|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 DIVERSITY AND CONTROL OF RALSTONIA SOLANACEARUM IN THE SOUTHEASTERN UNITED STATES By JASON CHRISTOPHER HONG 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 2011
2 2011 Jason Christopher Hong
3 To my wife Nina and son Aiden for their love, encouragement, and bring joy in my life. We share this degree.
4 ACKNOWLEDGMENTS I would like to thank Dr Jeff Jones for his support and encouragement. It was through his mentoring and belief in me that I gained the confidence to become a better scientist. I would like to thank Dr David Reed for introducing me to phylogenetics and his counsel regarding the simple sequence repeats. I would like to thank Dr Dave Norman for his help in running different experiments in his lab and the counsel and encouragement he gave thro ugh the years. I would like to thank Dr Steve Olson for mentoring me during the fieldwork. I am sorry for destroying the pressure gages, CO 2 tanks, and anything else I might have touched. I would like to thank Dr Caitlyn Allen for allowing me to come int o her lab and learn essential techniques for working with Ralstonia I would also like to thank Dr Tim Momol for introducing me to Ralstonia It was because of you I strived to become a better scientist. I would like to thank all the biological scientists, lab technicians, faculty, staff, and students that I worked with since starting my project. I am grateful for Jerry Minsavage, Dr Robert Stall and the students in Dr. Jones lab for helping me with my experiments and their counsel. I am especia lly grateful for Patrice Champoiseau for his advice and being an ally. I would like to thank the faculty and staff at the NFREC in Quincy, FL, especially Laura Ritchie and Pingsheng Ji for their help with running the experiments I am eternally grateful fo r my family for their love and support during my schooling years. I am grateful for my in laws for their aid and support during this process Most importantly I would like to thank my wife. I am so grateful for your patience and encouragement throughout th e years. Thank you for assisting me in the lab and waiting all those times in the parking lot. I am truly blessed to have you in my life.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 LITERATURE REVIEW ................................ ................................ .......................... 14 Introduction ................................ ................................ ................................ ............. 14 History of Bacterial Wilt ................................ ................................ ........................... 14 Di sease Cycle, Detection, and Epidemiology ................................ ......................... 16 Classification ................................ ................................ ................................ ........... 19 Disease Management ................................ ................................ ............................. 21 Project go als and objectives ................................ ................................ ................... 22 2 MANAGEMENT OF BACTERIAL WILT IN TOMATOES WITH THYMOL AND ACIBENZOLAR S METHYL ................................ ................................ ................... 24 Introduction ................................ ................................ ................................ ............. 24 Materials an d Methods ................................ ................................ ............................ 26 Bacterial Culture and Inoculum Preparation ................................ ..................... 26 Application of Thymol and ASM ................................ ................................ ....... 27 Tomato Plants and Experimental Design ................................ ......................... 27 Disease and Yield Assessment and Statistical Analysis ................................ .. 28 Results ................................ ................................ ................................ .................... 30 Field Experiment 2006 ................................ ................................ ..................... 30 Field Experiment 2008 ................................ ................................ ..................... 30 Discussion ................................ ................................ ................................ .............. 31 3 DIVERSITY AMONG RALSTONIA SOLANACEARUM STRAINS FROM THE SOUTHEASTERN UNITED STATES ................................ ................................ ..... 39 Introduction ................................ ................................ ................................ ............. 39 Materials and Methods ................................ ................................ ............................ 41 Bacterial Culture, Biovar and Inoculum Preparation ................................ ......... 41 Pathogenicity Tests on Tomato, Pepper and Banana a nd HR test on Tobacco. ................................ ................................ ................................ ....... 43 PCR Amplification and Phylogenetic Analysis of the egl ................................ .. 44 Results ................................ ................................ ................................ .................... 47 Bacterial Strains ................................ ................................ ............................... 47
6 Pathogenicity Tests ................................ ................................ .......................... 47 Phylogenetic Analysis of the egl ................................ ................................ ....... 48 Bacteriocin Test Between RS37 and U.S. Southeastern Race 1 Biovar 1 Strains ................................ ................................ ................................ ........... 49 Musa Multiplex PCR and Triploid Musa Pathogenicity Test ............................. 50 Pathogenicity Test on Cucurbits ................................ ................................ ....... 51 Discussion ................................ ................................ ................................ .............. 52 4 MULTIPLE LOCI VARIABLE NUMBER TANDEM REPEAT ANALYSIS AMONG RALSTONIA SOLANACEARUM STRAINS FROM THE SOUTHEASTERN UNITED STATES ................................ ................................ ................................ ... 70 Introduction ................................ ................................ ................................ ............. 70 Materials and Methods ................................ ................................ ............................ 75 Bacterial Cultures ................................ ................................ ............................. 75 Genomic DNA Extraction and Multiplex PCR ................................ ................... 75 Determining SSRs ................................ ................................ ............................ 76 PCR, Sequencing and Phylogenetic Analysis ................................ .................. 76 Results ................................ ................................ ................................ .................... 79 Determining the SSRs ................................ ................................ ...................... 79 Comparis on of SSRs ................................ ................................ ........................ 79 Phylogenetic Analysis of the SSRs ................................ ................................ .. 80 Analysis of MALs ................................ ................................ .............................. 81 Combination of the Three MALs ................................ ................................ ....... 82 Discussion ................................ ................................ ................................ .............. 83 5 OVERALL SUMMARY AND DISCUSSION ................................ .......................... 109 Management of Bacterial Wilt in Tomatoes with Thymol and Acibenzolar S Methyl ................................ ................................ ................................ ................ 109 Diversity Amongst Ralstonia solanacearum strains from the southeastern United States ................................ ................................ ................................ ..... 110 Multiple Loci Variable Number Tandem Repeat Analysis Among Ralstonia solanacearum Strains from the southeastern United States .............................. 112 APPENDIX: SEQUENCE S AND TREES ................................ ................................ .... 114 LITERATURE CITED ................................ ................................ ................................ .. 120 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 134
7 LIS T OF TABLES Table page 1.1 Characteristics of races and their relationship to other sbdivisions of R. solanacearum ................................ ................................ ................................ ..... 23 2.1 Effect thymol and the combination of thymol and ASM on bacterial wilt incidence and marketable yield for tomato cultivars Phoenix, BH669, and FL7514 (fal l 2006, Quincy, FL). ................................ ................................ .......... 36 2.2 Effect of soil fumigation with thymol, foliar application of ASM, and the combination thymol and ASM on tomato plants in bacterial wilt field experiment on disease incidence of the tomato plants and marketable fruit yield in 2008 (fall, Quincy, FL). ................................ ................................ ........... 37 3.1 List of strains used in this study ................................ ................................ .......... 56 3.2 List of strains obtained from GenBank a used for phyl ogenetic analysis ............. 62 4.1 Strains sequenced and used in this study ................................ .......................... 89 4.2 List of strains obtained from GenBank a used for phylogenetic study b ............... 93 4.3 Simple sequence repeats (SSRs) primers designed from the conserved regions of GMI1000, UW551, Molk2, and IPO1690. ................................ ........... 94
8 LIST OF FIGURES Figure page 2.1 Effect of thymol and the combination of thymol and ASM on the number of plants wilted (Graph A) and marketable fruit yield (Graph B) when applied to susceptible and moderately resistant tomato cultivars in a bacterial wilt field experiment (fall 2006, Quincy, FL). Means and SE (standard error of the mean) were the results of 6 replications, and a total of 16 plants per plot. Treatments for each cultivar were tested for significance. Same letter over each bar indicates no significant difference according range test at P = 0.05. Untreated control (UTC). ................................ ................ 34 2.2 Effect of thymol, ASM, and the combination of thymol and ASM on the number of plants wilted (Graph A) and marketable fruit yield (Graph B) when applied to susceptible and moderately resistant tomato cultivars in a bacterial wilt field experiment (fall 2008, Quincy, FL). Means and SE (standard error of the m ean) were the results of 4 replications, and a total of 18 plants per plot. Treatments for each cultivar were tested for significance. Same letter over range test at P = 0.05. Untr eated control (UTC). ................................ ................ 35 3.1 A 50% majority rule consensus rooted tree created by Bayesian analysis of the egl Weighted line indic ates posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. Strains from Carolina are red, Georgia green, Florida blue, Caribbean purple, and unknown location black. ................................ ................................ ..................... 65 3.2 A 50% majority rule consensus unrooted tree created by Bayesian analysis of the egl Weighted line indicates posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. .... 66 3.3 The zone of inhibition, indicated by the red circle, produced by Ralstonia solanacearum strains RS 37 and RS 5. The lawn strain was sprayed 24hr 48hr after the lawn strain was sprayed ................................ .............................. 67 3.4 Gel displaying bands created by using primers for multiplex Musa, lanes 2 4, 759/760 Ralstonia specific primers, lanes 5 7, and one primer set from the Musa multiplex PCR Musa06, lane 8. Lane 1 contains Ladder 10 100 bp markers. Strain RS37 was used in lanes 2, 5 and 8, RS5 in lanes 3 and 6, and 527 in lanes 4 and 7. ................................ ................................ ................... 68 3.5 Comparison of dried M inoculation of water control (WC), RS37 and Race 2 strains; UW2, UW70, and UW170. Same letter over each bar indicates no significant difference ................................ ................................ ......... 69
9 3.6 Comparison of inoculating plants with RS 5 (Race 1), UW2 (Race 2), and RS 37. Same letter over each ............. 69 4.1 Simple sequence repeat 1 as found on Ralstonia solanacearum strain GMI1000. All R. solanacearum strains were obtained from NCBI BLAST. The repeat was highlighted in orange by the author. ................................ ................. 95 4.2 Simple sequence repeat 3 as found on Ralstonia solanacearum strain GMI1000. All R. solanacearum strains were obtained from NCBI BLAST. The repeat was highli ghted in orange by the author. ................................ ................. 95 4.3 Simple sequence repeat 9 as found on Ralstonia solanacearum strain GMI1000. All R. solanacearum strains were obtained from NCBI BLAST. The repeat was highlighted in orange by the author. ................................ ................. 96 4.4 A majority rule co nsensus rooted tree created by Bayesian analysis of SSR 1. Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. Strains from Carolina are red, Georgia green, Florida blu e, Caribbean purple and unknown location black. R. solanacearum was isolated from potato (circle), tobacco (triangle), tomato (diamond), pothos/anthurium (square), and diverse location (star). Strains were not marked if location and host were unknown. The from GenBank. ................................ ................................ ................................ ... 97 4.5 An unrooted majority rule consensus tree created by Bayesian analysis of SSR 1. Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. The m GenBank. ................................ ................................ ................................ ........... 98 4.6 A majority rule consensus rooted tree created by Bayesian analysis of SSR 9. Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. Strains from Carolina are red, Georgia green, Florida blue, Caribbean purple, and unknown location black. R. solanacearum was isolated from potato (circle), tobacco (triangle), tomato (diamond), pothos/anthurium (square), and diverse location (star). Strains were not marked if location and host were unknown. uired from GenBank. ................................ ................................ ................................ ... 99 4.7 An unrooted majority rule consensus tree created by Bayesian analysis of SSR 9. Weighted line indicate s a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. The GenBank. ................................ ................................ ................................ ......... 100
