitable. The lack of labeled alternatives prompted us to evaluate 19 different chemicals including the grower copper-mancozeb standard. The efficacy of these chemicals was evaluated for managing BST in tomato seedlings for transplant operations and in field-grown plants. ASM is gaining wide preference for managing BST and was also included. ASM was one of the most efficient chemicals in reducing BST severity in tomato seedlings. However, other compounds such as copper octanoate were also efficient in reducing disease alone or in combination with ASM or other compounds. Oxytetracycline was generally ineffective against Xp, on the other hand, streptomycin was inconsistent in its efficacy. Antibiotics are better off avoided due to rapid resistance build-up. We recommend the use of copper octanoate and ASM as a first line of controlling BST outbreaks in transplant houses. Also, compounds such as OSN and quinoxyfen provide good control, but the possibility of phytotoxicity is a concern. The efficacy of spraying compounds on seedlings was much higher in reducing disease than in field plants. Nevertheless,

ASM showed to be a reliable alternative. Further studies should be conducted to determine the effect of planting transplants treated with compounds on BST severity in the field and on the efficacy of such compounds in reducing BST movement in a transplant house.

118

APPENDIX A ENVIRONMENT AND STATISTICS SUPPLEMENTARY TABLES

Table A-1. Average daily rainfall, relative humidity (RH) and temperature for the time period of each trial.

Rainfall (mm) RH (%) Temp (°C) Fall 2015a Aug 4.95 89.0 26.1 Sept 3.09 86.8 26.5 Oct 0.80 85.9 24.1

Spring 2016b March 2.27 79.4 21.6 April 1.89 72.7 22.3 May 9.08 74.1 23.6

Spring 2017c April 2.49 67.5 23.1 May 1.46 72.9 24.8 Jun 8.01 88.2 24.8 a Trial period from 8/26/2015 to 10/14/2015. b Trial period from 4/15/2016 to 5/17/2016. c Trial period from 4/3/2017 to 6/13/2017.

119

Table A-2. Type III tests of fixed effects for total bacterial populations levels of Xanthomonas perforans from in planta infiltrations in greenhouse trials. Effect Num DF Den DF F Value P>F Xp GEV872 Treatmenta 1 20 1.81 0.1931 Time 4 20 289.6 0.0001 Treatment*Time 4 20 3.11 0.0384

Xp GEV1001 Treatmenta 1 10 0.64 0.4409 Time 4 8 291.42 0.0001 Treatment*Time 4 10 3.84 0.0383 a Treatment effect includes comparison between wild-type and mutant strains of the designated Xanthomonas perforans isolate.

120

Table A-3. Type III tests of fixed effects for colony recovery of Xanthomonas perforans GEV1001 from co-infiltration leaf assays in greenhouse trials. Effecta Num DF Den DF F Value P>F Treatment 1 8 7.91 0.0227 Time 3 8 0.58 0.6437 Treatment*Time 3 8 0.22 0.8783 a Treatment effect includes comparison between GEV1001 wild-type and mutant strains.

121

Table A-4. Type III tests of fixed effects for total bacterial populations levels recovered from field plants in spring 2016. Effect Num DF Den DF F Value P>F Xp GEV872 Plant distance 14 411 8.22 0.0001 Time 6 411 59.90 0.0001 Plant distance*Time 84 411 1.01 0.4704

Xp GEV1001 Plant distance 14 414 5.71 0.0001 Time 6 414 20.43 0.0001 Plant distance*Time 84 414 0.93 0.6522

122

Table A-5. Type III tests of fixed effects for total bacterial populations recovered from field plants in spring 2017. Effect Num DF Den DF F Value P>F Xp GEV872 Plant distance 14 456 12.36 0.0001 Time 7 456 89.21 0.0001 Plant distance*Time 98 456 1.28 0.0516

Xp GEV1001 Plant distance 14 457 4.20 0.0001 Time 7 457 60.17 0.0001 Plant distance*Time 98 457 0.71 0.9787

123

Table A-6. Type III tests of fixed effects for movement of Xanthomonas perforans GEV872 in field trials. Effecta Num DF Den DF F Value P>F Fall 2015 Treatment 1 52 0.07 0.7909 Time 6 52 11.58 0.0001 Treatment*Time 6 52 1.11 0.3720 Spring 2016 Treatment 1 52 3.18 0.0802 Time 6 52 1.68 0.1443 Treatment*Time 6 52 0.60 0.7254 Spring 2017 Treatment 1 76 8.44 0.0048 Time 9 76 9.46 0.0001 Treatment*Time 9 76 0.42 0.9205 a Treatment effect includes comparison between GEV872 wild-type and mutant strains.

124

Table A-7. Type III tests of fixed effects for movement of Xanthomonas perforans GEV1001 in field trials. Effecta Num DF Den DF F Value P>F Fall 2015 Treatment 1 52 14.13 0.0004 Time 6 52 11.34 0.0001 Treatment*Time 6 52 1.46 0.2092 Spring 2016 Treatment 1 52 5.47 0.0232 Time 6 52 2.36 0.0433 Treatment*Time 6 52 1.31 0.2711 Spring 2017 Treatment 1 76 7.30 0.0085 Time 9 76 12.52 0.0001 Treatment*Time 9 76 0.90 0.5316 a Treatment effect includes comparison between GEV1001 wild-type and mutant strains.

125

Table A-8. Type III tests of fixed effects for colony recovery of Xanthomonas perforans GEV872 in field trials. Effecta Num DF Den DF F Value P>F Fall 2015 Treatment 1 35 2.73 0.1075 Time 5 35 9.11 0.0001 Treatment*Time 5 35 2.29 0.0665 Spring 2016 Treatment 1 51 59.33 0.0001 Time 6 51 7.89 0.0001 Treatment*Time 6 51 1.54 0.1847 Spring 2017 Treatment 1 50 14.87 0.0003 Time 9 50 13.14 0.0001 Treatment*Time 8 50 1.98 0.0689 a Treatment effect includes comparison between GEV872 wild-type and mutant strains.

126

Table A-9. Type III tests of fixed effects for colony recovery of Xanthomonas perforans GEV1001 in field trials. Effecta Num DF Den DF F Value P>F Fall 2015 Treatment 1 28 15.97 0.0004 Time 5 28 3.73 0.0103 Treatment*Time 5 28 0.91 0.4876 Spring 2016 Treatment 1 52 34.53 0.0001 Time 6 52 3.20 0.0095 Treatment*Time 6 52 0.53 0.7861 Spring 2017 Treatment 1 47 29.96 0.0001 Time 7 47 10.44 0.0001 Treatment*Time 7 47 5.26 0.0002 a Treatment effect includes comparison between GEV1001 wild-type and mutant strains.

127

APPENDIX B ADDITIONAL STRAIN CHARACTERIZATION DETAILS

Table B-1. Additional Xanthomonas strains used in this study. Strain Year Location Reference Xanthomonas perforans Xp1-7 Xp2-12 Xp3-15 Xp4-20 Xp5-6 Xp7-12 Xp8-16 Xp9-5 2006 Xp10-13 Xp11-2 Xp15-11 Xp17-12 Xp18-15 Xp19-10 Xp4B GEV839 GEV872 Florida Schwartz et al., 2015 GEV893 GEV904 GEV909 GEV915 GEV917 GEV936 2012 GEV940 GEV968 GEV993 GEV1001 GEV1026 GEV1044 GEV1054 GEV1063 TB6 TB9 2013 TB15 NI1 2015 Ethiopia Timilsina et al., 2015 91-118 1991 Florida Astua-Monge et al., 2000 X. vesicatoria ATCC35937 1955 New Zealand ATCC X. euvesicatoria 85-10 1985 Florida Minsavage et al., 1990 X. gardneri ATCC19865 1953 Yugoslavia ATCC

128

Table B-2. Genome details of sequenced strains collected from grower A. Strain Size (bp) Coverage Genes GC% Contigs GEV1989 5201369 38.427 4632 64.7 247 GEV1991 5192752 33.85 4611 64.7 238 GEV1992 5191666 38.14 4642 64.7 262 GEV1993 5155078 33.682 4622 64.69 279 GEV2004 5057458 18.884 4646 64.66 425 GEV2009 5136944 43.905 4620 64.71 324 GEV2010 5174097 29.26 4585 64.67 264 GEV2011 5115304 34.18 4542 64.71 261 GEV2013 5135549 28.585 4561 64.68 345 GEV2015 5144994 35.359 4576 64.7 250 GEV2047 5185550 26.688 4677 64.69 304 GEV2048 5177283 47.385 4655 64.69 304 GEV2049 5167720 40.976 4660 64.7 302 GEV2050 5165201 41.52 4639 64.7 281 GEV2052 5190899 36.7 4619 64.69 249 GEV2055 5177880 39.08 4696 64.7 368 GEV2058 5210673 56.96 4633 64.71 233 GEV2059 5179416 45.92 4686 64.69 331 GEV2060 5183330 45.644 4699 64.71 359 GEV2063 5128950 31.09 4608 64.7 279

129

Table B-3. Genome details of sequenced strains collected from grower B. Strain Size (bp) Coverage Genes GC% Contigs GEV1911 5137800 21.73 4681 64.67 404 GEV1912 5167005 30.35 4598 64.7 253 GEV1913 5183227 25.47 4678 64.68 340 GEV1914 5113830 25.9 4671 64.61 409 GEV1915 5174692 25.2 4695 64.61 361 GEV1916 5102395 30.32 4591 64.67 311 GEV1917 5161255 30.76 4601 64.67 267 GEV1918 5092768 29.66 4548 64.69 255 GEV1919 5186082 46.14 4658 64.68 286 GEV1920 5147727 25.43 4647 64.61 330 GEV1921 5172196 29.37 4663 64.71 333 GEV2065 5129485 31.28 4603 64.64 328 GEV2067 5213895 32.66 4635 64.64 248 GEV2072 5203120 34.32 4622 64.68 240 GEV2087 5122760 34.58 4582 64.67 272 GEV2088 5139743 28.97 4643 64.69 315 GEV2089 5102991 28.06 4623 64.67 328 GEV2097 5129171 32.19 4560 64.68 299 GEV2098 5174608 29.59 4661 64.7 323 GEV2099 5110801 30.99 4550 64.71 303 GEV2108 5105539 37.75 4566 64.71 345 GEV2109 5200764 40.2 4661 64.67 277 GEV2110 5120298 42.92 4567 64.77 237 GEV2111 5100511 52.07 4481 64.81 181 GEV2112 5089796 27.45 4530 64.64 301 GEV2113 5163947 25.24 4671 64.62 345 GEV2114 5172135 28.94 4648 64.61 314 GEV2115 5179887 29.46 4692 64.66 377 GEV2116 5153501 25.64 4620 64.67 299 GEV2117 5070119 26.75 4539 64.8 272 GEV2118 5161403 26.18 4654 64.68 314 GEV2119 5114562 23.92 4580 64.66 294 GEV2120 5196625 25.85 4647 64.62 276 GEV2121 5079594 21.69 4532 64.64 286 GEV2122 5127626 22.04 4544 64.64 235 GEV2123 5002536 19.8 4592 64.63 420 GEV2124 5031454 25.9 4426 64.71 221 GEV2125 4944651 14.02 4591 64.66 525 GEV2126 5089637 30.01 4475 64.79 204 GEV2127 5031736 22.31 4490 64.69 296 GEV2128 5129344 33.19 4517 64.69 247