10 4.8 A majority rule consensus rooted tree created by Bayesian analysis of MAL 1. Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. Strains from Carolina are red, Georgia green, Florida blue, Caribbean purple, and unknown location black. R. solanacearum was isolated from potato (circle), tobacco (triangle), tomato (diamond), pothos/anthurium (square), and diverse location (star). Strains were not marked if location and host were unknown. from GenBank. ................................ ................................ ................................ 101 4.9 An unrooted majority rule consensus tree created by Bayesian analysis of MAL 1. Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater fo r parsimony and likelihood. The GenBank. ................................ ................................ ................................ ......... 102 4.10 A maj ority rule consensus rooted tree created by Bayesian analysis of MAL 3. Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. Strains from Carolina are red, Georgia green Florida blue, Caribbean purple, and unknown location black. R. solanacearum was isolated from potato (circle), tobacco (triangle), tomato (diamond), pothos/anthurium (square), and diverse location (star). Strains were not marked if location and host were unknown. uired from GenBank. ................................ ................................ ................................ 103 4.11 An unrooted majority rule consensus tree created by Bayesian analysis of MAL 3. Weighted line indicat es a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. The GenBank. ................................ ................................ ................................ ......... 104 4.12 A majority rule consensus rooted tree created by Bayesian analysis of MAL 9. Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. Strains from Carolina are red, Georgia green, Florida blue, Caribbean purple, and unknown location black. R. solanacearum was isolated from potato (circle), tobacco (triangle), tomato (diamond), pothos/anthuriu m (square), and diverse location (star). Strains were not marked if location and host were unknown. from GenBank. ................................ ................................ ................................ 105 4.13 An unrooted majority rule consensus tree created by Bayesian analysis of MAL 9. Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. The GenBank. ................................ ................................ ................................ ......... 106
11 4.14 A m ajority rule consensus rooted tree created by Bayesian analysis of all three MALs. Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. Strains from Carolina are red, Ge orgia green, Florida blue, Caribbean purple, and unknown location black. R. solanacearum was isolated from potato (circle), tobacco (triangle), tomato (diamond), pothos/anthurium (square), and diverse location (star). Strains were not marked if location and host were acq uired from GenBank. ................................ ................................ ................... 107 4.15 An unrooted majority rule consensus tree created by Bayesian analysis of all three MALs. Weighted lin e indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. Groups A indicates the sequence was acquired from GenBank. ................................ ...... 108 A.1 An example of the diversity of SSR 9 of R. solanacearum strains isolated. Strains without P9 at the end of its name indicate s trains obtained from NCBI BLAST. ................................ ................................ ................................ ... 114 A.2 An example of the diversity of SSR 1 of R. solanacearum strains isolated. Strains withou t P1 at the end of its name indicate strains obtained from NCBI BLAST. ................................ ................................ ................................ ... 115 A.3 A majority rule consensus rooted tree created by Bayesian analysis of SSR 3. Numbers indicated the confidences of correct position of clades based on the sequence was acquired from GenBank. ................................ ..................... 116 A.4 A majority rule consensus unrooted tree created by Bayesian analysis of SSR 3. Numbers indicated the confidences of correct position of clades based on creating indicates the sequence was acquired from GenBank. ................................ ...... 117 A.5 A majority rul e consensus rooted tree created by Bayesian anal ysis of MAL 9 Numbers indicated the confidences of correct position of clades based on the sequence was acquired from GenBank. ................................ ..................... 118 A.6 A majority rule consensus rooted tree created by parsimony analysis of MAL 9 Numbers indicated the confidences of correct pos ition of clades based on the sequence was acquired from GenBank. ................................ ..................... 119
12 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 DIVERSITY AND CONTROL OF RALSTONIA SOLANACEARUM IN THE SOUTHEASTERN UNITED STATES By Jason Christopher Hong May 2011 Chair: Jeffery B. Jones Major: Plant Pathology Ralstonia solanacearum the causal agent of bacterial wilt, is a soil born phytobacterium native to the southeastern U.S. The pathogen can infect over 200 different plant species, including the agriculturally important crops for the southeastern U.S.; tomato, potato, and tobacc o. Once the bacterium is established in the field and conditions are right, within 5 years growers may see 80% or more crop loss R. solanacearum has been described as a species complex due the diversity that exist within the species. Although native stai ns have been a threat for growers, exotic strains have been known to infect new hosts, can be more aggressive, and deem current cultural practices to control the disease ineffective. The nonnative race 3 biovar 2 strains have been listed as a select agent under the Agricultural Bioterrorism Protection Act of 2002 due to the potential economic losses and effect it would have on international trade. In order to identify if an exotic strains has become established, one must first know the characteristics of th e native strains. This study gathered strains from different collections throughout the southeastern U.S. These strains were then characterized
13 based on classical and current methods for typing R. solanacearum strains. Using the current classification, phylogenetic analysis of the egl the majority of strains isolated in the southeast belong to sequevar 7. It was determined that there were also strains belonging to sequevar 4, 5 and not yet defined sequevar. On e of the strains from the undefined sequevar was further characterized by performing a pathogenicity test on triploid bananas and cucurbits and determining if the bacterium is able to produce a bacteriocin. The preferred method identifying relatedness amon gst organisms is by genotypic typing. Comparison of genome would give the greatest amount of information, however it is both timely and expensive. Microsatellites and microsatellite associated loci (MALs) have been used previously to show diversity between genotypes of the same species. MALs have been proven to be effective to show diversity with in human pathogenic bacterial species. This was the first study to use microsatellites and MALs in phytobacteria. Methods for controlling bacterial wilt in the fie ld are constrained to limiting the exposure of pathogen in the field or, when the field is infested changing cultural methods to plant when the pathogen population is low. Previously reports of using Acibenzolar S methyl (ASM) a foliar spray, and thymol, as soil fumigant, as chemical controls were effective limiting disease symptoms and increasing yield production for tomatoes. This study wanted to determine if using a combination of thymol, ASM and moderately resistant plants would elevate the level of efficacy to control bacterial wilt
CHAPTER 1 LITERATURE REVIEW Introduction Ralstonia solanacearum the causal agent of bacterial wilt, is one of the most important phytopathogenic bacteria in the world. The bacterium is a soilborn e pathogen that can cause major economic loss for growers worldwide The pathogen complex is able to cause disease in over 200 different plant species including important cash crops such as tobacco, tomato, potato, and b anana (Hayward, 1995). Due to the economic impact caused by this pathogen it has been studied extensively. R. solanancearum was one of the first plant pathogens to have its genome sequenced (Salanoubat et al. 2002). The bacterium een studied in great detail, and has served as a model system for host pathogen interactions (Genin and Boucher 2002). R. solanacearum is an aerobic Gram negative rod, 0.5 1.5 m in length, and is motile by 1 4 polar flagell a The bacterium is grouped wi th non fluorescent pseudomonads, and is catalase and oxidase positive and forms nitr i te from nitrates (McCarter, 1991). Bacterial colonies on agar surfaces are initially smooth, shiny and opalescent, but become brown with age (Lelliott and Stead, 1987). T he optimum temperature range for the bacterium is 30 37C; however, previous tests have shown tha t the bacterium is unable to grow at 4 or 40 C (Denny, 2006). History of Bacterial Wilt The first report describing the pathogen was by Burrill in 1890 when i t was to confirm that the disease was indeed caused by the pathogen. Six years later the bacterium was reported in other hosts. It was suspected that the bacterium was native to Florida soils due to rep orts of
frequent outbreaks of the disease on solanaceous crops grown on virgin land (Kelman, 1953). The disease first gained attention in the United States in 1903, when losses in the tobacco growing region of in Granville County, North Carolina caused far mers to abandon their land. By 1907, all the farms in the county reported 25 100% crop loss es. Buildings and lands were abandoned, and what were once prosperous cities became ghost towns. From the combined loss of income and the forced s ale of farms, it was estimated that the county lost over $40 million (Kelman, 1953). As a result of the impact the disease had, it is sometimes referred to as Granville wilt when infecting tobacco R. solanacearum can be problematic for communities that grow solanaceous plants in both tropical and temperate regions. The greatest economic damage has been reported on potatoes, tobacco and tomatoes in the southeastern United States, Indonesia, Brazil, Colombia, Taiwan, and South Africa. The Philippines reported average losses of 15% in tomatoes, 10% in eggplant and pepper and 2 5% in tobacco (Zehr, 1969). Along the Amazon basin in Peru, rapid spread of the pathogen threatened to destroy half of the banana plantations ( French and Sequeira, 1968). In Taiwan, incidence of bacterial wilt on tomato has been reported to range from 15 55% in the summer (Wang and Lin, 2005). The disease has affected most states of India, and in extreme incidences growers have reported up to 90% crop loss (Anuratha and Gnanamanickam, 1990). Extensive economic losses were due to outbreaks in potato in Israel (Volcani and Palti, 1960), and Greece (Zachos, 1957). Most of the regions of the world that are impacted by the pathogen are some of the poor est. Due to the demands of growing cash crops to provide money, growers plant the same crop year after year. This creates more
opportunities for the pathogen to become established in the field. Once the field is infested, farmers can no longer grow economi cally important crops (Smith et al., 1998). Presently R. solanacearum is not only a concern pertaining to crop loss, but has become a matter of national security. The race 3 biovar 2 strains were placed on the top 10 list of potential plant pathogens tha t could be hazardous to U.S. crops and this organism was included in the Agricultural Bioterrorism Act of 2002 (Hawks, 2002). Disease C ycle Detection, and Epidemiology In order for the bacterium to cause disease, it must first enter the host. Most of the time R solanacearum strains enter through wounds on the host Normal agricultural practices can result in wounding. The roots can also be injured by nematodes and during normal plant growth as the roots expand and lateral roots are produced (Denny, 2 006). The pathogen can be vector ed by bees or other insects unintentionally as has been observed in bacterial wilt on banana and plantain (Buddenhagen and Kelman 1964). The disease is favored by high temperature and moist soil. The disease develops rapidl y when temperatures are greater than 20C. Other soil factors that affect debris, and soil type (Denny, 2006). The pathogen has been reported to be present in virgin soi l (Kelman, 1953). Once the bacterium has entered and colonized the host, it multiplies and moves systemically through the xylem. Survival is one of the main obstacles a pathogen faces in the disease cycle. R. solanacearum is able to survive in soil in the absence of host plants for up to 2 years S urvival is dependent on soil temperature and soil moisture (McCarter, 1991). The bacterium has been reported to survive in pure water at 20 25C for more than 40 years
(Denny, 2006). The bacterium can overw inter in irrigation water in close association with a symptom atic aquatic weed hosts thus aiding in the spread of the pathogen throughout the waterways in Europe (Elphinstone et al. 1998; Janse et al. 1997) A similar phenomenon was observed in Florida (Hong et al., 2004). Bacterial wilt can cause many symptoms and some are unique and differentiate bacterial wilt from other types of wilt. In Florida, symptoms have been reported mostly during summer tomato production compared to spring production. Sympto ms of the disease in tomato include stunting, wilting of leaves and sometimes complete stem collapse. Advanced stages can include yellowing or browning and occasionally maceration of the pith. In the field, wilting can be observed in the afternoon in the t erminal leaves or on one side of the plant. The following morning no sign of wilt is evident and the plant appears healthy. Depending on the temperature and humidity, complete wilting of the plant will occur between 5 and14 days after the first symptoms oc cur. Bacterial wilt can be distinguished from other wilt diseases by the leaf color. The leaves of infected plants will appear green, while other wilts, such as Fusarium wilt, will develop a yellowing of leaves. Various methods have been used to detect th e pathogen. A simple test to determine if a plant is infected with R. solanacearum is the ooze test. The test is performed by cutting the plant at the crown, and then squeezing the stem. A whitish ooze will be excreted from the surface of the cut. A varian t of this is to place the cut surface of the stem in water and in a few minutes bacteria will stream out, resulting in visible milky strands emanating from the base of the stem. Recently, increased interest in methods of detection and identification of R. solanacearum has occurred as a result
of R3B2 being considered a threat to potato production in temperate environments (Caruso et al., 2005, Elphinstone et al., 2005, and Janse et al., 2005). Detection and identification assays have been designed to distin guish R. solanacearum from other species of phytobacteria. In some instances the bacterium may be present at concentrations below the limits of detection. In those instances, the population must be enriched to reach a detectable count. The semi selective m edium, SMSA, or the R solanacearum strains (Englebrecht, 1994 and French et al., 1995). Serological techniques such as ELISA or flow cytometry have been used for identif ication of R solanacearum (Seal, 1998). Immunostrips also be used in the field for quick diagnosis (Agdia, Inc., Elkhart, Indiana). The use of PCR for identification is accurate and can detect strains at low concentrations, approximately 100 CFU/ml (Alvar ez, 2005). Many primer sets have been developed for the identification of R solanacearum (Gillings et al. 1993, Pastrik and Maiss, 2000, Seal et al. 1993, Opina et al., 1997. Two other methods for identification are fatty acid methyl ester (FAME) analys is and BIOLOG (MIDI, 2001, Li and Hayword, 1993). Many virulence factors are associated with p athogenesis by the bacterial wilt pathogen. E xtracellular polysaccharide (EPS1) production is a major virulence factor for the pathogen. A major cause of w ilting is due to EPS which clogs the vascular tissue and prevents sufficient water from reaching the leaves (Buddenhagen and Kelman 1964). As the bacterial population increases, an increased amount of EPS1 is produced (McGravey et al. 1999). The bacterium also produces 6 extracellular plant cell wall degrading enzymes, which are delivered through the type II secretion system (T2SS).
Inactivation of any one of these genes does not completely inhibit the disease (Lui et al., 2005). R. solanacea rum has a type III secretion system (T3SS) common to Gram negative plant pathogenic bacteria. Mutants lacking a functional T3SS are nonpathogenic on hosts (Schell, 2000). The T3SS is used to translocate effector proteins into the plant cell where they supp ress basal defense or aid in nutrient release (Hueck, 1998). Plants have evolved to recognize effectors, thus triggering a rapid defense response called a hypersensitive response (HR). This phenomenon aids plant breeders in developing resistant cultivars. Classification Classification of R. solanacearum has been in question for more than a century (Kelman, 1953). In the past 20 years the bacterium has been classified in 3 different genera (Kelman, 1953, Yabuuchi et al., 1992, and Yabuuchi et al, 1995). Howe ver, due to phenotypic differences fatty acid composition and genotypic traits identified through rRNA DNA hybridization and phylogenetic analysis of 16s rDNA the bacterium was placed in the new genus Ralstonia (Yabuuchi et al, 1995). Although R. sola nacearum strains share common traits, groups within the species have different characteristics such as: host range, optimal temperature for causing disease, and differential utilization of carbohydrates. These differences prompted many researchers to class ify the bacterium to subspecies level. Hayward (1964), distinguished 5 different biovars (Table sugar alcohols. The bacterium has also been divided into subspecies based on the host range (Table 1). Race 1 strains infect tobacco, tomato, many solanaceous weeds, non solanaceous weeds and diploid bananas (Buddenhagen et al. 1962). Additional hosts have
been added to include groundnut, potato, pepper, eggplant, olive, ginge r, strawberry, geranium, and Eucalyptus (Denny, 2006). Race 2 strains infect triploid bananas and members of the Heliconia species (Buddenhagen et al. 1962). Race 3 strains were originally thought to be pathogenic on potato and tomato, but weakly path ogenic on other solanaceous crops. This race is considered to have a narrow host range and is able to survive in colder temperatures than other races (Buddenhagen et al. 1962). Race 4 strains aggressively wilt ginger and can wit tomato, pepper, eggplant and some native weeds (Denny, 2006). Race 5 strains are only found in China and cause wilt on mulberry trees (Denny, 2006). Classical restriction fragment length polymorphism (RFLP) analysis was performed on 62 R. sola nacearum strains representing 4 races and 5 biovars isolated in the Americas, Asia and Oceania (Cook and Sequeira, 1988). Southern blot analysis of restriction endonuclease digested genomic DNAs probed with 9 DNA fragments cloned from R. solanacearum reve aled that this bacterium can be divided into two distinct divisions (Cook and Sequeira, 1988) Phylogenetic analysis of select genes is currently the preferred method for classifying R. solanacearum (Fegan & Prior, 2005). Phylotyping is based on performin g a multiplex PCR, which include s using primers that are specific to 16S 23S internal transcribed spacer region (ITS region). R. solanacearum strains will display 1 of 4 possible bands in the gel ; each band represents a different phylotype. A phylotype cor responds to a group of strains that originate from different regions of the world: phylotype 1, Asia; phylotype 2, the Americas; phylotype 3, Africa; and phylotype 4, Indonesia. The phylotypes are subdivided into sequevars. Sequevars are created by sequenc ing and phylogenetic analysis of the endoglucanase gene ( egl ) Three or more strains that group together in a clade form a sequevar.
Disease Management Control of bacterial wilt is difficult once the pathogen has been introduced into the field. Hence the best method for controlling the disease is for growers to use an integrated approach to lower the impact of bacterial wilt in their production. Control methods can be targeted to 3 components of a production cycle: field preparation, crop production and po st harvest practices (Wang and Lin, 2005). The objective of the field preparation period is to limit introduction of the bacterium into the field, or if the field is infested, the goal is to reduce the pathogen population. In infested fields, resistant or moderately resistant cultivars should be chosen. Resistant genotypes provide the best method for control ; however it is extremely difficult with groundnut being the only host where dominant resistant genes have been identified (Boshou, 2005). Moderately resistant cultivars of tomato are commercially available ; however, they tend to be limited in use to certain geographic regions (McCarter, 1991). Grafting on resistant rootstock has proven to be useful in controlling the disease, while providing the oppor tunity to harvest the desired fruit; usually from susceptible cultivars (Nakaho et al., 2000). If moderately resistant cultivars are not available, it is recommended to use pathogen free transplants. Seedlings can be symptomless carriers of the pathogen (W ang and Lin, 2005). It is also recommended to add soil amendments, such as compost and use a fumigant, such as thymol, when fields are infested with the bacterium (Saddler, 2005, Ji et al, 2007, Santos et al. 2006). It is advised to avoid planting in path ogen infested soil, irrigat e with pathogen free water, and use proper sanitation practices to exclude or reduce th e pathogen (Pradhanang et al., 2005). One of the simplest methods for avoiding disease is to plant in cooler and/or drier conditions that would be less favorable for disease development (Denny, 2006).
The use of non hosts as cover crops for crop rotation has helped to reduce th e incidence of bacterial wilt in the field These non host crops include sorghum sudan, rye and other grass crops (Denny, 2006). High soil moisture can increase the population, thus it is suggested to plant in well drained fields. The goal for controlling disease during production is to maintain low pathogen populations and to minimize spread of the organism. Contamination can be reduced by applying strict sanitation practices such as pathogen free irrigation water, transplants, machinery, and stakes. Excessive irrigation will increase pathogen population; thus irrigation should be minimized and based on water requirements. (Wang and Lin, 2005). Following harvest the intention is to reduce bacterial populations for the next season. The crop residue should be plowed under or destroyed immediately. Infected plants and fruit sometimes are left in the field, thus allowing the pathogen to survive and provide inoculum for the next ye ar. Rotation with cover crops should begin after harvest to minimize weeds that may aid in the survival of the pathogen. Project goals and objectives The objectives of this study were to: Determine the effectiveness of using the combination of thymol, a c ibenzolar S methyl and moderately resistant plants for control of bacterial wilt. Characterize R. solanacearum strains in the southeastern U.S to determine diversity and to identify exotic strains. Investigate the effectiveness of microsatellites for dete rmining infrasubspecies groups, and determine their effectiveness for identifying exotic strains in the southeastern U.S.