130

Table B-3. Continued. Strain Size (bp) Coverage Genes GC% Contigs GEV2129 5145947 24.72 4585 64.66 256 GEV2130 5120789 33.1 4480 64.71 195 GEV2132 5059184 33.19 4452 64.8 193 GEV2133 5149660 32.43 4551 64.71 185 GEV2134 5106930 27.59 4513 64.67 221 GEV2135 5092715 28.1 4474 64.81 175

131

Table B-4. Average nucleotide identity showing a one-way comparison of Xanthomonas perforans transplant (top panel) and field (side panel) grower A strains. GEV2047 GEV2052 GEV2063 GEV2049 GEV2055 GEV2059 GEV2060 GEV2048 GEV2058 GEV2050 GEV2011 99.99 99.99 99.96 99.99 99.99 99.98 99.98 99.99 99.98 99.98 GEV1989 99.99 99.99 99.96 99.99 100 99.99 99.99 99.99 100 99.99 GEV2009 99.97 99.99 99.96 99.97 99.98 99.98 99.98 99.99 99.99 99.98 GEV1993 99.99 99.98 99.96 99.99 100 99.99 100 100 99.99 99.99 GEV2010 99.64 99.64 99.61 99.65 99.65 99.65 99.66 99.64 99.65 99.64 GEV2015 99.98 99.99 99.94 99.99 100 99.99 99.99 99.99 99.99 99.99 GEV2004 99.95 99.97 99.95 99.96 99.97 99.97 99.97 99.97 99.97 99.97 GEV2013 99.98 99.98 99.95 99.98 99.98 99.98 99.98 99.98 99.98 99.98 GEV1992 99.98 99.99 99.95 99.99 100 100 99.99 100 99.99 99.98 GEV1991 99.99 100 99.96 99.99 100 99.99 99.99 100 99.99 99.98

132

Table B-5. Average nucleotide identity showing a one-way comparison of Xanthomonas perforans transplant (top panel) and field (side panel) grower B strains. GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV 2099 2072 1914 1918 2089 2098 1920 1913 1917 2088 2067 2097 1921 2087 1916 1912 1915 2065 1919 1911 GEV 99.6 99.96 99.97 99.96 99.91 99.98 99.95 99.96 99.96 99.97 99.98 99.53 99.95 99.96 99.96 99.96 99.97 99.52 99.96 99.96 2110 GEV 99.98 99.58 99.64 99.65 99.5 99.58 99.58 99.66 99.66 99.57 99.59 99.98 99.59 99.6 99.67 99.66 99.65 99.91 99.65 99.65 2108 GEV 99.56 99.95 99.99 99.96 99.92 99.98 99.95 99.96 99.97 99.98 99.98 99.5 99.96 99.96 99.96 99.96 99.98 99.48 99.96 99.95 2133 GEV 99.98 99.58 99.65 99.66 99.52 99.59 99.6 99.68 99.68 99.57 99.61 99.97 99.61 99.58 99.68 99.67 99.66 99.92 99.67 99.67 2127 GEV 99.57 99.98 99.95 99.98 99.94 99.97 99.93 99.98 99.97 99.96 99.94 99.52 99.96 99.93 99.98 99.98 99.96 99.46 99.97 99.97 2123 GEV 99.5 99.95 99.98 99.95 99.9 99.98 99.96 99.96 99.95 99.96 99.96 99.45 99.93 99.95 99.95 99.95 99.99 99.47 99.95 99.95 2114 GEV 99.57 99.98 99.96 99.98 99.95 99.97 99.94 99.99 99.98 99.96 99.96 99.51 99.95 99.93 99.97 99.98 99.97 99.48 99.98 99.98 2118 GEV 99.94 99.55 99.6 99.61 99.47 99.54 99.55 99.63 99.64 99.52 99.55 99.95 99.54 99.52 99.61 99.63 99.6 99.93 99.61 99.6 2134 GEV 99.54 99.95 99.97 99.95 99.87 99.96 99.95 99.95 99.95 99.95 99.99 99.49 99.94 99.94 99.95 99.95 99.98 99.45 99.95 99.95 2113

1 GEV

33 99.98 99.57 99.66 99.65 99.45 99.58 99.59 99.66 99.66 99.53 99.58 99.97 99.59 99.58 99.67 99.66 99.64 99.98 99.66 99.65 2128

GEV 99.98 99.61 99.66 99.68 99.54 99.6 99.63 99.69 99.67 99.61 99.62 99.96 99.61 99.6 99.69 99.69 99.67 99.92 99.68 99.67 2124 GEV 99.52 99.95 99.98 99.97 99.92 99.98 99.95 99.97 99.97 99.97 99.98 99.46 99.93 99.93 99.97 99.97 99.99 99.44 99.97 99.96 2120 GEV 99.58 99.97 99.97 99.96 99.91 99.98 99.95 99.96 99.96 99.98 99.97 99.54 99.94 99.96 99.95 99.97 99.98 99.52 99.96 99.95 2117 GEV 99.98 99.57 99.64 99.66 99.49 99.58 99.59 99.67 99.67 99.57 99.6 99.95 99.6 99.56 99.65 99.66 99.66 99.92 99.66 99.66 2130 GEV 99.6 99.96 99.97 99.96 99.94 99.98 99.95 99.96 99.96 99.98 99.98 99.55 99.96 99.96 99.96 99.96 99.98 99.52 99.96 99.95 2135 GEV 99.94 99.55 99.61 99.64 99.44 99.54 99.56 99.64 99.63 99.51 99.55 99.96 99.55 99.51 99.64 99.64 99.63 99.98 99.64 99.63 2112 GEV 99.94 99.56 99.62 99.63 99.47 99.55 99.56 99.63 99.63 99.54 99.57 99.92 99.57 99.58 99.63 99.63 99.62 99.88 99.62 99.62 2125 GEV 99.59 99.98 99.97 99.98 99.96 99.97 99.94 99.99 99.99 99.97 99.97 99.56 99.98 99.95 99.99 99.98 99.97 99.49 99.99 99.98 2129 GEV 99.94 99.52 99.58 99.6 99.45 99.52 99.52 99.59 99.59 99.5 99.5 99.96 99.52 99.52 99.6 99.61 99.57 99.92 99.59 99.58 2121 GEV 99.54 99.96 99.96 99.96 99.94 99.96 99.93 99.97 99.97 99.95 99.97 99.5 99.96 99.94 99.97 99.97 99.96 99.47 99.96 99.96 2116 GEV 99.59 99.98 99.96 99.98 99.92 99.96 99.94 99.98 99.98 99.95 99.97 99.54 99.97 99.92 99.98 99.98 99.97 99.49 99.98 99.97 2109 GEV 99.59 99.97 99.98 99.96 99.9 99.98 99.96 99.97 99.95 99.98 99.98 99.59 99.96 99.96 99.95 99.97 99.98 99.55 99.97 99.96 2111

Table B-5. Continued. GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV GEV

2099 2072 1914 1918 2089 2098 1920 1913 1917 2088 2067 2097 1921 2087 1916 1912 1915 2065 1919 1911 GEV 99.6 99.97 99.98 99.96 99.92 99.99 99.96 99.97 99.97 99.98 99.98 99.58 99.96 99.95 99.96 99.97 99.98 99.55 99.97 99.96 2132 GEV 99.61 99.96 99.99 99.95 99.89 99.98 99.95 99.95 99.95 99.97 99.98 99.59 99.93 99.95 99.95 99.96 99.97 99.55 99.95 99.95 2126 GEV 99.93 99.54 99.59 99.61 99.41 99.54 99.54 99.62 99.62 99.49 99.54 99.96 99.55 99.53 99.61 99.62 99.6 99.97 99.61 99.61 2122 GEV 99.53 99.97 99.96 99.97 99.94 99.97 99.94 99.97 99.97 99.97 99.98 99.49 99.97 99.95 99.98 99.97 99.97 99.43 99.97 99.96 2115 GEV 99.56 99.95 99.97 99.95 99.9 99.98 99.96 99.95 99.96 99.96 99.97 99.5 99.92 99.95 99.95 99.95 99.98 99.48 99.95 99.94 2119

134

Table B-6. Singe nucleotide polymorphism (SNP) details for individually sequenced strains in this study. non- non- Strain #SNPa Coverage CDSb Synonymous Truncation Extension CDS synonymousc GEV839 5643 46.3 4381 1262 1044 3321 8 7 GEV1001 5725 48.2 4389 1336 1050 3324 8 7 GEV872 1575 49.8 1261 314 339 919 2 1 GEV993 1560 47.3 1248 312 330 915 2 1 Xp17-12 5847 506.2 4602 1245 1105 3489 6 2 GEV1911 5557 22.5 4342 1215 1054 3271 9 8 GEV1912 5657 31.3 4385 1272 1063 3307 8 7 GEV1913 5608 26.9 4359 1249 1051 3291 9 8 GEV1914 5524 25.2 4286 1238 1034 3236 8 8 GEV1915 5574 25.4 4340 1234 1045 3281 7 7 GEV1916 5680 31.5 4404 1276 1072 3316 9 7 GEV1917 5666 33 4387 1279 1058 3313 8 8

1 GEV1918 5715 29.9 4420 1295 1087 3319 8 6 35 GEV1919 5752 48.4 4423 1329 1077 3330 9 7 GEV1920 5561 25.7 4308 1253 1028 3266 7 7 GEV1921 5643 30.8 4364 1279 1039 3310 7 8 GEV1989 5675 38.6 4361 1314 1030 3316 8 7 GEV1991 5661 33.6 4341 1320 1020 3306 8 7 GEV1992 5673 37.9 4352 1321 1026 3311 8 7 GEV1993 5632 33.3 4323 1309 1008 3300 8 7 GEV2004 5251 17.7 4125 1126 957 3151 10 7 GEV2009 5708 45.6 4425 1283 1077 3332 9 7 GEV2010 5976 26.6 4680 1296 1126 3546 6 2 GEV2011 5632 35 4383 1249 1052 3316 8 7 GEV2013 5611 30.2 4386 1225 1057 3313 8 8 GEV2015 5638 34.1 4345 1293 1024 3304 10 7 GEV2047 5570 26.4 4302 1268 1012 3275 8 7 GEV2048 5698 46.6 4385 1313 1038 3331 8 8 GEV2049 5654 39.8 4368 1286 1044 3309 8 7

Table B-6. Continued. non- non- Strain #SNPa Coverage CDSb Synonymous Truncation Extension CDS synonymousc GEV2050 5664 40.2 4361 1303 1031 3315 8 7 GEV2052 5640 34.9 4336 1304 1024 3297 8 7 GEV2055 5685 39.1 4362 1323 1024 3323 8 7 GEV2056 5713 56.3 4371 1342 1041 3315 8 7 GEV2059 5683 45 4363 1320 1026 3322 8 7 GEV2060 5728 46.2 4367 1361 1026 3325 8 8 GEV2063 5679 30.2 4354 1325 1024 3314 8 8 GEV2065 6043 28.3 4683 1360 1133 3540 8 2 GEV2067 5634 34.3 4372 1262 1055 3304 7 6 GEV2072 5724 36.5 4406 1318 1073 3318 8 7 GEV2087 5664 35.2 4385 1279 1061 3309 8 7 GEV2088 5612 28.5 4389 1223 1074 3298 10 7

1 GEV2089 5677 28.7 4402 1275 1074 3312 9 7 36 GEV2097 6022 29.3 4681 1341 1126 3544 9 2 GEV2098 5643 29.9 4371 1272 1049 3304 10 8 GEV2099 6002 28.8 4659 1343 1121 3530 6 2 GEV2108 6019 35.1 4661 1357 1112 3541 6 2 GEV2109 5712 40.8 4397 1315 1079 3301 9 8 GEV2110 5804 43.6 4429 1375 1083 3331 8 7 GEV2111 5804 53.8 4441 1363 1082 3340 12 7 GEV2112 5974 25.1 4641 1333 1118 3513 8 2 GEV2113 5586 25.5 4353 1233 1046 3287 12 8 GEV2114 5584 28.6 4345 1239 1044 3288 7 6 GEV2115 5614 29.9 4392 1222 1069 3307 9 7 GEV2116 5608 27.3 4360 1247 1052 3293 8 7 GEV2117 5657 27 4342 1315 1027 3300 8 7 GEV2118 5660 26.8 4375 1285 1059 3301 8 7 GEV2119 5562 24.1 4354 1208 1057 3280 11 6 GEV2120 5549 26.6 4311 1238 1022 3275 8 6 GEV2121 5837 19.9 4591 1246 1090 3494 5 2

Table B-6. Continued. non- non- Strain #SNPa Coverage CDSb Synonymous Truncation Extension CDS synonymousc GEV2122 5877 20.4 4618 1259 1112 3497 7 2 GEV2123 5308 20.6 4117 1191 1054 3271 7 8 GEV2124 5943 24.2 4638 1305 1102 3528 6 2 GEV2125 4998 14.1 3946 1052 916 3022 6 2 GEV2126 5697 30.3 4388 1309 1058 3313 10 7 GEV2127 5852 20.8 4551 1301 1072 3469 8 2 GEV2128 6049 30.9 4699 1350 1147 3542 8 2 GEV2129 5640 26.2 4378 1262 1060 3301 8 9 GEV2130 6009 38.8 4681 1328 1130 3542 7 2 GEV2132 5711 33.1 4403 1308 1065 3320 8 8 GEV2133 5657 32.7 4383 1274 1071 3296 9 7 137 GEV2134 5972 25.4 4679 1293 1128 3543 6 2

GEV2135 5692 27.9 4395 1297 1045 3334 8 8 a Compared to reference genome Xanthomonas perforans 91-118. b CDS: coding DNA sequence. Average number of SNPs in CDS and non-CDS. c Effect of SNP on coding DNA sequence (CDS) protein product. Non-synonymous: amino acid change; synonymous: no amino acid change; truncation: early stop codon in CDS; extension: amino acid addition.

Table B-7. The range of average genetic distance of transplant strains compared to grower A field strains recovered from various cultivars based on single nucleotide polymorphisms. Cultivar Field strain Min Max H GEV1993 0.059 0.074 H GEV1989 0.064 0.089 H GEV1992 0.075 0.1 H GEV2015 0.063 0.088 H GEV1991 0.047 0.094 Transplant rangea 0.05 0.10 Cultivar Strain Min Max M GEV2004 0.117 0.128 M GEV2011 0.094 0.106 M GEV2013 0.096 0.107 M GEV2009 0.096 0.108 Transplant range 0.046 0.081 a Range of genetic distance for pairwise comparison between transplant strains recovered from each cultivar. Compare genetic distance range of field strains with transplant range. Field strains with distances greater than transplant range indicate distantly related strains from transplant strains.

138

Table B-8. The range of average genetic distance of transplant strains compared to grower B field strains recovered from various cultivars based on single nucleotide polymorphisms. a P8 B7 H P0 P9 Min Max Min Max Min Max Min Max Min Max GEV2109b 0.073 0.137 0.069 0.12 0.078 0.099 0.083 0.13 0.065 0.115 GEV2118 0.09 0.154 0.086 0.137 0.095 0.116 0.1 0.147 0.069 0.132 GEV2129 0.079 0.123 0.079 0.106 0.077 0.089 0.078 0.116 0.07 0.101 GEV2135 0.135 0.157 0.135 0.139 0.133 0.145 0.12 0.133 0.134 0.147 GEV2117 0.132 0.154 0.132 0.137 0.131 0.143 0.118 0.131 0.132 0.145 GEV2132 0.133 0.155 0.133 0.137 0.131 0.143 0.119 0.132 0.133 0.145 GEV2126 0.13 0.152 0.131 0.135 0.129 0.141 0.116 0.129 0.13 0.143 GEV2111 0.134 0.156 0.134 0.139 0.133 0.145 0.12 0.133 0.134 0.147 GEV2110 0.139 0.161 0.139 0.144 0.138 0.15 0.125 0.138 0.139 0.151 GEV2119 0.117 0.139 0.117 0.121 0.115 0.127 0.092 0.116 0.117 0.129 GEV2133 0.124 0.146 0.124 0.129 0.123 0.135 0.105 0.123 0.124 0.137 GEV2113 0.099 0.117 0.099 0.099 0.097 0.109 0.098 0.113 0.095 0.111 GEV2114 0.08 0.108 0.072 0.106 0.104 0.116 0.105 0.121 0.068 0.118 GEV2120 0.082 0.114 0.085 0.112 0.11 0.122 0.11 0.126 0.081 0.124 GEV2116 0.102 0.119 0.101 0.105 0.103 0.115 0.104 0.119 0.096 0.117 GEV2115 0.091 0.108 0.091 0.095 0.093 0.105 0.093 0.109 0.086 0.107 GEV2123 0.136 0.188 0.139 0.17 0.137 0.149 0.138 0.18 0.143 0.166 Transplant 0.077 0.131 0.111d 0.078 0.1 0.088 0.12 0.097 0.119 rangec a Strains were recovered from tomato cultivars P8, B7 and H in December 2015 and from cultivars P0 and P9 in March 2016 in transplant houses from grower B. b Field strains were recovered from cultivar P8 in March 2016. c Range of genetic distance for pairwise comparison between transplant strains recovered from each cultivar. Compare genetic distance range of field strains with transplant range. Field strains with distances greater than transplant range indicate distantly related strains from transplant strains. d Genetic distance between only two transplant strains recovered from B7.

139

Table B-9. Race profile of characterized strains collected from grower A. Tomato Tomato Tomato Pepper Strain Year Location Sourcea Bonnie Bestb Hawaii7998 FL216 ECW GEV1989 2015 Manatee Co. F S S S HR GEV1991 2015 Manatee Co. F S S S HR GEV1992 2015 Manatee Co. F S S S HR GEV1993 2015 Manatee Co. F S S S HR GEV2004 2015 Manatee Co. F S S S HR GEV2009 2015 Manatee Co. F S S S S GEV2010 2015 Manatee Co. F S S S HR GEV2011 2015 Manatee Co. F S S S S GEV2013 2015 Manatee Co. F S S S S GEV2015 2015 Manatee Co. F S S S HR GEV2047 2015 Polk Co. T S S S HR GEV2048 2015 Polk Co. T S S S HR GEV2049 2015 Polk Co. T S S S HR GEV2050 2015 Polk Co. T S S S HR GEV2052 2015 Polk Co. T S S S HR GEV2055 2015 Polk Co. T S S S HR GEV2058 2015 Polk Co. T S S S HR GEV2059 2015 Polk Co. T S S S HR GEV2060 2015 Polk Co. T S S S HR GEV2063 2015 Polk Co. T S S S HR a F: field; T; transplant. b S: susceptible; HR: hypersensitive reaction.

140

Table B-10. Race profile of characterized strains collected from grower B. Tomato Tomato Tomato Pepper Strain Year Location Sourcea Bonnie Bestb Hawaii7998 FL216 ECW GEV1911 2015 Collier Co. T S S S HR GEV1912 2015 Collier Co. T S S S HR GEV1913 2015 Collier Co. T S S S HR GEV1914 2015 Collier Co. T S S S HR GEV1915 2015 Collier Co. T S S S HR GEV1916 2015 Collier Co. T S S S HR GEV1917 2015 Collier Co. T S S S HR GEV1918 2015 Collier Co. T S S S HR GEV1919 2015 Collier Co. T S S S HR GEV1920 2015 Collier Co. T S S S HR GEV1921 2015 Collier Co. T S S S HR GEV2065 2016 Collier Co. T S S S HR GEV2067 2016 Collier Co. T S S S HR GEV2072 2016 Collier Co. T -c - - - GEV2087 2016 Collier Co. T S S S HR GEV2088 2016 Collier Co. T S S S HR GEV2089 2016 Collier Co. T S S S HR GEV2097 2016 Collier Co. T S S S^ HR GEV2098 2016 Collier Co. T S S S HR GEV2099 2016 Collier Co. T S S S S GEV2108 2016 Collier Co. F S S S S GEV2109 2016 Collier Co. F S S S HR GEV2110 2016 Collier Co. F S S S HR GEV2111 2016 Collier Co. F S S S S GEV2112 2016 Collier Co. F S S^ S^ HR GEV2113 2016 Collier Co. F S S S HR GEV2114 2016 Collier Co. F S S S HR GEV2115 2016 Collier Co. F S S S HR GEV2116 2016 Collier Co. F S S S S* GEV2117 2016 Collier Co. F S S S S* GEV2118 2016 Collier Co. F S S S HR GEV2119 2016 Collier Co. F S S S HR GEV2120 2016 Collier Co. F S S S HR GEV2121 2016 Collier Co. F S S S HR GEV2122 2016 Collier Co. F S S S^ HR GEV2123 2016 Collier Co. F S S S HR GEV2124 2016 Collier Co. F S S S^ HR* GEV2125 2016 Collier Co. F S S S^ HR* GEV2126 2016 Collier Co. F S S S S* GEV2127 2016 Collier Co. F S S S^ HR* GEV2128 2016 Collier Co. F S S S HR* GEV2129 2016 Collier Co. F S S S HR GEV2130 2016 Collier Co. F S S S HR*

141

Table B-10. Continued. Tomato Tomato Tomato Pepper Strain Year Location Source Bonnie Best Hawaii7998 FL216 ECW GEV2132 2016 Collier Co. F S S S S GEV2133 2016 Collier Co. F S S S HR GEV2134 2016 Collier Co. F S S S HR GEV2135 2016 Collier Co. F S S S HR a F: field; T; transplant. b S: susceptible; HR: hypersenstivie reaction. c Not tested. ^ Weak HR-like symptoms after 36 hours. *'HR’ or ‘S’ phenotype does not match genotype in these strains.