Table 1.1 Characteristics of races and their relationship to other sbdivisions of R. solanacearum Race Host Range Geographical Distr ibution Biovar RFLP Division a Optimum Temperature 1 wide Asia, Australia, Americas 3,4 1 I II 30 37C 2 banana other Musa spp Caribbean, Brazil, Philippines 1 II 30 37C 3 primarily potato Worldwide b 2 c II 25 37C 4 ginger Asia 3,4 I 30 37C 5 d mulberry China 5 I 30 37C a Based on restriction fragment length polymorphism (RFLP) analysis b Originating in the Andes, but disseminated worldwide on latently infected potato tubers c Typical race 3 strain are sometimes referred to as biovar 2A. Strai ns from the Amazon basin have been placed in a new biovar, designated by various authors as 2T or N2 d Although originally designated as race 4, the prior designation of the ginger strains as race 4 takes precedence
CHAPTER 2 MANAGEMENT OF BACTER IAL WILT IN TOMATOES WITH THYMOL AND ACIBENZOLAR S METHYL Introduction Bacterial wilt, caused by the soilborne pathogen Ralstonia solanacearum occurs worldwide in tropical and subtropical regions of the world ( Yabuuchi et al., 1995 ) The bacterium can cause disease symptoms in over 200 different plant species ( Buddenhagen and Kelman, 1964; Hayward, 1991). In the southeastern United States economic los s e s for important solanaceous crops including tomato, tobacco, and eggplant can be at tributed to bacterial wilt. The bacterium enters the plant through the root and colonizes the vascular tissue in the stem. In field conditions signs of the disease usually appear in mature tomato plants. Leaves will wilt during the day and recover at night or the early hours of the morning. If the weather is favorable for the disease, with high humidity and high temperatures, complete wilting of the plant will occur and eventually death The leaves of wilted plants remain green and the vascular tissue usual ly turns a brownish yellow in the advanced stages of wilt In the field the disease occurs mostly in areas where water accumulates ; however plants showing signs of the disease can be found sporadically throughout. Plants affected by R solanacearum can al so be stunted due to the lack of water and poor uptake of nutrients. Current integrated management strategies include the use of resistant cultivars, pathogen free transplants, and crop rotation with non host cover crops (Pradhanang et al., 2005). However these strategies have proven to be limited due to the complex nature of soilborne pathogens. Resistant cultivars have been developed for fresh market production in the U.S. ; however growers have only adopted moderately
resistant cultivars (Scott et al. 1995). Resistant cultivars are limited to locations, climate, and strains of the pathogen (Saddler, 2004). Transplants limit the spread of the bacterium, yet due to it being a soilborne pathogen most plants in the field can be infected. Cover crops or c rop rotation can be difficult due to the diverse host range of R. solanacearum strain s and the fact that pathogen is able to survive or colonize various weeds that surround the field (Hayward, 1991). With the limited control measures and the gravity of b acterial wilt on important economical crops, investigating other methods for controlling the disease has become critical. Plants are able to activate a protective mechanism after contact by a pathogen, their metabolites, or by a diverse group of structural ly unrelated organic and inorganic compounds. This phenomenon has been dubbed as systemic acquired resistance (SAR) (Kuc, 2001). SAR inducers are ideal for controlling diseases because they trigger a response that may protect the plant from fungal, bacteri al and viral pathogens if the product is applied at the correct time. Acibenzolar S methyl ( ASM; Actigard 50 WG Syngenta, Basel, Switzerland) is a chemical compound that triggers SAR when applied to plants (Oostendrop et al. 2001). ASM has been used to reduce the incidence of fire blight in pear and apple, bacterial spot and speck in tomato and pepper, and common bunt in wheat seedlings (Louws et al. 2001; Lu et al. 2006; Norelli et al. 2003; Obradovic et al. 2005). Previously it wa s reported that ASM enhanced host resistance in moderately resistant tomato cultivars against bacterial wilt ( Pradhanang et al. 2005) Thymol (2 isopropyl 5 methylphenol) is a monoterpene phenol derivative of thyme (Aeschbach et al., 1994). Essential oil s have been used in the past for flavoring and preserving food, for their antioxidant power, and for their antimicrobial activity (Lambert
et al., 2001; Rojano et al., 2008; Scheie, 1989). Both medical and food sciences have shown that thymol is able to in hibit both Gram positive and Gram negative bacteria (Cailet and Lacroix, 2006; Evans and Martin, 2000; Lambert et al., 2001; Shapira and Mimran, 2007; Walsh et al., 2003). Previously thymol applied as a biofumigant was reported to be effective to control b acterial wilt. Thymol applications in the field on susceptible tomato cultivars were able to reduce the incidence of bacterial wilt and increase yield (Ji et al., 2005). In previous studies bacterial wilt was reduced by applying ASM in combination with mo derately resistant tomato cultivars ( Pradhanang et al., 2005 ) or by applying thymol and using susceptible tomato plant s (Ji et al., 2005 ). In this study, we wanted to determine if using a combination of thymol, ASM and moderately resistant plants would elevate the level of efficacy in control ling bacterial wilt. This would be the first time that the two products had been applied together on moderately resistant tomato cultivars in a field trial. It was unknown if the chemicals would work synergistically or would have little to no effect in enhancing disease control. Success with both of the chemicals in controlling the disease would provide another tool in a small arsenal to control bacterial wilt. Materials and Methods Bacterial Culture and Inoculum P reparation R. solanacearum strain RS5 isolated from tomato in Quincy, Florida was used in this study ( Pradhanang and Momol 2001) Pathogenicity was determined by performing s by inoculating tomato plants and re isolating RS5. Bacteria were plated on modified semi selective agar, SMSA (Engelbrecht, 1994) and casamino acid peptone glucose agar, CPG ( Schaad et al. 2001 ). Plates were stored at 28C. The
inoculum contained bacteri a grown on CPG for 24 h and suspended in sterile deionized water. The bacterial suspension was adjusted to 10 7 CFU/ml using sterile deionized water. Inoculum concentration was estimated using a spectrophotometer (Sigma Aldrich Co., Milwaukee, WI) at 600 nm The actual bacterial concentration (cfu/ml) was determined by performing 10 fold dilution s of the inoculum suspension and plating on CPG. Where each tomato plant was to be transplanted, 15 cm holes were created in the soil and 50 ml of the bacterial susp ension was poured into each hole (Ji et al, 2005) The holes were covered with tape prior to the thymol fumigation. Application of T hymol and ASM Thymol was a pplied as a soil fumigant 24 h after the field was inoculated. The field was aerated 7 days post t hymol application by removal of the tape. Thymol was applied at 9.42 kg per ha in a solution consisting of water, 70% ethanol and detergent. ASM was applied as a foliar spray at a volume of 10ml of ASM solution (25g/ml) per plant. The ASM solution was ap plied 6 times: 1 week before the seedlings were transplanted, 1 day after transplanting, followed by 2 treatments that were applied once a week, and then 2 treatments that were applied biweekly. Tomato P lants and Experiment al Design In the 2006 trial tomato cultivars Phoenix FL7514 and BHN669 were used in the field experiment the first being susceptible and the last 2 moderately resistant to bacterial wilt For the 2008 trial, only Phoenix and FL7514 were used. Tomato pla nts were grown in Terra Lite agricultural mix (Scott Sierra Horticultural Products Co., Marysville, OH) in expanded polystyrene flats with 3.5 3.5 cm cells. For each experiment 5 week old tomato seedlings were transplanted 1 week after the thymol applica tion.
The experiment for both years was conducted in experimental fields at the Uni versity of Florida North Florida Research and Education Center located in Quincy Previously, the field s were used for growing tomatoes. The beds were fumigated with methyl bromide (67%) and chloropicrin (33%) at a broadcast equivalent rate of 392 kg a.i./ha for control of weeds and other soilborne pathogens, fertilized with 218 31 181 kg/ha of N P K, and covered with polyethylene mulch. The plots consisted of 4 rows, 5 m lon g with the raised beds, 10 cm high by 91 c m wide and centered 1.8 m apart. Tomato plants were treated with standard foliar sprays for insecticides and fungicides at weekly intervals until harvest. Over time the plants were tied and staked. Experimental plo ts were comprised of 2 rows 10 to 12 m long with 14 tomato seedlings transplanted per block in 2006 and 18 in 2008. Thus each treatment consisted of 84 plants per cultivar in 2006 and 72 plants per cultivar in 2008. In the 2006 experiment each block of plants received one of the following treatments: thymol, the combination of thymol and ASM, or neither thymol nor ASM which was the untreated control (UTC). The treatments for the 2008 experiment consisted of thymol, ASM, both thymol and ASM, or the UTC. I n between each block was a 2 m buffer where no tomato seedlings were planted. A randomized complete block design was used including 6 blocks in 2006 and 4 in 2008. Each block constituted a replication. Disease and Y ield A sse ssment and Statistical Analysis C ompletely wilted tomato p lants were removed from the field weekly and a few of the plants were tested for presence of the bacterium R. solanacearum was confirmed by performing a bacterial ooze test and either isolation on SMSA and confirmation by gas chr omatographic profiling of whole cell fatty acid methyl esters (FAME) (MIDI, Newark, DE), as described previously (Pradhanang et al., 2003; Stead, 1992), or by
using specific immunoassay strips (Agdia, Inc., Elkhart, IN). RS5 was used as positive control fo r each test. In both 2006 and 2008 completely wilted plants were counted weekly after transplanting. Bacterial wilt incidence was recorded at weekly intervals and was quantified as the percentage of plants wilted. P ercentage of plants wilted was calculated by dividing the number of completely wilted plants by total number of transplanted plants. Two harvests were conducted for each crop The total marketable and unmarketable yield was determined for both years according to the USDA standards by using a frui t and vegetable processing machine (Model No. 1650 Roller, TEW Manufacturing Corp., Penfield, NY). Marketable fruit size was categorized as extra large, large and medium (Stavisky et al., 2002; USDA, 1976). The average fruit size and average fruit number w as calculated for each size, cultivar and treatment. The variance of the treatments on bacterial wilt incidence and tomato yield was analyzed by using a general linear model (GLM) conducted in Statistical Analysis System version 9.1 (SAS Institute Cary, NC). To determine the significance of interaction of the treatments, the differences between means of the disease incidence and yield were contrasted using least significant difference (lsd) test The results were tested for normality. In the 200 8 experiment, a week after transplanting, Hurricane Fay descended on the Florida panhandle and did not move for 72 hours. During that time the experimental station received 45 cm of rain. In a normal year for the month of August the station receives on ave rage 19 cm of rain. The w ater collected at the north end of the field and many of the plants were submerged. Some replications of the trial were destroyed, but data was collected from at least 2 of the 4 replications.
Results Field Experiment 2006 Typical bacterial wilt symptoms were observed as early as 1 week post transplanting. Wilted plants were sampled for R. solanacearum by performing a bacterial ooze test, FAME, or using the immunoassay strips. All the plants that were sampled were positive f or the presence of the bacterium. In all the replications the susceptible cultivar, Phoenix, was affected the most by the pathogen; by the end of the experiment the Phoenix plants in the UTC produced the least amount of fruit compared to the 2 resistant cu ltivars, BHN669 and FL7514 ( Figure 2.1 ). Phoenix plants that received the thymol or thymol and ASM treatments had an over 200 fold increase of fruit production, and a 3 fold decrease of plants wilting for thymol and an almost 5 fold for thymol and ASM comp ared to the UTC. By the end of the experiment 94% of the UTC Phoenix plants were completely wilted while 30% of the thymol treated plants wilted and 19% of the thymol and ASM plants wilted ( Table 2.1 ). The plants treated with thymol and ASM resulted over 70% increase in marketable fruit yield and a 30% reduction of disease incidence for all three cultivars. In addition all three cultivars treated with thymol or with thymol and ASM were sig nificant statistical different than the UTC when compari ng disease i ncidence (Table 2.1 ) A significant statistical difference was also observed when FL5714 was treated with thymol and ASM compared to thymol alone (Table 2.1). Field Experiment 2008 Typical bacterial wilt symptoms were observed as early as week 1, and all wilted plants tested by the bacterial ooze test and with the immunoassay strips assay were positive for R. solanacearum In this experiment the Phoenix cultivar survived better
than FL7514, the moderately resistant cultivar, which might be due to the amoun t of rain received from the hurricane. Regardless of the differences between the 2 cultivars, the thymol, ASM and thymol and ASM treated plants resulted in a greater yield and had fewer plants wilt than the untreated contro ls for both cultivars ( Figure 2.2). Even in unfavorable weather conditions, s ignificant statistical difference were observed with the yield for the susceptible plants and both the disease incidence and yield for the moderately resistant plants treated with thymol and ASM when contrast ed with the UTC (Table 2.2). The treatments alone for both cultivars were not statistically significant when contrasted with the UTC. The difference in bacterial wilt incidence between thymol and ASM and the UTC, thymol, or ASM treatments for FL7514 was si gnificant ( Figure 2.2 ). Discussion Controlling bacterial wilt in field conditions has been studied for decades (Kelman, 1953), and to date a single strategy proven to be effective to reduce the incidence of disease or severity of bacterial wilt does not ex ist (Denny, 2006). Factors such as the et al. 2008), the longevity of the bacterium in fallow soil and water (Hayward, 1991), and its ability to persist in infested plant debris ( Granada and Sequeira 1983) have made it difficult to control the disease once it has e stablished itself in the field. Good cultural practices also referred to as Integrated Disease Management (IDM) encompasses multiple strategies for controlling the disease. Included in IDM is avoid ing planting in pathogen infested soil with pathogen free crops irrigating with pathogen free water, and proper sanitation practices of operation tools, which are all important to exclude or reduce the pathogen ( Anith et al., 2004; Champoiseau et al.,
2009; Hong et al., 2008; Denny, 2006 ). Complete resistance is only found in groundnut, but semi resistant cultivars are available, however resistance is limited to geographical location (Denny, 2006). With the decreased use of methyl bromide, alte rnatives to control soil pathogens have been increasingly studied (Martin, 2003 ; Noling and Becker 1994 ; Santos et al. 2006). Thymol ha s proven to be effective in controlling pest s such as fungi, nematodes, insects, and bacteria ( Delespaul, 2000 ; Ji et a l. 2005 ; Lee, 1997 ; et al. 2006). Acibenzolar S methyl too has been proven to be effective against soilborn e fungi nematode s ; Chinnasri et al. 2003 ; Pradhanang et al 2005) Further studies to determine the effectiveness of thymol alone, in a non pretreated field could aid in determining its use as an alternative to methyl bromide. Previously, we determined that the use of thymol and acibenzolar S methyl in field conditions was able to decrease disease incidence and increase fruit yield (Ji et al. 2005 ; Pradhanang et al. 2005). This study was the first time the application of thymol and acibenzolar S methyl was used together in field conditions to control bacteri al wilt on moderately resistant tomato cultivars We report that the use of both products will not have a negative affect on the tomato production. The combination of both products numerically increased the fruit yield and decreased the disease incidence f or the susceptible cultivar In both trials the moderately resistant plants that received the thymol, acibenzolar S methyl, the combination of both chemicals increased fruit yield and the lower disease incidence when compared to the UTC. In both studies th e combination of thymol and acibenzolar S methyl was significantly statistically different
from the UTC, thymol or acibenzolar S methyl alone, when focusing on disease incidence for the different treatments on moderately resistant cultivars. Thus correlati ng to what Pradhanang et al. (2005) reported, in which a greater difference in disease incidence was shown with resistant plants than susceptible plants when both cultivars were treated with acibenzolar S methyl. Susceptible tomato cultivars treated with a cibenzolar S methyl were resistant to the pathogen only when the bacterial populations were low, 10 5 10 6 ; a cibenzolar S methyl was determined ineffective in increasing resistance when the pathogen populations were 10 7 or higher (Anith et al., 2004; Pradhan ang et al., 2005). Again, it is recommend using moderately resistant cultivars to lower disease incidence and for maximum yield. We showed that if a grower were to use both chemicals, neither would be detritus to yield production. As shown in before ment ioned studies, both products are effective at decreasing the incidence of different plant diseases. Thus, the combination of both products could offer a wider protection against multiple biological inhibitory factors. Further studies would include determin ing the minimum inhibitory concentration (MIC) in field conditions for the most effective and economical benefit for the growers. In conjunction with determining the MIC, further studied of the effect of the combination of thymol and acibenzolar S methyl would have on other plant pathogens or on multiple diseases. Further research would also need to grafting could be a new method for controlling the disease (Rivard and Louws 2008).