142

Figure B-1. Neighbor-joining phylogenetic tree of the YopJ-like homologs of plant pathogenic bacteria. The avrBsT-homolog refers to the xopJ6 gene. Bootstrap values are indicated on branches.

143

LIST OF REFERENCES

Abbasi, P. A., Khabbaz, S. E., Weselowski, B., and Zhang, L. (2015). Occurrence of copper- resistant strains and a shift in Xanthomonas spp. causing tomato bacterial spot in Ontario. Can. J. Microbiol., 61:753-761. doi: http://dx.doi.org/10.1139/cjm-2015-0228

Abbasi, P. A., and Weselowski, B. (2015). Efficacy of Bacillus subtilis QST 713 formulations, copper hydroxide, and their tank mixes on bacterial spot of tomato. Crop Prot., 74:70-76. doi: https://doi.org/10.1016/j.cropro.2015.04.009

Alfano, J. R., & Collmer, A. (2004). TYPE III SECRETION SYSTEM EFFECTOR PROTEINS: Double Agents in Bacterial Disease and Plant Defense. Annual Review of Phytopathology, 42:385-414. doi: http://dx.doi.org/10.1146/annurev.phyto.42.040103.110731

Aiello, D., Scuderi, G., Vitale, A., Firrao, G., Polizzi, G., and Cirvilleri, G. (2013). A pith necrosis caused by Xanthomonas perforans on tomato plants. Eur. J. Plant Path., 137:29-41. doi: http://dx.doi.org/10.1007/s10658-013-0214-7

Albuquerque, P., Caridade, C. M. R., Rodrigues, A. S., Marcal, A. R. S., Cruz, J., Cruz, L., Santos, C. L., Mendes, M. V., and Tavares, F. (2012). Evolutionary and experimental assessment of novel markers for detection of Xanthomonas euvesicatoria in plant samples. PLoS One, 7:e37836. doi: http://dx.doi.org/10.1371/journal.pone.0037836

Almeida, N. F., Yan, S., Cai, R., Clarke, C. R., Morris, C. E., Schaad, N. W., Schuenzel, E. L., Lacy, G. H., Sun, X., Jones, J. B., Castillo, J. A., Bull, C. T., Leman, S., Guttman, D. S., Setubal, J. C., and Vinatzer, B. A. (2010). PAMDB, A multilocus sequence typing and analysis database and website for plant-associated microbes. Phytopathology, 100:208-215. doi: http://dx.doi.org/10.1094/phyto-100-3-0208

Anonymous. (1996). Data Sheets on Quarantine Pests - Xanthomonas vesicatoria. EPPO Bulletin.

Araújo, E. R., Costa, J. R., Ferreira, M. A. S. V., and Quezado-Duval, A. M. (2017). Widespread distribution of Xanthomonas perforans and limited presence of X. gardneri in Brazil. Plant Pathology, 66:159-168. doi: http://dx.doi.org/10.1111/ppa.12543

Araújo, E. R., Costa, J. R., Pontes, N. C., and Quezado-Duval, A. M. (2015). Xanthomonas perforans and X. gardneri associated with bacterial leaf spot on weeds in Brazilian tomato fields. Eur. J. Plant Path., 143:543-548. doi: http://dx.doi.org/10.1007/s10658-015-0705-9

Araújo, E., Pereira, R., Ferreira, M., Quezado-Duval, A., and Café-Filho, A. (2012). Sensitivity of xanthomonads causing tomato bacterial spot to copper and streptomycin and in vivo infra- specific competitive ability in Xanthomonas perforans resistant and sensitive to copper. J. Plant Path., 94:79-87.

144

Aritua, V., Harrison, J., Sapp, M., Buruchara, R., Smith, J., and Studholme, D. J. (2015). Genome sequencing reveals a new lineage associated with lablab bean and genetic exchange between Xanthomonas axonopodis pv. phaseoli and Xanthomonas fuscans subsp. fuscans. Front. Microbiol., 6:1080. doi: http://dx.doi.org/10.3389/fmicb.2015.01080

Astua-Monge, G., Minsavage, G. V., Stall, R. E., Vallejos, C. E., Davis, M. J., and Jones, J. B. (2000). Xv4-vrxv4: a new gene-for-gene interaction identified between Xanthomonas campestris pv. vesicatoria race T3 and the wild tomato relative Lycopersicon pennellii. Mol. Plant-Microbe In., 13:1346-1355. doi: http://dx.doi.org/10.1094/mpmi.2000.13.12.1346

Barak, J. D., Vancheva, T., Lefeuvre, P., Jones, J. B., Timilsina, S., Minsavage, G. V., Vallad, G. E., and Koebnik, R. (2016). Whole-genome sequences of Xanthomonas euvesicatoria strains clarify taxonomy and reveal a stepwise erosion of type 3 effectors. Front. Plant Sci., 7:1805. doi: http://dx.doi.org/10.3389/fpls.2016.01805

Bashan, Y., Okon, Y., and Henis, Y. (1982). Long-term survival of Pseudomonas syringae pv. tomato and Xanthomonas campestris pv. vesicatoria in tomato and pepper seeds. Phytopathology, 72:143-1144.

Basim , H., Minsavage, G. V., Stall, R. E., Wang, J.-F., Shanker, S., and Jones, J. B. (2005). Characterization of a unique chromosomal copper resistance gene cluster from Xanthomonas campestris pv. vesicatoria. Appl. Environ. Microb., 71:8284-8291. doi: http://dx.doi.org/10.1128/aem.71.12.8284-8291.2005

Behlau, F., Canteros, B. I., Minsavage, G. V., Jones, J. B., and Graham, J. H. (2011). Molecular characterization of copper resistance genes from Xanthomonas citri subsp. citri and Xanthomonas alfalfae subsp. citrumelonis. Appl. Environ. Microb., 77:4089-4096. doi: http://dx.doi.org/10.1128/aem.03043-10

Bent, A. F., and Mackey, D. (2007). Elicitors, effectors, and R genes: the new paradigm and a lifetime supply of questions. Annu. Rev. Phytopath., 45:399-436. doi: doi: http://dx.doi.org/10.1146/annurev.phyto.45.062806.094427

Bernal, R., and Berger, R. (1996). The spread of epiphytic populations of Xanthomonas campestris pv. vesicatoria on pepper in the field. J. Phytopath., 144:479-484.

Boch, J., and Bonas, U. (2010). Xanthomonas AvrBs3 family-type III effectors: discovery and function. Annu. Rev. Phytopath., 48:419-436. doi: http://dx.doi.org/10.1146/annurev-phyto- 080508-081936

Bock, C. H., Gottwald, T. R., Parker, P. E., Cook, A. Z., Ferrandino, F., Parnell, S., and van den Bosch, F. (2009). The Horsfall-Barratt scale and severity estimates of citrus canker. Eur. J. Plant Path., 125:23-38. doi: http://dx.doi.org/10.1007/s10658-009-9455-x

145

Bouzar, H., Jones, J. B., Stall, R. E., Louws, F. J., Schneider, M., Rademaker, J. L. W., de Bruijn, F. J., and Jackson, L. E. (1999). Multiphasic analysis of xanthomonads causing bacterial spot disease on tomato and pepper in the Caribbean and Central America: evidence for common lineages within and between countries. Phytopathology, 89:328-335. doi: http://dx.doi.org/10.1094/phyto.1999.89.4.328 Bouzar, H., Jones, J., Somodi, G., Stall, R., Daouzli, N., Lambe, R., Gastelum, R. F., and Correa, R. T. (1996). Diversity of Xanthomonas campestris pv. vesicatoria in tomato and pepper fields of Mexico. Can. J. Plant Path., 18:75-77.

Bouzar, H., Jones, J., Stall, R., Hodge, N., Minsavage, G., Benedict, A., and Alvarez, A. (1994). Physiological, chemical, serological, and pathogenic analyses of a worldwide collection of Xanthomonas campestris pv. vesicatoria strains. Phytopathology, 84:663-671.

Briceno-Montero, G., and Miller, S. A. (2004). Evaluation of biological control options for bacterial spot management during tomato transplant production. Acta Hortic., 695:357-365.

Brumfield, R. T., Beerli, P., Nickerson, D. A., and Edwards, S. V. (2003). The utility of single nucleotide polymorphisms in inferences of population history. Trends Ecol. Evol., 18:249- 256. doi: http://dx.doi.org/10.1016/S0169-5347(03)00018-1

Büttner, D., and Bonas, U. (2010). Regulation and secretion of Xanthomonas virulence factors. FEMS Microbiol. Rev., 34:107-133. doi: http://dx.doi.org/10.1111/j.1574- 6976.2009.00192.x

Casper-Lindley, C., Dahlbeck, D., Clark, E. T., and Staskawicz, B. J. (2002). Direct biochemical evidence for type III secretion-dependent translocation of the AvrBs2 effector protein into plant cells. P. Natl. Acad. Sci. USA, 99:8336-8341. doi: http://dx.doi.org/10.1073/pnas.122220299

Cha, J. S., and Cooksey, D. A. (1991). Copper resistance in Pseudomonas syringae mediated by periplasmic and outer membrane proteins. P. Natl. Acad. Sci. USA, 88:8915-8919. doi: http://dx.doi.org/10.1073/pnas.88.20.8915

Chang, R., Ries, S., and Pataky, J. (1992). Local sources of Clavibacter michiganensis ssp. michiganensis in the development of bacterial canker on tomatoes. Phytopathology, 82:553- 560.

Cheong, M. S., Kirik, A., Kim, J.-G., Frame, K., Kirik, V., and Mudgett, M. B. (2014). AvrBsT acetylates Arabidopsis ACIP1, a protein that associates with microtubules and is required for immunity. PLoS Path., 10:e1003952. doi: http://dx.doi.org/10.1371/journal.ppat.1003952

Cock, P. J. A., Antao, T., Chang, J. T., Chapman, B. A., Cox, C. J., Dalke, A., Friedberg, I., Hamelryck, T., Kauff, F., Wilczynski, B., and de Hoon, M. J. L. (2009). Biopython: freely

146

available Python tools for computational molecular biology and bioinformatics. Bioinformatics, 25:1422-1423. doi: http://dx.doi.org/10.1093/bioinformatics/btp163

Conover, R. A., and Gerhold, N. R. (1981). Mixtures of copper and maneb or mancozeb for control of bacterial spot of tomato and their compatibility for control of fungus diseases [Phytophthora infestans, Stemphylium solani, Xanthomonas campestris pv. vesicatoria, Florida]. P. Florida State Hortic. Soc., 94:154–156.

Constantin, E. C., Cleenwerck, I., Maes, M., Baeyen, S., Van Malderghem, C., De Vos, P., and Cottyn, B. (2016). Genetic characterization of strains named as Xanthomonas axonopodis pv. dieffenbachiae leads to a taxonomic revision of the X. axonopodis species complex. Plant Path., 65:792-806. doi: http://dx.doi.org/10.1111/ppa.1246

Contreras-Moreira, B., and Vinuesa, P. (2013). GET_HOMOLOGUES, a versatile software package for scalable and robust microbial pangenome analysis. Appl. Environ. Microb., 79:7696-7701. doi: http://dx.doi.org/10.1128/aem.02411-13

Croucher, N. J., Finkelstein, J. A., Pelton, S. I., Mitchell, P. K., Lee, G. M., Parkhill, J., Bentley, S. D., Hanage, W. P., and Lipsitch, M. (2013). Population genomics of post-vaccine changes in pneumococcal epidemiology. Nature Genet., 45:656-663. doi: http://dx.doi.org/10.1038/ng.2625

Cunnac, S., Wilson, A., Nuwer, J., Kirik, A., Baranage, G., and Mudgett, M. B. (2007). A conserved carboxylesterase is a SUPPRESSOR OF AVRBST-ELICITED RESISTANCE in Arabidopsis. Plant Cell, 19:688-705. doi: http://dx.doi.org/10.1105/tpc.106.048710

Cuppels, D. A., and Elmhirst, J. (1999). Disease Development and Changes in the Natural Pseudomonas syringae pv. tomato populations on field tomato plants. Plant Dis., 83:759-764. doi: http://dx.doi.org/10.1094/pdis.1999.83.8.759

Durrant, W. E., and Dong, X. (2004). Systemic acquired resistance. Annu. Rev. Phytopath., 42:185-209. doi: doi: http://dx.doi.org/10.1146/annurev.phyto.42.040803.140421

Dutta, B., Gitaitis, R., Smith, S., and Langston, D., Jr. (2014). Interactions of seedborne bacterial pathogens with host and non-host plants in relation to seed infestation and seedling transmission. PLoS One, 9:e99215. doi: http://dx.doi.org/10.1371/journal.pone.0099215

Dye, D., Bradbury, J., Goto, M., Hayward, A., Lelliott, R., and Schroth, M. N. (1980). International standards for naming pathovars of phytopathogenic bacteria and a list of pathovar names and pathotype strains. Rev. Plant Pathol., 59:153-168.

Escolar, L., Van Den Ackerveken, G., Pieplow, S., Rossier, O., and Bonas, U. (2001). Type III secretion and in planta recognition of the Xanthomonas avirulence proteins AvrBs1 and

147

AvrBsT. Mol. Plant Pathol., 2:287-296. doi: http://dx.doi.org/10.1046/j.1464- 6722.2001.00077.x

Fayette, J., Roberts, P. D., Pernezny, K. L., and Jones, J. B. (2012). The role of cymoxanil and famoxadone in the management of bacterial spot on tomato and pepper and bacterial leaf spot on lettuce. Crop Prot., 31:107-112. doi: https://doi.org/10.1016/j.cropro.2011.09.006 Figurski, D. H., and Helinski, D. R. (1979). Replication of an origin-containing derivative of plasmid RK2 dependent on a plasmid function provided in trans. P. Natl. Acad. Sci. USA, 76:1648-1652.

Freeman, J. H., McAvoy, E. J., Boyd, N. S., Kinessary, R., Ozores-Hampton, M., Smith, H. A., Noling, J. W., and Vallad, G. E. (2017). Tomato production in Florida. In G. E. Vallad, H. A. Smith, P. J. Dittmar and J. H. Freeman (Eds.), Vegetable Production Handbook (Vol. HS739). University of Florida/IFAS: EDIS.

Gabriel, D. W. (1999). Why do pathogens carry avirulence genes? Physiol. Mol. Plant Pathol., 55:205-214. doi: http://dx.doi.org/10.1006/pmpp.1999.0230

Galán, J. E., and Collmer, A. (1999). Type III secretion machines: bacterial devices for protein delivery into host cells. Science, 284:1322-1328. doi: http://dx.doi.org/10.1126/science.284.5418.1322

Gitaitis, R., McCarter, S., and Jones, J. (1992). Disease control in tomato transplants produced in Georgia and Florida. Plant Dis., 76:651-656.

Goris, J., Konstantinidis, K. T., Klappenbach, J. A., Coenye, T., Vandamme, P., and Tiedje, J. M. (2007). DNA–DNA hybridization values and their relationship to whole-genome sequence similarities. Int. J. Syst. Evol. Micr., 57:81-91. doi: http://dx.doi.org/10.1099/ijs.0.64483-0

Griffin, K., Gambley, C., Brown, P., and Li, Y. (2017). Copper-tolerance in Pseudomonas syringae pv. tomato and Xanthomonas spp. and the control of diseases associated with these pathogens in tomato and pepper. A systematic literature review. Crop Prot., 96:144-150. doi: http://dx.doi.org/10.1016/j.cropro.2017.02.008

Gürlebeck, D., Thieme, F., and Bonas, U. (2006). Type III effector proteins from the plant pathogen Xanthomonas and their role in the interaction with the host plant. J. Plant Physiol., 163(3), 233-255. doi: https://doi.org/10.1016/j.jplph.2005.11.011

Hamza, A. A., Robène-Soustrade, I., Jouen, E., Gagnevin, L., Lefeuvre, P., Chiroleu, F., and Pruvost, O. (2010). Genetic and pathological diversity among Xanthomonas strains responsible for bacterial spot on tomato and pepper in the Southwest Indian Ocean Region. Plant Dis., 94:993-999. doi: http://dx.doi.org/10.1094/pdis-94-8-0993

148

Hirano, S. S., and Upper, C. D. (1983). Ecology and epidemiology of foliar bacterial plant pathogens. Annu. Rev. Phytopath., 21:243-270. doi: http://dx.doi.org/10.1146/annurev.py.21.090183.001331

Horvath, D. M., Stall, R. E., Jones, J. B., Pauly, M. H., Vallad, G. E., Dahlbeck, D., Staskawicz, B. J., and Scott, J. W. (2012). Transgenic resistance confers effective field level control of bacterial spot disease in tomato. PLoS One, 7:e42036. doi: https://doi.org/10.1371/journal.pone.0042036

Huang, C. -H. and Vallad, G. E. (2012). Evaluation of bactericides and Actigard for management of bacterial spot of tomato, fall 2010. Plant Dis. Manag. Rep. 6:V039

Huang, C.-H., Vallad, G. E., Zhang, S., Wen, A., Balogh, B., Figueiredo, J. F. L., Behlau, F., Jones, J. B., Momol, M. T., and Olson, S. M. (2012). Effect of application frequency and reduced rates of acibenzolar-S-methyl on the field efficacy of induced resistance against bacterial spot on tomato. Plant Dis., 96:221-227. doi: http://dx.doi.org/10.1094/pdis-03-11- 0183

Huang, C.-L., Pu, P.-H., Huang, H.-J., Sung, H.-M., Liaw, H.-J., Chen, Y.-M., Chen, C.-M., Huang, M.-B., Osada, N., Gojobori, T., Pai, T.-W., Chen, Y.-T., Hwang, C.-C., and Chiang, T.-Y. (2015). Ecological genomics in Xanthomonas: the nature of genetic adaptation with homologous recombination and host shifts. BMC Genomics, 16:188. doi: http://dx.doi.org/10.1186/s12864-015-1369-8

Incani, V., Omar, A., Prosperi-Porta, G., and Nadworny, P. (2015). Ag5IO6: novel antibiofilm activity of a silver compound with application to medical devices. Int. J. Antimicrob. Ag., 45:586-593. doi: https://doi.org/10.1016/j.ijantimicag.2014.09.008

Itako, A. T., Tolentino-Júnior, J. B., Rauer Demant, L. A., and Maringoni, A. C. (2014). Control of bacterial spot of tomato and activation of enzymes related to resistance by chemicals under field conditions. J. Agric. Sci., 6:100-109.

Jeger, M. J. (2004). Analysis of disease progress as a basis for evaluating disease management practices. Annu. Rev. Phytopath., 42:61-82. doi: http://dx.doi.org/10.1146/annurev.phyto.42.040803.140427

Jones, J. B., and Miller, S. A. (2014). Bacterial leaf spot. In J. B. Jones, T. A. Zitter, M. T. Momol and S. A. Miller (Eds.), Compendium of Tomato Diseases and Pests (2 ed.). St. Paul, Minnesota: APS Press.

Jones, J. B., Bouzar, H., Somodi, G. C., Stall, R. E., Pernezny, K., El-Morsy, G., and Scott, J. W. (1998a). Evidence for the preemptive nature of tomato race 3 of Xanthomonas campestris pv.

149

vesicatoria in Florida. Phytopathology, 88:33-38. doi: http://dx.doi.org/10.1094/phyto.1998.88.1.33

Jones, J. B., Bouzar, H., Stall, R. E., Almira, E. C., Roberts, P. D., Bowen, B. W., Sudberry, J., Strickler, P. M., and Chun, J. (2000). Systematic analysis of xanthomonads (Xanthomonas spp.) associated with pepper and tomato lesions. Int. J. Syst. Evol. Micr., 50:1211-1219. doi: http://dx.doi.org/10.1099/00207713-50-3-1211

Jones, J. B., Jackson, L. E., Balogh, B., Obradovic, A., Iriarte, F. B., and Momol, M. T. (2007). Bacteriophages for plant disease control. Annu. Rev. Phytopath., 45:245-262. doi: http://dx.doi.org/10.1146/annurev.phyto.45.062806.094411

Jones, J. B., Lacy, G. H., Bouzar, H., Minsavage, G. V., Stall, R. E., and Schaad, N. W. (2005). Bacterial spot-worldwide distribution, importance and review. Acta Hortic., 695:27-34.