Figure 2.1 Effect of thymol and the combination of thymol and ASM on the number of plants wilted (Graph A) and marketable fruit yield (Graph B) when applied to susceptible and moderately resistant tomato cultivars in a bacterial wilt field experiment ( fall 2006 Quincy, FL ). Means and SE (standard error of the mean) were the results of 6 replications and a total of 16 plants per plot Treatments for each cultivar were tested for significance. Same letter over each bar range test at P = 0.05. Untreated control (UTC). b
Figure 2.2 Effect of thymol ASM, and the combination of thymol and ASM on the number of plants wilted (Graph A) and marketable fruit y ield (Graph B) when applied to susceptible and moderately resistant tomato cultivars in a bacterial wilt field experiment ( fall 2008 Quincy, FL ). Means and SE (standard error o f the mean) were the results of 4 replications and a total of 18 plants per pl ot Treatments for each cultivar were tested for significance. Same letter over each bar indicates no significant difference according to range test at P = 0.05. Untreated control (UTC).
Table 2. 1 Effect thymol and the combination of thymol and ASM on bacterial wilt incidence and marketable yield for tomato cultivars Phoenix, BH669, and FL7514 (fall 2006, Quincy, FL) Cultivar v Treatment w Disease incidence (%) x Mark etable yield (kg/ha) x Phoenix UTC 94.3 ID y Thymol 30.0 28,100.9 Thymol + ASM 19.3 39,569.4 BH669 UTC 51.4 15,519.3 Thymol 9.3 64,915.4 Thymol + ASM 3.6 76,452.0 FL7514 UTC 66.4 6,294.6 Thymol 12.9 53,225.7 Thymol + ASM 5.0 65,699.6 Contrast z df F P>F df F P>F Phoenix Thymol vs. UTC 1 177.8 0.0001 1 49.1 0.0001 Thymol + ASM vs. UTC 1 204.6 0.0001 1 26.7 0.0004 Thymol + ASM vs Thymol 1 2.79 0.1256 1 1.8 0.2130 BH669 Thymol vs. UTC 1 46.1 0.0001 1 38.6 0.0001 Thymol + ASM vs UTC 1 65.6 0.0001 1 73.7 0.0001 Thymol + ASM vs Thymol 1 2.1 0.1556 1 2.0 0.1895 FL7514 Thymol vs. UTC 1 69.8 0.0001 1 31 0.0002 Thymol + ASM vs. UTC 1 91.4 0.0001 1 48.2 0.0001 Thymol + ASM vs Thymol 1 6.6 0.0277 1 1.8 0.2066 v BHN669 and FL7514 are moderately resistant cultivars, and Phoenix is a susceptible cultivar to bacterial wilt. w Thymol was applied once before transplanting. ASM was applied by foliar spray 6 times: once before transplanting and 5 times afterwards. x Disease incidence was the final percentage of wilted plants. Disease incidence and yield values we re m eans from 6 replications. y Insignificant Data. z Contrast determined by using a GLM (gen eral linear model) and the means of disease incidence and yield treatments for each cultivar were compared using least significant difference (lsd).
3 7 Table 2 2 Effect of soil fumigation with thymol, foliar application of ASM, and the combination thymol and ASM on tomato plants in bacterial wilt field experiment on disease incidence of the tomato plants and marketable fruit yield in 2008 (fall, Quincy, FL). Cultivar w Treatment x Disease incidence (%) y Marketable yield (kg/ha) y Phoenix UTC 38.9 3,061.0 Thymol 37.2 5,316.9 ASM 23.9 6,553.8 Thymol + ASM 26.7 9,440.4 FL7514 UTC 51.7 1,494.3 Thymol 40.6 2,804.5 ASM 38.9 5,018.7 Thymol + ASM 21.1 5,727.8 Contrast z df F P>F df F P>F Phoenix Thymol vs. UTC 1 0.3 0.6519 1 1.4 0.3251 ASM vs. UTC 1 2.1 0.2241 1 5.5 0.0785 Thymol + ASM vs UTC 1 1.3 0.3018 1 7.1 0.0376 Thymol + ASM vs Thymol 1 0.6 0.4923 1 1.2 0.3242 Thymol + ASM vs ASM 1 0.1 0.8498 1 0.1 0.7678 Thymol vs. ASM 1 0.1 0.7406 1 0.1 0.8338 FL7514 Thymol vs. UTC 1 0.6 0.4609 1 1.0 0.3671 ASM vs. UTC 1 1.6 0.2571 1 10.47 0.0178 Thymol + ASM vs UTC 1 15.5 0.0077 1 6.0 0.0499 Thymol + ASM vs. Thymol 1 2.8 0.1556 1 0.6 0.4910
38 Table 2.2 Continued Thymol + ASM vs ASM 1 4.4 0.0805 1 0.2 0.6972 Thymol vs. ASM 1 0.1 0.9001 1 0.35 0.5783 w BHN669 and FL7514 are moderately resistant cultivars, and Phoenix is a susceptible cultivar to bacterial wilt. x Thymol was applied once before transplanting. ASM was applied by foliar spray 6 times: once before transplanting and 5 times afterwards. y Disease incidence was the final percentage of wilted plants. Disease incidence and yield values were means from 4 re plications. z Contrast determined by using a GLM (general linear model) and the means of disease incidence and yield treatments for each cultivar were compared using least significant difference (lsd).
39 CHAPTER 3 DIVERSITY AMONG RALSTONIA SOLANACEAR UM STRAINS FROM THE SOUTHEASTERN UNITED STATES Introduction Bacterial wilt, caused by the bacterial pathogen Ralstonia solanacearum (Yabuuchi et al., 1995), is one of the most destructive bacterial plant diseases in the tropical, sub tropical, and temperate regions of the world. R. solanacearum as a group, is known to infect over 200 different plant species, but more importantly can cause serious yield losses on important agricultural crops such as tomato, potato, pepper eggplant, tobacco, banana and geranium ( Pelargonium ) (Hayward, 1991). Once the bacterium is established in the field and conditions are right, within 5 years growers may experience 80% or more crop loss. As a result of the fear of the potential economic impact and yield loss the bacterium could cause in potato production in the temperate regions of the U.S., R. solanacearum race 3 biovar 2 is listed as a select agent in the U.S. under the Agricultural Bioterrorism Protection Act of 2002 (Hawks, 2002). Cl assification of R. solanacearum has become a highly debated topic for more than a century. In the past 20 years the bacterium has been classified in 3 different genera. Previously the bacterium was placed in the genus Burkholderia but due to phenotypic and genotypic differences, it was transferred to the new genus, Ralstonia (Yabuuchi et al., 1995). Although R.solanacearum strains share common traits, groups within the species have different characteristics such as host range, opt imal temperature for causing disease symptoms, and differential utilization of an array of carbohydrates. These differences prompted many researchers to classify bacterial strains to subspecies level based on biovar determination (Hayward, 1964), host rang e
40 (Buddenhagen & Kelman, 1964), restriction fragment length polymorphism (RFLP) analysis (Cook and Sequeira 1989) and phylogenetic analysis of specific genes (Fegan & Prior, 2005) Phylogenetic analysis has made it possible to determine the diversity of strains quickly and with a degree of accuracy. Previously, various techniques could take up to a month for definitive results (i.e. host tests or biovar test), or the tests were difficult to analyze (i.e. amplified fragment length polymorphisms (AFLP), RF LP, or pulsed field gel electrophoresis). Within the past few years many studies have been published using phylogenetic analyses to determine the diversity of Ralstonia strains (Xu et al., 2009, Liu et al., 2009, Jeong et al., 2007, Lewis Ivey et al., 2007 Sanchez Perez et al., 2008, Cardozo et al., 2009, Toukam et al., 2009, Ji et al., 2007 and Hong et al., 2008). Most R. solanacearum strains isolated from the southeast have been characterized as race 1 biovar 1 (Martin et al. 1982; McLaughilin and Seque ria, 1989; Robertson et al., 2001; Ji et al., 2008), Previous reports had not demonstrated diversity within these strains. Robertson et al. (2004) determined that strains from the Carolinas contained a truncated non functional avirulence gene, compared to the Georgia and Florida strains, which had the full length functional avirulence gene. Strains with the truncated avirulence gene also had a broader host range, including tobacco, whereas strains with the functional avirulence gene, avrA caused a hypersen sitive reaction (HR) on tobacco. Exotic R.solanacearum strains recently have been identified in northern, central, and southern Florida (Ji et al., 2007, Hong et al., 2008 and Norman et al., 2009). These strains may have originated from Asia or the Caribb ean and have a wider host range
41 than the native Florida strains. It is believed that many of the bacterial wilt outbreaks may have been due to these exotic strains instead of the native strains (Norman et al., 2009). Plant species previously considered non hosts of R. solanacearum were reported to develop bacterial wilt symptoms by strains of Asiatic origin (Ji et al., 2007). In 2008 we reported o n an exotic strain that was found in the waterways in northern Florida (Hong et al. 2008) Based on previous research, phylogenetic analysis of the egl of this strain was 100% identical to new emerging strains reported in Martinique (Wicker et al. 2007) In both papers, it was reported that this strain was more aggressive when compared to the typical native stra in. Furthermore th e new strains found in Martinique, although not race 2, was able to colonize and move systemically in triploid banana, without caus ing symptoms. The bacteri al strain also had a larger host range than the typical strains. During the sampli ng that encompassed over 2 years, haplotypes of RS5, the typical northern Florida R. solanacearum strain, w ere never detected. We determined that this exotic strain not only colonized triploid banana, but also was deleterious to the growth of the banana. In this study an extensive analysis was performed on strains isolated from the southeastern U.S. Strains were characterized based on biovar, pathogenicity tests, HR on tobacco, and phylogenetic analyses of egl sequences. Materials and Methods Bacterial C u lture Biovar and Inoculum P reparation All the R. solanacearum strains used in this study are listed in Table 1. Strains were renamed to reflect the origin of labs from where they came. Upon receiving bacterial cultures, strains were streaked on modified s emi selective medium (SMSA;
42 Englebrecht, 1994 ) were transferred to casamino acid peptone glucose (CPG) agar ( Kelman, 1954) and then stored at 80C in 30% glycerol solution. sources as described by Hayward (Hayward, 1991). RS5, race 1 biovar 1, GMI1000, race 1 biovar 3, and UW447, race 3 biovar 2T, were used as standards and positive controls for the biovar tests. A bacterial culture grown on CPG for 12 hr was suspended in st erile tap water and adjusted to approximately10 8 cfu/ml with a spectrophotometer (Sigma Aldrich Co., Milwaukee, WI) The suspension was added to plates containing the panel of carbon sources For all pathogenicity tests, each strain was grown on CPG aga r for 24 hr and then suspended in sterile de ionized water. Suspensions were adjusted to 10 8 cfu/ml. A few postulates. Plants were sampled at the stem and were soaked in 70% et hanol and then flame sterilized. The plant material was crushed using a morta r and pestle The residue was suspended in 1ml of de ionized water and the suspension was streaked on SMSA. The plates were stored at 28C. R. solanacearum colonies were confirmed by PCR using the R. solanacearum specific primers, 759/760 (Opina et al., 1997). The specific details for the PCR amplification are described later in the document. The bacteriocin test consisted of growing a loopful of the producing strain in the cente r of the plate for 24 hr. The indicator strain adjusted to 10 8 cfu/ml was sprayed on the plate and the plates were incubated for 48 hr at 28C and checked every 12 hr for a zone of inhibition. Three different types of media, King B, Luria Bertani and CPG, were tested to determine which would produce the clearest zone. The indicator strain was
43 grown in CPG broth for 24 hr prior to spraying onto the agar surface. To determine if the agent causing the zone of inhibition was a bacteriocin or a lytic b acteriophage, the zone of inhibition was cut out of the agar and washed in 1ml of sterile de ionized water. The sample was then vortexed and centrifuge d at 10,000 rpm and the supernatant was removed and added to CPG broth with the indicator bacterial strai n. The suspension was incubated at 28C on a shaker, set at 300 rpm, for 24 hr. The suspension was then centrifuged at 10,000 rpm and the supernatant was then sterilized using a low protein binding Microcon filter (Amicon, Beverly, MA) with a 0.22 m pore size. The n 1 ml of the filtered product was added to CPG agar A 10 fold series dilution of the susceptible strain was prepared and 0.1 ml was added to the amended CPG agar Plates were then observed for formation of plaques. Pathogenicity T ests o n Tomato, Pepper and Banana and HR test on T obacco. A pathogenicity test was performed using tomato (c.v. Bonny Best) and pepper (c.v. Aruba). Strains were inoculated on 3 to 4 week old plants in 10 cm diameter pots. Root inoculation consisted of wounding the roots and then pouring 30 ml of a bacterial suspension on the soil surrounding in close proximity to the crown of the plant. The inoculated plants were transferred to a growth room with 12 hr light/dark cycle and the air temperature was maintained at 30 C. Eac h trial consisted of 3 plants per strain, and was repeated twice. The Musa pathogenicity test was performed on 3 different Musa genotypes, inoculated with RS5, RS37, and th e race 2 strains UW2, UW70, UW170, and sterile de ionized water. For the first experiment, 8 plants of each genotype were inoculated by root inoculation or by syringe. The plants were at the 3 full leaf stage when inoculated.
44 The root inoculation was perfo rmed as described previously. For syringe inoculation, 1 ml of the inoculum was injected into the plant tissue at the corm. Further Musa Thirty days post inoculation (dp i) plant roots were rinsed, air dried overnight and then conjunction with an analysis of variance (ANOVA) by the SAS System for Windows program (release 9.1 ; SAS, Cary, NC) For the tobacco HR assay, R. solanacearum suspensions were infiltrated into the intercellular space of fully expanded tobacco leaves. Each strain was inoculated twice on the same leaf and inoculations were repeated on 3 other plants. Inoculated plants were placed in a growth chamber with conditions previously described. Infiltrated areas were assessed for HR 24 hr after infiltration. PCR Amplification and Phylogenetic Analysis of the egl Each PCR amplification reaction contained 1 unit of Taq DNA polymerase (GoTaq Flexi DNA Polymerase (Promega Corp., Madison, WI)) with 5.0 l of 5x buffer, 1.5 l of MgCl 2 4.0 l dNTP, each primer at 10 pmol, and 100 ng of DNA. The total volume was adjusted to 25 l with sterile deionized water. The amount of wate r was adjusted if the protocol required more than two primers or a greater amount of DNA. A hot start at 95C for 5 min followed by 30 cycles at 95C for 45 sec, 68C for 30 sec and 72C for 60 sec, and finished with a 10 min extension at 72C was used for the amplification of a DNA sequence in a thermocycler (BioRad, Hercules, CA). The annealing temperature was adjusted according to the composition of the oligonucleotide sequence.
45 Each strain was test ed to determine phylotype and to ensure it was not rac e 3 biovar 2 Phylotype classification, as outlined by Fegan & Prior (2005), consisted of doing PCR using the primers; Nmult21:1F, Nmult21:2F, Nmult23:AF, Nmult22:InF, and Nmult22:RR. The PCR products were visualized on a gel, and according to the band len gth the phylotype was assigned. P rimer pair 630/631 specific for race 3 biovar 2 strains, was used to amplify DNA from each strain (Fegan et al. 1998). DNA from UW551, a race 3 biovar 2 strain, was used as a positive control. The egl gene was amplifi ed by PCR and the product was sequenced and phylogenetically analyzed to determine the relationship of each strain to each other and strains previously described (Table 2). The primer pair JHFegl: GACGATGCATGCCGCTGGTCGC and JHRegl: CACGAACACCACGTTGCTCGCATT GG were designed based on the egl sequences of GMI1000 ( AL646053.1) UW551 ( DQ657596.1 ), and Molk2 ( CU694393.1 ) accessed from Genbank Closely related R. syzygii strains, blood disease bacterium R223 (DQ011538.1) and R. syzygii strain R058 (DQ011543.1) ser ved as outgroups. Amplicons were sequenced in both the forward and reverse directions to create a consensus sequence. The sequences were aligned using Muscle (Edgar, 2004) via http://www.ebi.ac.uk/Tools/muscle/index.html Se Al version 2.0a11 was used for manual manipula tion of the sequences. The program is available at http://tree.bio.ed.ac.uk/software/seal/ (Rambaut, 1996 ). Phylogenetic analysis using maximum parsimony (MP) and maximum likelihood (ML) were run on PAUP* version 4.0b10 (Sworfford, 2003). The MP and ML dendrograms were created by performing tree bisection/reconnection (TBR) branch swapping and heuristic searc hes with random
46 stepwise addition. Modeltest 3.7 was used to select a nucleotide model that best fit the data (Posada and Crandall, 1998). ML settings were adjusted to t he TIM+I+G nucleotide substitution model; the best fit model suggested by rmation criterion. Branch support for both the MP and ML trees was estimated by nonparametric bootstrapping (n=1,000 replicates for MP and n=200 replicates for ML) by TBR swapping (n= 1,000 replicates for MP and n= 100 replicates for ML ; Felsenstein, 1985). RAxML VI HPC was used to run additional ML and ML bootstrap analysis which was performed on the Fisher Cluster at The University of Florida Genetics Institute Gainesville, FL (Stamatakis et al. 2005). The settings in RAxML could not be adjusted to the TI M+I+G model, thus a GTR+G model was used. For bootstrap analysis, 5000 runs were used to generate 5000 trees. Treefig was used to create and visualize the consensus tree (Rambaut, 2006). MrBayes 3.1 was used to perform Baysesian analysis (Huelsenbeck et al 2001). The settings were adjusted to resemble TIM+I+G and were set for 1,000,000 generations being sampled at every 500 generations. The burn in was placed at 88,500, and the consensus tree of the posterior probabilities was created by PAUP* 4.0b10. Pr eviously RS37 was reported to be identical to nonpathogenic banana strains from Martinique (Hong et al. 2008). To determine if RS37 was a Musa strain, the Musa multiplex PCR protocol as described by Prior and Fegan was used (2005). The following primers w ere used : Mus20 F, Mus20 R, Mus35 F, Mus35 R, Mus06 F, Mus06 R, Si28 F and Si28 R. RS5 was used as a negative control and 527, a race 2 strain, was used as a positive control.