Jones, J. B., Lacy, G. H., Bouzar, H., Stall, R. E., and Schaad, N. W. (2004). Reclassification of the xanthomonads associated with bacterial spot disease of tomato and pepper. Syst. Appl. Microbiol., 27:755-762. doi: http://dx.doi.org/10.1078/0723202042369884

Jones, J. B., Stall, R. E., and Bouzar, H. (1998b). Diversity among xanthomonads pathogenic on pepper and tomato. Annu. Rev. Phytopath., 36:41-58. doi: http://dx.doi.org/10.1146/annurev.phyto.36.1.41

Jones, J. B., Vallad, G. E., Iriarte, F. B., Obradović, A., Wernsing, M. H., Jackson, L. E., Balogh, B., Hong, J. C., and Momol, M. (2012). Considerations for using bacteriophages for plant disease control. Bacteriophage, 2:208-214. doi: http://dx.doi.org/10.4161/bact.23857

Jones, J., and Scott, J. (1986). Hypersensitive response in tomato to Xanthomonas campestris pv. vesicatoria. Plant Dis.e, 70:337-339.

Jones, J., Pohronezny, K., Stall, R., and Jones, J. (1986). Survival of Xanthomonas campestris pv. vesicatoria in Florida on tomato crop residue, weeds, seeds, and volunteer tomato plants. Phytopathology, 76:430-434.

Jones, J., Stall, R., Scott, J., Somodi, G., Bouzar, H., and Hodge, N. (1995). A third tomato race of Xanthomonas campestris pv. vesicatoria. Plant Dis., 79:395-398.

Jones, J., Woltz, S., Jones, J., and Portier, K. (1991). Population dynamics of Xanthomonas campestris pv. vesicatoria on tomato leaflets treated with copper bactericides. Phytopathology, 81:714-719.

Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M., Sturrock, S., Buxton, S., Cooper, A., Markowitz, S., Duran, C., Thierer, T., Ashton, B., Meintjes, P., and Drummond, A. (2012). Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics, 28:1647-1649. doi: http://dx.doi.org/10.1093/bioinformatics/bts199

150

Kebede, M., Timilsina, S., Ayalew, A., Admassu, B., Potnis, N., Minsavage, G. V., Goss, E. M., Hong, J. C., Strayer, A., Paret, M., Jones, J. B., and Vallad, G. E. (2014). Molecular characterization of Xanthomonas strains responsible for bacterial spot of tomato in Ethiopia. Eur. J. Plant Pathol., 140:677-688. doi: http://dx.doi.org/10.1007/s10658-014-0497-3

Kim, N. H., and Hwang, B. K. (2015). Pepper heat shock protein 70a interacts with the type III effector AvrBsT and triggers plant cell death and immunity. Plant Physiol., 167:307-322. doi: http://dx.doi.org/10.1104/pp.114.253898

Kim, N. H., Choi, H. W., and Hwang, B. K. (2010a). Xanthomonas campestris pv. vesicatoria effector AvrBsT induces cell death in pepper, but suppresses defense responses in tomato. Mol. Plant-Microbe In., 23:1069-1082. doi: http://dx.doi.org/10.1094/mpmi-23-8-1069

Kim, S. H., Olson, T. N., Peffer, N. D., Nikolaeva, E. V., Park, S., and Kang, S. (2010b). First report of bacterial spot of tomato caused by Xanthomonas gardneri in Pennsylvania. Plant Dis., 94:638-638. doi: http://dx.doi.org/10.1094/pdis-94-5-0638b

Kizheva, Y., Vancheva, T., Hristova, P., Stoyanova, M., Bogatzevska, N., and Moncheva, P. (2011). Diversity of Xanthomonas spp. causal agents of bacterial spot on pepper and tomato plants in Bulgaria. Biotechnol. Biotec. Eq., 25(sup1), 98-104. doi: http://dx.doi.org/10.5504/bbeq.2011.0126

Kornev, K. P., Matveeva, E. V., Pekhtereva, E. S., Polityko, V. A., Ignatov, A. N., Punina, N. V., and Schaad, N. W. (2009). Xanthomonas species causing bacterial spot of tomato in the Russian federation. Acta Hortic., 808:243-246. doi: http://dx.doi.org/10.17660/ActaHortic.2009.808.36

Kousik, C., and Ritchie, D. (1996). Race shift in Xanthomonas campestris pv. vesicatoria within a season in field-grown pepper. Phytopathology, 86:952-958.

Kumar, S., Stecher, G., and Tamura, K. (2016). MEGA7: Molecular Evolutionary Genetics Analysis Version 7.0 for Bigger Datasets. Mol. Biol. Evol., 33:1870-1874. doi: http://dx.doi.org/10.1093/molbev/msw054

Lewis Ivey, M. L., Strayer, A., Sidhu, J. K., and Minsavage, G. V. (2016). Bacterial leaf spot of tomato (Solanum lycopersicum) in Louisiana is caused by Xanthomonas perforans, tomato race 4. Plant Dis., 100:1233-1233. doi: http://dx.doi.org/10.1094/pdis-12-15-1451-pdn

Lewis, J. D., Lee, A. M. Y., Ma, W., Zhou, H., Guttman, D. S., and Desveaux, D. (2011). The YopJ superfamily in plant-associated bacteria. Mol. Plant Pathol., 12:928-937. doi: http://dx.doi.org/10.1111/j.1364-3703.2011.00719.x

Louws, F. J., Wilson, M., Campbell, H. L., Cuppels, D. A., Jones, J. B., Shoemaker, P. B., Sahin, F., and Miller, S. A. (2001). Field control of bacterial spot and bacterial speck of tomato

151

using a plant activator. Plant Dis., 85:481-488. doi: http://dx.doi.org/10.1094/pdis.2001.85.5.481

Ma, K.-W., and Ma, W. (2016). YopJ family effectors promote bacterial infection through a unique acetyltransferase activity. Microbiol. Mol. Biol. R., 80:1011-1027. doi: http://dx.doi.org/10.1128/mmbr.00032-16

Ma, X., Lewis Ivey, M. L., and Miller, S. A. (2011). First report of Xanthomonas gardneri causing bacterial spot of tomato in Ohio and Michigan. Plant Dis., 95:1584-1584. doi: http://dx.doi.org/10.1094/pdis-05-11-0448

Maiden, M. C. J., van Rensburg, M. J. J., Bray, J. E., Earle, S. G., Ford, S. A., Jolley, K. A., and McCarthy, N. D. (2013). MLST revisited: the gene-by-gene approach to bacterial genomics. Nat. Rev. Micro., 11:728-736. doi: http://dx.doi.org/10.1038/nrmicro3093

Marco, G., and Stall, R. (1983). Control of bacterial spot of pepper initiated by strains of Xanthomonas campestris pv. vesicatoria that differ in sensitivity to copper. Plant Dis., 67:779-781.

Markowitz, V. M., Chen, I. M. A., Palaniappan, K., Chu, K., Szeto, E., Grechkin, Y., Ratner, A., Jacob, B., Huang, J., Williams, P., Huntemann, M., Anderson, I., Mavromatis, K., Ivanova, N. N., and Kyrpides, N. C. (2012). IMG: the integrated microbial genomes database and comparative analysis system. Nucleic Acids Res., 40:115-122. doi: http://dx.doi.org/10.1093/nar/gkr1044

Martin, H. L., Hamilton, V. A., and Kopittke, R. A. (2004). Copper tolerance in Australian populations of Xanthomonas campestris pv. vesicatoria contributes to poor field control of bacterial spot of pepper. Plant Dis., 88:921-924. doi: http://dx.doi.org/10.1094/pdis.2004.88.9.921

McGuire, R. G., Jones, J. B., and Scott, J. W. (1991). Epiphytic populations of Xanthomonas campestris pv. vesicatoria on tomato cultigens resistant and susceptible to bacterial spot. Plant Dis., 75:606-609.

McInnes, T., Gitaitis, R., McCarter, S., Jaworski, C., and Phatak, S. (1988). Airborne dispersal of bacteria in tomato and pepper transplant fields. Plant Dis., 72:575-579.

Minsavage, G., Dahlbeck, D., Whalen, M., Kearney, B., Bonas, U., Staskawicz, B., and Stall, R. (1990). Gene-for-gene relationships specifying disease resistance in Xanthomonas campestris pv. vesicatoria―pepper interactions. Mol. Plant-Microbe In., 3:41-47.

Mirik, M., Aysan, Y., and Cinar, O. (2007). Copper-resistant strains of Xanthomonas axonopodis pv. vesicatoria (Doidge) Dye in the eastern Mediterranean region of Turkey. J. Plant Pathol., 89:153-154.

152

Nadworny, P., Omotoso, O., and Zheng, Z. (2015). Comparison of oxysilver nitrate and oxysilver bisulfate, Part I: Synthesis and physicochemical properties. Polyhedron, 99: 204- 215. doi: https://doi.org/10.1016/j.poly.2015.07.038

NASS. 2008. Vegetables 2007 Summary, Census of Agriculture, United States Summary and State Data. United States Department of Agriculture, National Agricultural Statistics Service. January 2008.

NASS. 2016. Vegetables 2015 Summary, Census of Agriculture, United States Summary and State Data. United States Department of Agriculture, National Agricultural Statistics Service. February 2016.