47 Results Bacterial S trains All R. solanacearum strains received from the various libraries or isolated and collected within the lab were first streaked on semi selective agar SMSA. The colonies that did not appear as typical R. solanacearum colonies were tested with specific R. solanacearum immunoassa y strips. All strains that were negative for the immunoassay strips (Agdia, Inc., Elkhart, IN) were tested a second time with the strips, and if these failed they were discarded M ultiplex PCR as described by Fegan and Prior, was performed for the remaini ng strains. All of these strains produced a 281 bp product, which is considered positive for the Opina 759/760 universal R. solanacearum primers. All strains, except for 3 isolated from Florida, had a 372 bp band that corresponds to phylotype II. The 3 Flo rid a strains isolated on pepper, were identified as phylotype I due to the 144 bp band they produced The 630/631 primers, specific for race 3 biovar 2 strains, were tested on all the collected strains. None of the strains produced a band with these primers Strains were classified based on their ability to oxidize a panel of different carbon sources. The three pepper strains were identified as biovar 3. These three strains were able to acidify mannitol, sorbitol, dulcitol, trehalose, and maltose. Th e re mainder of the strains corresponded to biovar 1, which only acidified trehalose. Pathogenicity Tests After plating the strains on SMSA it was discovered that some appeared dry and not shiny which is typical of non pathogenic R. solanacearum strains (Table 1) All strains wilted onn y B est tomato plants except for the 6 non mucoid strains. Wilt
48 symptoms were observed 7 12 dpi for all 3 replicates. Of the 142 strains tested 136 caused wilt symptoms in the susceptible cultivar. A few plants from eac h trial were sampled and bacterial colonies were isolated from the plants and then streaked for individual colonies on SMSA. The bacterial colonies were confirmed as R. solanacearum by immunoassay using immuno strips. A pathogenicity test on pepper, culti var Aruba was performed for all the strains. This variety previously was reported to be susceptible to phylotype I strains (Ji et al., 2007) All three of the phylotype I strains wilted pepper within 7 10 dpi. Symptoms were observed on the three replicates. Besides the three phylotype I strains five other strains from Florida and three strains from Georgia also wilted pepper, but we re biovar 1 and phylotype II. A few plants from each trial were sampled for R. solanacearum and the bacterium was c onfirmed using Immuno strips. A n HR test on tobacco was performed because it was previously reported that Florida strains produce an HR while Carolina strains do not HR was observed 24 hr after inoculation. A n HR was observed in 9.7% of the Carolina strai ns, 46.2% of the Georgia strains, and 56.2% of the Florida strains. Even though 38 strains were isolated in Florida, these strains were isolated from plants that originated from the Caribbean (Norman et al. 2009) Phylogenetic analysis of these Caribbean l ike strains grouped them with strains originating from the Caribbean. The total number of strains originating in Florida was 67. Thus, with the adjusted total 88.1% of the Florida strains gave a positive HR on tobacco. Phylogenetic A nalysis of the egl Th e trees created by the three different analyses ( Figure s 3.1 3.2): MP, ML and Bayesian, were similar in respect to the cladist placement of the representative strains
49 from our library and strains acquired from GenBank (Table 3.2). Strains acquired from Gen Bank represent all of the R. solanacearum egl sequences submitted to Genbank. These strains originated from different regions of the world and were isolated from multiple hosts. The phylogenetic trees confirm the phylotype determination based on multiplex PCR in that all the strains are phylotype II, except for strain 232, which will be discussed in Chapter 4. Furthermore, p hylogenetic analysis of the egl reve a led that the strains isolated from Florida were more diverse than strains isolated from the Carolinas and Georgia. The majority of the strains in our collection grouped with the sequevar 7 strains. All the Carolina and Georgia strains were found in this clade. The strains isolated in Florida, including those with exotic origins, grouped with sequevar 4, 5/6, and 7, and an undefined sequevar. The undefined clade did not contain any previously reported strains. The likelihood score using RAxML and a GTR+G model was lnL 2575.71 The likelihood score using Paup*4.0b10 with a TIM+I+G model was lnL 2597.05. Bacteriocin T est B etween RS37 and U S S outheastern Race 1 Biovar 1 S trains RS37 was tested to determine if it produced a bacteriocin that would inhibit the growth o f other race 1 biovar 1 strains. In t he first test RS37 was used as the producing strain and RS5 as the indicator. A zone of inhibition was detected when RS37 was first spotted on the agar and RS5 was sprayed on it 24 hr later. A zone was not detected when RS5 was spotted and RS37 was sprayed, nor when RS5 was tested against itself, or when RS37 was tested against itself ( Figure 3.3). The largest zone of inhibition was Medium B agar when compared to nutrient agar, CPG agar, or LB agar. The t est to determine if the causal agent was a bacteriocin or a lytic phage confirmed it to be a bacteriocin. Attempts were made to isolate the bacteriocin RS37 was grown in
50 nutrient broth and the cells were spun down and the supernatant was extracted and fil tered. The supernatant was poured in well s was sprayed on it immediately, 8 or 24 hr later. However, the bacteriocin was not detected when the supernatant was run through a series of different size filters. RS37 did not have activity against 3 other R. solanacearum strains: AW1, P553, and V 170 representing Alabama, an older Florida strain, collected in 1997, and a typical Carolina strain, respectfully. Musa Multiplex PCR and Triploid Musa Pathogenicity Test Based on our previous study (Hong et al., 2008) it was assumed that RS37 belonged to sequevar 4NPB. The Musa multiplex PCR, a diagnostic test for identifying race 2 strains (Prior and Fegan, 2005), was performed on RS37 with 527, a race 2 strain used as a positive control, and RS5, a race 1 strain used as a negative control. Strain 527 produced a 220 bp band, thus corresponding to sequevar 6. RS5 did not produce a band. RS37 produced a 167 bp band (Fig ure 3.4). As reported, a strain that produces a 167 bp b and also produces a band at 400 bp belongs to sequevar 4. However, RS37 never produced a 400 bp band. Sequevar based on phylogenetic analysis of the partial egl placed RS37 in an undefined clade. It was below sequevars 5 and 6 and above sequevar 19 23 ( Fig ure 3.1). The other strains collected at the same location over a 2 year period belonged to the same clade as RS37 and also produced a 167 bp band for the Musa multiplex. Due to the presence of a positive but unique band in RS37 with the Musa specific prim ers, a triploid banana pathogenicity test was performed. Two different experiments were done. In the first experiment, three different Musa
51 su spension was injected into the corm with a hypodermic needle and syringe. Plants were kept in greenhouse conditions for 80 dpi, and sampled every 10 days. The first symptoms for the race 2 corm inoculated plants were observed 7 10 dpi, while root inoculati on symptoms were observed 15 20 dpi. For sampling, plants were divided in 3 parts; roots, corm to the first leaf, and from the first leaf to the top of the plant with the unfolded leaves removed. The bacterium was recovered mostly in the roots and corm sec tion for the race 2 strains and RS37 for the root inoculated plants. In plants inoculated with RS5, the bacterium was recovered close to the source of inoculation or sometimes not at all. However, in plants inoculated by the syringe injection technique, RS 5 and the other strains were detected in all 3 parts of the plants. The root mass of the 3 Musa genotypes inoculated with the race 2 strains, RS5, and RS37 were weighed. The plants inoculated by the root wounding method had less root mass than those in oculated by syringe injection (f igure not shown). Plants inoculated with UW2 had the lowest mass. RS37 reduced the root mass in all 3 genotypes, however symptoms were more se igure not shown). of the plants inoculated with RS37 were more similar to the root mass of the plants inoculated with race 2 strains than the root masses of the plants inoculated with either water inoculated or RS5 ( Figures 3. 5 an d 3.6). Pathogenicity T est on C ucurbits Wicker et al. (2007) indicated that a new emerging R. solanacearum strain was identified in Martinique These strains had a wider host range than typical biovar 1 strains found on the island. These strains were pathogenic on cucurbits, solanaceous crops, and ornamental plants. RS37 was inoculated on a panel of cucurbit plants to
52 determine if it could cause wilt symptoms. This panel of plants included watermelon 30 dpi for either RS5 or RS37. Discussion In the conti nental U.S. R. solanacearum is primarily found in the southeast. All the native strains are classified as race 1 biovar 1 and belong to phylotype II. Although these strains are similar, differences were noted in HR, host range tests, and phylogenetic analy sis of the egl We determined that the greatest diversity was found in Florida with strains isolated belonging to two different phylotypes and at least two different sequevars. One of the exotic strains isolated in FL, RS37 was further characterized. This strain was able to cause symptoms on tomato plants, like the native strains, but it also was deleterious to growth of triploid banana. The greatest diversity of R. solanacearum strains was found predominantly in Florida. The strains from the Carolinas wer e all identified as the same haplotype. Furthermore, all the Carolina strains reacted similarly to all the tests and analyses performed in this study. The diversity of haplotypes found in Florida could be due to several factors. The most plausible explanat ion is that Florida is a sub tropical region where many exotic crops are imported. The importation of these plants offers an avenue for introduction of exotic strains of the pathogen. Phylogenetic analysis to determine diversity can be a useful tool; howe ver, if not done correctly or interpreted incorrectly it can result in incorrect conclusions. One of the examples of using phylogenetics incorrectly was the sequevar placement of RS37 strain
53 in previous studies. RS37 was isolated in northern Florida and wa s identical to a Caribbean strain based on egl sequence. Previous reports placed this strain and other clones of it in sequevar 5 and 4NPB (Ji et al. 2007; Hong et al., 2008). It appears that those studies incorrectly assigned sequevar due to a lack of tax a used. Hence, phylogenetic trees in other papers may not be correct due to their lack of representative strains or the analysis used is based on old and outdated techniques. Using the current methods for phylogenetic analysis, the trees created are reprod ucible. However support for many sequevar clades was low, being less than 75% for bootstrap and less than 95 for posterior probability. The phylogenetic trees created in this study indicate that strains isolated from Florida were more diverse than those isolated in Georgia and the Carolinas. Yet, phylogenetic analysis was not able to differentiate between strains which caused wilt on pepper or the results of an HR test. The majority of strains from Florida grouped with sequevar 7 strains; however, a few strains grouped with sequevar 4 and 5/6. Previously it was reported that strains belonging to sequevar 4 can infect triploid bananas and can be detected throughout the plant. Some of the sequevar 4 strains are true race 2 strains and cause wilt in triploid banana, while others may not cause disease (Wicker et al. 2007). Most of the strains that grouped within sequevar 4 originated in pothos and anthurium cuttings. A M usa (AAA) with these strains, and they did not cause symptoms (Norman et al., 2009). In this study, we observed that the severity of symptoms caused by RS37 depended on the genotype of banana plant. He nce Musa pathogenicity tests using a variety of
54 genotypes and a Musa multiplex PCR should be performed with all strains that grouped with sequevar 4. As stated previously, it was assumed that RS37 was similar to Caribbean strains that infected triploid b ananas. Musa multiplex PCR was performed to determine if it belonged to a race 2 sequevar. Unlike the race 1 which produced no bands, RS37 produced a single 167bp band. For some sequevar 4 strains this band is amplified, but it is never amplified for the o ther race 2 sequevars. However, all the seqeuvar 4 strains produced a single 400bp band or one band at 400 bp and one at 167 bp. Based on the results for the PCR test, RS37 was inoculated on a panel of M usa genotypes to determine if could cause symptoms Th e bacterium was detected systemically on all three genotypes. The observed symptoms caused by RS37, less root mass and reduced plant growth, were more severe on the ABB genotype than the other genotypes. We also determined that the preferred technique empl oyed to infect Musa plants was by root inoculation. The race 1 strains were periodically detected in the banana leaves. We believe that when the bacterium was injected into the corm, it was readily transported as the leaves grew, thus indicating a false mo vement. Race 2 strains only infect triploid banana. RS37 is unique in that it can infect and wilt tomato and also move systemically through a banana plant. Further research will need to be done to determine the differences and similarities between this str ain and typical race 1 and 2 strains. Hopefully, further research using RS37 could indicate which genes are responsible for determining host specificity. In our previous study it was noted that only haplotypes of RS37 were found and not the native sequeva r 7 strains (Hong et al. 2008). This led us to investigate the
55 potential reasons for RS37 to outcompete RS5. Bacteria that compete for the same resources commonly produce bacteriocins that inhibit other bacteria from growing in close proximity (Bradley, 19 67). The putative bacteriocin from RS37 did not inhibit growth of all representative strains from the collection. This could be attributed to some bacteriocins being specific to the strains they target. Further research could be done to characterize the ba cteriocin for possible use as a control agent in the field. Creating and characterizing a collection of R. solanacearum strains from the southeast provides a reference for future comparison. Future researchers can now determine if a population shift has ta ken place in the Carolinas or if a new strains have been introduced.