Noël, L., Thieme, F., Nennstiel, D., and Bonas, U. (2001). cDNA-AFLP analysis unravels a genome-wide hrpG-regulon in the plant pathogen Xanthomonas campestris pv. vesicatoria. Mol. Microbiol., 41:1271-1281. doi: http://dx.doi.org/10.1046/j.1365-2958.2001.02567.x

Obradovic, A., Jones, J. B., Balogh, B., and Momol, M. T. (2008). Integrated Management of Tomato Bacterial Spot. In A. Ciancio and K. G. Mukerji (Eds.), Integrated Management of Diseases Caused by Fungi, Phytoplasma and Bacteria (pp. 211-223). Dordrecht: Springer Netherlands. Obradovic, A., Jones, J. B., Momol, M. T., Balogh, B., and Olson, S. M. (2004). Management of tomato bacterial spot in the field by foliar applications of bacteriophages and SAR inducers. Plant Dis., 88:736-740. doi: http://dx.doi.org/10.1094/pdis.2004.88.7.736

Obradovic, A., Jones, J. B., Momol, M. T., Olson, S. M., Jackson, L. E., Balogh, B., Guven, K., and Iriarte, F. B. (2005). Integration of biological control agents and systemic acquired resistance inducers against bacterial spot on tomato. Plant Dis., 89:712-716. doi: http://dx.doi.org/10.1094/pd-89-0712

Orth, K., Xu, Z., Mudgett, M. B., Bao, Z. Q., Palmer, L. E., Bliska, J. B., Mangel, W. F., Staskawicz, B., and Dixon, J. E. (2000). Disruption of signaling by Yersinia effector YopJ, a ubiquitin-like protein protease. Science, 290:1594-1597. doi: http://dx.doi.org/10.1126/science.290.5496.1594

Osdaghi, E., Taghavi, S. M., Hamzehzarghani, H., and Lamichhane, J. R. (2016). Occurrence and characterization of the bacterial spot pathogen Xanthomonas euvesicatoria on pepper in Iran. J. Phytopathol., 164:722-734. doi: http://dx.doi.org/10.1111/jph.12493

Osdaghi, E., Taghavi, S. M., Hamzehzarghani, H., Fazliarab, A., & Lamichhane, J. R. (2017). Monitoring the occurrence of tomato bacterial spot and range of the causal agent Xanthomonas perforans in Iran. Plant Pathology, 66(6), 990-1002.doi: http://dx.doi.org/10.1111/ppa.12642

153

Paret, M. L., Vallad, G. E., Averett, D. R., Jones, J. B., and Olson, S. M. (2012). Photocatalysis: effect of light-activated nanoscale formulations of TiO2 on Xanthomonas perforans and control of bacterial spot of tomato. Phytopathology, 103:228-236. doi: http://dx.doi.org/10.1094/phyto-08-12-0183-r

Pearson, T., Busch, J. D., Ravel, J., Read, T. D., Rhoton, S. D., U'Ren, J. M., Simonson, T. S., Kachur, S. M., Leadem, R. R., Cardon, M. L., Van Ert, M. N., Huynh, L. Y., Fraser, C. M., and Keim, P. (2004). Phylogenetic discovery bias in Bacillus anthracis using single- nucleotide polymorphisms from whole-genome sequencing. P. Natl. Acad. Sci. USA, 101:13536-13541. doi: http://dx.doi.org/10.1073/pnas.0403844101

Pearson, T., Okinaka, R. T., Foster, J. T., and Keim, P. (2009). Phylogenetic understanding of clonal populations in an era of whole genome sequencing. Infection, Genetics and Evolution, 9:1010-1019. doi: http://dx.doi.org/10.1016/j.meegid.2009.05.014

Pernezny, K., and Collins, J. (1997). Epiphytic Populations of Xanthomonas campestris pv. vesicatoria on Pepper: Relationships to Host-Plant Resistance and Exposure to Copper Sprays. Plant Dis., 81:791-794. doi: http://dx.doi.org/10.1094/pdis.1997.81.7.791

Pernezny, K., Nagata, R., Havranek, N., and Sanchez, J. (2008). Comparison of two culture media for determination of the copper resistance of Xanthomonas strains and their usefulness for prediction of control with copper bactericides. Crop Prot., 27:256-262. doi: http://dx.doi.org/10.1016/j.cropro.2007.05.012

Pohronezny, K., and Volin, R. (1983). The effect of bacterial spot on yield and quality of fresh market tomatoes. HortScience, 18:69-70.

Pohronezny, K., Hewitt, M., Infante, J., and Datnoff, L. (1992). Wind and wind-generated sand injury as factors in infection of pepper by Xanthomonas campestris pv. vesicatoria. Plant Dis., 76:1036-1039.

Pohronezny, K., Moss, M. A., Dankers, W., and Schenk, J. (1990). Dispersal and management of Xanthomonas campestris pv. vesicatoria during thinning of direct-seeded tomato. Plant Dis., 74:800-805.

Pohronezny, K., Waddill, V., Schuster, D., and Sonoda, R. (1986). Integrated pest management for Florida tomatoes. Plant Dis., 70:96-102.

Popov, G., Fraiture, M., Brunner, F., and Sessa, G. (2016). Multiple Xanthomonas euvesicatoria type III effectors inhibit flg22-triggered immunity. Mol. Plant-Microbe In., 29:651-660. doi: http://dx.doi.org/10.1094/mpmi-07-16-0137-r

154

Potnis, N., Krasileva, K., Chow, V., Almeida, N. F., Patil, P. B., Ryan, R. P., Sharlach, M., Behlau, F., Dow, J. M., Momol, M., White, F. F., Preston, J. F., Vinatzer, B. A., Koebnik, R., Setubal, J. C., Norman, D. J., Staskawicz, B. J., and Jones, J. B. (2011). Comparative genomics reveals diversity among xanthomonads infecting tomato and pepper. BMC Genomics, 12:146. doi: http://dx.doi.org/10.1186/1471-2164-12-146

Potnis, N., Timilsina, S., Strayer, A., Shantharaj, D., Barak, J. D., Paret, M. L., Vallad, G. E., and Jones, J. B. (2015). Bacterial spot of tomato and pepper: diverse Xanthomonas species with a wide variety of virulence factors posing a worldwide challenge. Mol. Plant Pathol., 16:907- 920. doi: http://dx.doi.org/10.1111/mpp.12244

Rai, M. K., Deshmukh, S. D., Ingle, A. P., and Gade, A. K. (2012). Silver nanoparticles: the powerful nanoweapon against multidrug-resistant bacteria. J. App. Microbiol., 112:841-852. doi: http://dx.doi.org/10.1111/j.1365-2672.2012.05253.x

Rashid, T. S., Kamaruzaman, S., Golkhandan, E., Nasehi, A., and Awla, H. K. (2016). First report of Xanthomonas gardneri causing bacterial spot of tomato in Malaysia. Plant Dis., 100:854-854. doi: http://dx.doi.org/10.1094/pdis-08-15-0886

Richter, M., and Rosselló-Móra, R. (2009). Shifting the genomic gold standard for the prokaryotic species definition. P. Natl. Acad. Sci. USA, 106:19126-19131. doi: http://dx.doi.org/10.1073/pnas.0906412106

Ritchie, D., and Dittapongpitch, V. (1991). Copper-and streptomycin-resistant strains and host differentiated races of Xanthomonas campestris pv. vesicatoria in North Carolina. Plant Dis., 75:733-736.

Roberts, P. D., Momol, M. T., Ritchie, L., Olson, S. M., Jones, J. B., and Balogh, B. (2008). Evaluation of spray programs containing famoxadone plus cymoxanil, acibenzolar-S-methyl, and Bacillus subtilis compared to copper sprays for management of bacterial spot on tomato. Crop Prot., 27:1519-1526. doi: http://dx.doi.org/10.1016/j.cropro.2008.06.007

Roden, J. A., Belt, B., Ross, J. B., Tachibana, T., Vargas, J., and Mudgett, M. B. (2004). A genetic screen to isolate type III effectors translocated into pepper cells during Xanthomonas infection. P. Natl. Acad. Sci. USA, 101:16624-16629. doi: http://dx.doi.org/10.1073/pnas.0407383101

Römer, P., Hahn, S., Jordan, T., Strauß, T., Bonas, U., and Lahaye, T. (2007). Plant pathogen recognition mediated by promoter activation of the pepper Bs3 resistance gene. Science, 318:645-648. doi: http://dx.doi.org/10.1126/science.1144958

Rouli, L., Merhej, V., Fournier, P. E., and Raoult, D. (2015). The bacterial pangenome as a new tool for analysing pathogenic bacteria. New Microb. New Infec., 7:72-85. doi: http://dx.doi.org/10.1016/j.nmni.2015.06.005

155

Saddler, G. S., and Bradbury, J. F. (2005). Xanthomonas. In D. J. Brenner, N. Krieg & J. Staley (Eds.), Bergey’s Manual of Systematic Bacteriology, Vol 2: The Proteobacteria (2nd ed., pp. 63-90). NY: Springer.

Sahin, F., and Miller, S. A. (1998). Resistance in Capsicum pubescens to Xanthomonas campestris pv. vesicatoria pepper race 6. Plant Dis., 82:794-799. doi: http://dx.doi.org/10.1094/pdis.1998.82.7.794

Schürch, A. C., and Siezen, R. J. (2010). Genomic tracing of epidemics and disease outbreaks. Microb. Biotechnol., 3: 628-633.

Schwartz, A. R., Potnis, N., Timilsina, S., Wilson, M., Patane, J., Martins, J., Minsavage, G. V., Dahlbeck, D., Akhunova, A., Almeida, N., Vallad, G. E., Barak, J. D., White, F. F., Miller, S. A., Ritchie, D., Goss, E., Bart, R. S., Setubal, J. C., Jones, J. B., and Staskawicz, B. J. (2015). Phylogenomics of Xanthomonas field strains infecting pepper and tomato reveals diversity in effector repertoires and identifies determinants of host specificity. Front. Microbiol., 6:535. doi: http://dx.doi.org/10.3389/fmicb.2015.00535

Scott, J., Hutton, S., Jones, J., Francis, D., and Miller, S. (2006). Resistance to bacterial spot race T4 and breeding for durable, broad-spectrum resistance to other races. Rpt. Tomato Genet. Coop., 56:33-36.

Sharabani, G., Manulis-Sasson, S., Borenstein, M., Shulhani, R., Lofthouse, M., Chalupowicz, L., and Shtienberg, D. (2013). The significance of guttation in the secondary spread of Clavibacter michiganensis subsp. michiganensis in tomato greenhouses. Plant Pathol., 62:578-586. doi: http://dx.doi.org/10.1111/j.1365-3059.2012.02673.x

Stall, R. E., Beaulieu, C., Egel, D., Hodge, N. C., Leite, R. P., Minsavage, G. V., Bouzar, H., Jones, J. B., Alvarez, A. M., and Benedict, A. A. (1994). Two genetically diverse groups of strains are included in Xanthomonas campestris pv. vesicatoria. Int. J. Syst. Evol. Micr., 44:47-53. doi: http://dx.doi.org/10.1099/00207713-44-1-47

Stall, R. E., Jones, J. B., and Minsavage, G. V. (2009). Durability of resistance in tomato and pepper to Xanthomonads causing bacterial spot. Annu. Rev. Phytopathol., 47:265-284. doi: http://dx.doi.org/10.1146/annurev-phyto-080508-081752

Stall, R. E., Loschke, D. C., and Jones, J. B. (1986). Linkage of copper resistance and avirulence loci on a self transmissible plasmid in Xanthomonas campestris pv. vesicatoria. Phytopathology, 76:240-243.

Stall, R., and Thayer, P. (1962). Streptomycin resistance of the bacterial spot pathogen and control with streptomycin. Plant Dis. Rep., 46:389-392.

Strayer, A., Ocsoy, I., Tan, W., Jones, J. B., and Paret, M. L. (2015). Low concentrations of a silver-based nanocomposite to manage bacterial spot of tomato in the greenhouse. Plant Disease, 100:1460-1465. doi: http://dx.doi.org/10.1094/pdis-05-15-0580-re

156

Strayer-Scherer, A. L., Liao, Y.-Y., Young, M., Ritchie, L., Vallad, G. E., Santa, S., Clark, D., Freeman, J. H., Jones, J. B., and Paret, M. L. (2017). Advanced copper composites against copper-tolerant Xanthomonas perforans and tomato bacterial spot. Phytopathology. (In press) doi: http://dx.doi.org/10.1094/phyto-06-17-0221-r

Szczesny, R., Büttner, D., Escolar, L., Schulze, S., Seiferth, A., and Bonas, U. (2010). Suppression of the AvrBs1-specific hypersensitive response by the YopJ effector homolog AvrBsT from Xanthomonas depends on a SNF1-related kinase. New Phytol., 187:1058-1074. doi: http://dx.doi.org/10.1111/j.1469-8137.2010.03346.x

Tamura, K., Stecher, G., Peterson, D., Filipski, A., and Kumar, S. (2013). MEGA6: Molecular Evolutionary Genetics Analysis Version 6.0. Mol. Biol. Evol., 30:2725-2729. doi: http://dx.doi.org/10.1093/molbev/mst197

Teper, D., Salomon, D., Sunitha, S., Kim, J.-G., Mudgett, M. B., and Sessa, G. (2014). Xanthomonas euvesicatoria type III effector XopQ interacts with tomato and pepper 14–3–3 isoforms to suppress effector-triggered immunity. Plant J., 77:297-309. doi: http://dx.doi.org/10.1111/tpj.12391

Tettelin, H., Masignani, V., Cieslewicz, M. J., Donati, C., Medini, D., Ward, N. L., Angiuoli, S. V., Crabtree, J., Jones, A. L., Durkin, A. S., DeBoy, R. T., Davidsen, T. M., Mora, M., Scarselli, M., Margarit y Ros, I., Peterson, J. D., Hauser, C. R., Sundaram, J. P., Nelson, W. C., Madupu, R., Brinkac, L. M., Dodson, R. J., Rosovitz, M. J., Sullivan, S. A., Daugherty, S. C., Haft, D. H., Selengut, J., Gwinn, M. L., Zhou, L., Zafar, N., Khouri, H., Radune, D., Dimitrov, G., Watkins, K., O'Connor, K. J. B., Smith, S., Utterback, T. R., White, O., Rubens, C. E., Grandi, G., Madoff, L. C., Kasper, D. L., Telford, J. L., Wessels, M. R., Rappuoli, R., and Fraser, C. M. (2005). Genome analysis of multiple pathogenic isolates of Streptococcus agalactiae: Implications for the microbial “pan-genome”. P. Natl. Acad. Sci. USA., 102:13950-13955. doi: http://dx.doi.org/10.1073/pnas.0506758102

Timilsina, S. (2016). Multi-locus and whole-genome sequence anlaysis of pseudomonads and xanthomonads impacting tomato production in Florida. PhD dissertation, University of Florida, Gainesville, FL.

Timilsina, S., Abrahamian, P., Potnis, N., Minsavage, G. V., White, F. F., Staskawicz, B. J., Jones, J. B., Vallad, G. E., and Goss, E. M. (2016). Analysis of sequenced genomes of Xanthomonas perforans identifies candidate targets for resistance breeding in tomato. Phytopathology, 106:1097-1104. doi: http://dx.doi.org/10.1094/phyto-03-16-0119-fi

Timilsina, S., Jibrin, M. O., Potnis, N., Minsavage, G. V., Kebede, M., Schwartz, A., Bart, R., Staskawicz, B., Boyer, C., Vallad, G. E., Pruvost, O., Jones, J. B., and Goss, E. M. (2015). Multilocus sequence analysis of Xanthomonads causing bacterial spot of tomato and pepper plants reveals strains generated by recombination among species and recent global spread of

157

Xanthomonas gardneri. Appl. Environ. Microb., 81:1520-1529. doi: http://dx.doi.org/10.1128/aem.03000-14

Timilsina, S., Vallad, G. E., and Jones, J. B. (2013). Xanthomonas perforans strains from Florida tomatoes exhibit widespread copper tolerance, changing race structure and sensitivity to other bactericides. [Abstract]. Phytopathology, 103:146.

Tudor-Nelson, S. M., Minsavage, G. V., Stall, R. E., and Jones, J. B. (2003). Bacteriocin-like substances from tomato race 3 strains of Xanthomonas campestris pv. vesicatoria. Phytopathology, 93:1415-1421. doi: http://dx.doi.org/10.1094/phyto.2003.93.11.1415

Üstün, S., and Börnke, F. (2015). The Xanthomonas campestris type III effector XopJ proteolytically degrades proteasome subunit RPT6. Plant Physiol., 168(1):107-19. doi: http://dx.doi.org/10.1104/pp.15.00132

Vaidya, G., Lohman, D. J., and Meier, R. (2011). SequenceMatrix: concatenation software for the fast assembly of multi-gene datasets with character set and codon information. Cladistics, 27:171-180. doi: http://dx.doi.org/10.1111/j.1096-0031.2010.00329.x

Vallad, G. E., and Goodman, R. M. (2004). Systemic acquired resistance and induced systemic resistance in conventional agriculture. Crop Sci., 44:1920-1934. doi: http://dx.doi.org/10.2135/cropsci2004.1920

Vallad, G. E., Pernezny, K. L., Balogh, B., Wen, A., L. Figueiredo, J. F., Jones, J. B., Momol, T., Muchovej, R. M., Havranek, N., Abdallah, N., Olson, S., and Roberts, P. D. (2010). Comparison of kasugamycin to traditional bactericides for the management of bacterial spot on tomato. HortScience, 45:1834-1840.

Vallad, G., Timilsina, S., Adkison, H., Potnis, N., Minsavage, G., Jones, J., and Goss, E. (2013). A recent survey of Xanthomonads causing bacterial spot of tomato in Florida provides insights into management strategies. Proc. Fl. Tomato Instit., 25-27.

Vanderplank, J. E. (1968). Disease Resistance in Plants (pp. 206). London-New York: Academic.

Vauterin, L., Swings, J., Kersters, K., Gillis, M., Mew, T. W., Schroth, M. N., Palleroni, N. J., Hildebrand, D. C., Stead, D. E., Civerolo, E. L., Hayward, A. C., Maraîte, H., Stall, R. E., Vidaver, A. K., and Bradbury, J. F. (1990). Towards an improved taxonomy of Xanthomonas. Int. J. Syst. Evol. Micr., 40:312-316. doi: http://dx.doi.org/10.1099/00207713- 40-3-312

Voloudakis, A. E., Reignier, T. M., and Cooksey, D. A. (2005). Regulation of resistance to copper in Xanthomonas axonopodis pv. vesicatoria. Appl. Environ. Microb., 71:782-789. doi: http://dx.doi.org/10.1128/aem.71.2.782-789.2005

158

Wang, J. F. (1992). A new race of Xcv on tomato and screening tomato genotypes for resistance to this new race. Resistance to Xanthomonas campestris pv. vesicatoraia in tomato. Doctoral dissertation, University of Florida, Gainesville, FL.

Wang, H., Hutton, S. F., Robbins, M. D., Sim, S.-C., Scott, J. W., Yang, W., Jones, J. B., & Francis, D. M. (2011). Molecular mapping of hypersensitive resistance from tomato ‘Hawaii 7981’ to Xanthomonas perforans race T3. Phytopathology, 101:1217-1223. doi: http://dx.doi.org/10.1094/phyto-12-10-0345

Wen, A., Balogh, B., Momol, M. T., Olson, S. M., and Jones, J. B. (2009). Management of bacterial spot of tomato with phosphorous acid salts. Crop Prot., 28:859-863. doi: https://doi.org/10.1016/j.cropro.2009.04.013

Wengelnik, K., and Bonas, U. (1996). HrpXv, an AraC-type regulator, activates expression of five of the six loci in the hrp cluster of Xanthomonas campestris pv. vesicatoria. J. Bact., 178:3462-3469. doi: http://dx.doi.org/10.1128/jb.178.12.3462-3469.1996

Wengelnik, K., Van den Ackerveken, G., and Bonas, U. (1996). HrpG, a key hrp regulatory protein of Xanthomonas campestris pv. vesicatoria ls homologous to two-component response regulators. Mol. Plant-Microbe In., 9:704-712.

Werner, N. A., Fulbright, D. W., Podolsky, R., Bell, J., and Hausbeck, M. K. (2002). Limiting Populations and Spread of Clavibacter michiganensis subsp. michiganensis on Seedling Tomatoes in the Greenhouse. Plant Dis., 86:535-542. doi: http://dx.doi.org/10.1094/pdis.2002.86.5.535

Whalen, M., Richter, T., Zakhareyvich, K., Yoshikawa, M., Al-Azzeh, D., Adefioye, A., Spicer, G., Mendoza, L. L., Morales, C. Q., Klassen, V., Perez-Baron, G., Toebe, C. S., Tzovolous, A., Gerstman, E., Evans, E., Thompson, C., Lopez, M., and Ronald, P. C. (2008). Identification of a host 14-3-3 protein that interacts with Xanthomonas effector AvrRxv. Physiol. Mol. Plant Path., 72:46-55. doi: http://dx.doi.org/10.1016/j.pmpp.2008.05.006

White, F. F., Potnis, N., Jones, J. B., and Koebnik, R. (2009). The type III effectors of Xanthomonas. Mol. Plant Pathol., 10:749-766. doi: http://dx.doi.org/10.1111/j.1364- 3703.2009.00590.x

Zhang, Y., Callaway, E. M., Jones, J. B., & Wilson, M. (2009). Visualisation of hrp gene expression in Xanthomonas euvesicatoria in the tomato phyllosphere. Eur. J. Plant Pathol. 124:379-390. doi: http://dx.doi.org/10.1007/s10658-008-9423-x

159

BIOGRAPHICAL SKETCH Peter Abrahamian was born in Lebanon. He attended the American University of Beirut

(AUB) in Beirut, Lebanon for a Bachelor of Science degree in computer science. Knowing that computer science was not his call, he changed majors and pursued a degree in Agricultural

Sciences and Ingenieur Agricole. Peter graduated with distinction from AUB in June 2011. Peter pursued a Master of Science in plant protection at AUB under Dr. Yusuf Abou-Jawdah and graduated in June 2013.

In August 2013, Peter started his doctorate degree in plant pathology at the University of

Florida, Gainesville. He continued his doctoral studies under the supervision of Dr. Gary E.

Vallad and Dr. Jeffrey B. Jones. Peter spent a year at the Jones Lab in Gainesville working on bacterial spot. Later, he moved to the Gulf Coast REC in Balm, FL to conduct his field experiments.

160