56 Table 3. 1 List of strains used in this study Pathogenicity Test d Strains a Origin Host Origin Other Name Year Biovar b Phylotype c Tomato Pepper HR e RS5 FL Tomato 1999 1 II + + + 102 FL Pond water 2004 1 II + 103 FL Pond water 2004 1 II + 104 FL Pond water 2004 1 II + 105 FL Pond water 2004 1 II + 106 FL Pond water 2004 1 II + 107 FL Polygonum pennsylvanicum 2004 1 II + 108 FL P. pennsylvanicum 2004 1 II + 109 FL P. pennsylvanicum 2004 1 II + 110 FL Hydrocotyle ranunculoides 2004 1 II + 111 FL H. ranunculoides 2004 1 II + 112 FL H. ranunculoides 2004 1 II + 113 FL Pond water 2004 1 II + 114 FL Pond water 2004 1 II + + 118 FL Pond water 2004 1 II + 119 FL Pond water 2004 1 II + 120 FL Pond water 2004 1 II + 125 FL Pond water 2004 1 II + 128 FL Pond water 2004 1 II + 130 FL Pond water 2004 1 II + 201 FL Pothos P487 1996 1 II + + 202 FL Tomato P503 NA g 1 II + 203 FL Tomato P504 NA g 1 II + + 204 FL Tomato P505 NA g 1 II + 205 FL Tomato P507 1996 1 II + + 206 FL Tomato P530 1997 1 II + + 207 f FL Tomato P532 1997 1 II +
57 Table 3.1 Continued 208 FL Tomato P533 1997 1 II + + 209 FL Tomato P534 1997 1 II + + 210 f FL Tomato P535 1997 1 II + 211 FL Tomato P536 1997 1 II + + 212 f FL Potato P541 1997 1 II + 213 FL Potato P543 1997 1 II + + 214 FL Potato P550 1997 1 II + + 215 FL Potato P553 1997 1 II + + 216 FL NA g P557 1997 1 II + 217 FL Pothos P564 1996 1 II + + + 218 FL Pothos P573 1999 1 II + + 219 FL Tomato P576 1999 1 II + + 220 FL Tomato P594 2000 1 II + 221 FL NA g 136 NA g 1 II + + 222 FL Heliconia 158 NA g 1 II + + 223 FL Pothos 485 NA g 1 II + + + 224 NA g Pothos NA g 1 II + + + 225 NA g Pothos 517 NA g 1 II + + + 226 NA g Pothos 521 NA g 1 II + + + 227 NA g 528 NA g 1 I + + 228 FL Potato 544 1994 1 II + + 229 NA g Pothos 545 NA g 1 II + + + 230 NA g Pothos 548 NA g 1 II + + 231 NA g Tomato NA g 1 II + + 232 Guadeloupe Tomato 506 1985 1 II + + 233 Martinique Tomato 609 1986 1 I + 234 Martinique Tomato 610 1987 1 II + + 235 NA g Anthurium 618 NA g 1 II + + 236 NA g Anthurium 621 NA g 1 II + + 237 USA Tomato 660 NA g 1 II + + 238 NA g Pothos 673 NA g 1 II + 239 NC 688 NA g 1 II + +
58 Table 3.1 Continued 240 NC Potato 689 NA g 1 II + + 241 NC Tomato 690 NA g 1 II + + 242 NC Tomato 691 NA g 1 II + + 301 f FL Soil RS14 2001 1 II + 302 f FL Tomato RS18 1991 1 II + 303 FL Geranium RS37 2001 1 II + 304 FL Geranium RS38 2001 1 II + + 305 FL Geranium RS39 2001 1 II + 306 FL Geranium RS40 2001 1 II + 307 FL Pond Water RS51 2001 1 II + 308 FL Bidens mitis RS55 2002 1 II + 309 FL P. pennsylvanicum RS56 2002 1 II + 310 FL P. pennsylvanicum RS57 2002 1 II + 311 FL H. ranunculoides RS58 2002 1 II + 312 FL H. ranunculoides RS59 2002 1 II + 313 FL H. ranunculoides RS60 2002 1 II + 314 FL H. ranunculoides RS61 2002 1 II + 315 FL Pond Water RS62 2002 1 II + 316 FL Pond Water RS65 2002 1 II + 317 FL B. mitis RS66 2002 1 II + 318 FL Pond Water RS67 2002 1 II + 319 FL P. pennsylvanicum RS70 2002 1 II + 320 FL Pond Water RS74 2003 1 II + 321 FL Pond Water RS77 2003 1 II + + 322 FL Pond Water RS79 2003 1 II + 323 FL Pond Water RS85 2003 1 II + + 324 FL Pond Water RS89 2003 1 II + 325 FL Pond Water RS90 2003 1 II + + 326 FL Pond Water RS101 2004 1 II + 327 FL Pond Water RS109 2004 1 II + + 328 FL Hydrangea RS116 2005 1 II + + 329 FL Hydrangea RS118 2005 1 II + +
59 Table 3.1 Continued 330 FL Pepper RS121 2005 3 I + + 331 FL Pepper RS122 2005 3 I + + 332 FL Pepper RS123 2005 3 I + + + 333 FL Hydrangea RS124 2005 1 II + 334 FL Geranium RS125 2005 1 II + + 335 FL Geranium RS126 2005 1 II + + 336 FL Geranium RS127 2005 1 II + 337 FL Geranium RS128 2005 1 II + + 338 FL Pond Water RS129 2005 1 II + + 339 FL Pond Water RS130 2005 1 II + 340 FL Tomato RS133 2005 1 II + 341 FL Tomato RS134 2005 1 II + 342 FL Tomato RS135 2005 1 II + + 343 FL Tomato RS136 2005 1 II + + 344 FL Tomato RS137 2005 1 II + + 348 NC Tobacco TD294 NA g 1 II + + 352 GA Tomato TD674 NA g 1 II + 353 FL Tomato RS73 2006 1 II + + 355 FL Tomato RS1 1999 1 II + + 356 f GA Tomato RS3 1999 1 II 357 GA Tomato UW26 1954 1 II + 359 NC Tomato UW123 NA g 1 II 362 NC Tobacco UW203 1969 1 II + 363 NC Tobacco UW209 1969 1 II + 368 FL Tomato NA g 1 II + + 369 FL Tomato NA g 1 II + + 370 NC Tobacco NA g 1 II + 373 GA Tomato NA g 1 II 374 GA Tomato 286 NA g 1 II 375 FL Tomato NA g 1 II 401 GA Tomato Rso81 5 1981 1 II + 402 GA Tobacco Rso96 41 1996 1 II + +
60 Table 3.1 Continued 403 GA Tomato Rso81 2 1981 1 2 + 404 GA Tomato Rso86 2 1986 1 2 + 405 GA Coffee Weed Rso80 1 1980 1 2 + + 406 GA Tomato Rso87 105 1987 1 2 + 407 GA Potato Rso84 1 1984 1 2 + + 501 NC Tobacco K60 NA g 1 2 + 502 NC Tobacco NC116 NA g 1 2 + 503 GA Tobacco GA122 NA g 1 2 + 504 GA Tobacco GA142 NA g 1 2 + 505 SC Tobacco SC06 NA g 1 2 + 506 SC Tobacco SC121 NA g 1 2 + 507 SC Tobacco SC108 NA g 1 2 + 508 NC Tobacco L 17 NA g 1 2 + 509 NC Tobacco L 18 NA g 1 2 + 510 NC Tobacco L 19 NA g 1 2 + 511 NC Tobacco J 12a NA g 1 2 + + 512 NC Tobacco T 24 NA g 1 2 + + 513 NC Tobacco T 30 NA g 1 2 + 514 NC Tobacco T 32 NA g 1 2 + 515 NC Tobacco T 33 NA g 1 2 + 516 NC Tobacco V 167a NA g 1 2 + 517 NC Tobacco V 167b NA g 1 2 + 518 NC Tobacco V 168 NA g 1 2 + 519 NC Tobacco V 170 NA g 1 2 520 NC Tobacco 40a NA g 1 2 + 521 NC Tobacco K66 NA g 1 2 + 522 NC Tobacco K68 NA g 1 2 + 523 NC Tobacco K69 NA g 1 2 + 524 NC Tobacco K71 NA g 1 2 + 525 NC Tomato K74 NA g 1 2 + 526 NC Tomato K136 NA g 1 2 + 527 HI Helicona A3908 NA g 1 2
61 Table 3.1 Continued UW2 NA g NA g NA g 1 2 NA NA NA UW70 Colombia Plantain NA g 1 2 NA NA NA UW17 0 Colombia Heliconia NA g 1 2 NA NA NA AW1 AL Tomato NA g 1 2 + + a from Tim Momol 400 from Ron 500 from Dan Kluepfel and Asimina Mila except for 527 which is part of b Biovar was determined as outlined by Hayward, 1991 c Phylotype was determined by multiplex PCR and confirmed by phylogenetic analysis of the egl as outlined by Fegan and Prior 2005 d Two week old plants were root inoculated with a bacterial suspension at 10 8 cfu/ml, three plants per trial each trial repeated twice. Only when all 3 plants for both trials were considered positive. Plants were stored in growth room conditions at 28 C. e Only fully expanded leaves were inoculated with a bacterial suspension at 10 8 cfu/ml. The HR test was repeated 3 times. Plants were stored in growth room conditions at 28C. f Strains lost pathogenicity and appear dry and circular on agar. g NA=information was not available.
62 Table 3.2 List of strains obtained from GenB ank a used for phylogenetic analysis Strain b Origin b Host b Other No Biovar b Phylotype b GenBank acc. no. A3909 USA Heliconia 1 II/6 EF371812.1 ACH 0 732 Australia Tomato UW433 2 IV/7 GQ907150.1 ANT307 FWI Anthurium CFBP6784 1 II/4 DQ657648.1 Aoyu Australia Potato 2 II/1 FJ561083.1 B1 China Sweet potato 4 I/15 FJ561159.1 Blooddisease R233 ND ND IV DQ011542.1 CFBP734 Madagascar Potato JS767 III/19 to 23 AF295274.1 CFBP765 Japan Tomato JS771 4 I/ND EF371810.1 CFBP1183 Costa Rica Heliconia JS793 1 II/3 EF371805.1 CFBP1409 Honduras Musa sp. K135, JS77 1 II/3 EF371808.1 CFBP2047 USA Tomato 1 II/7 AF295262.1 CFBP2957 FWI Tomato MT5 1 II/5 EF371807.1 CFBP2958 FWI Tomato GT4 1 II/5 AF295266.1 CFBP2968 Guadeloupe Eggplant RUN58 I/13 to 18 EF371806.1 CFBP2972 Martinique Tomato RUN27 1 II/5 EF371809.1 CFBP3059 Burkina Faso Eggplant JCG.AU28 1 III/23 AF295270.1 CFBP3858 Netherlands Potato JS907 1 II/1 AF295259.1 CFBP6786 Martinique Tomato SPV98 1537 II/4 EF371823.1 CIP301 Peru Potato R311 1 II/5 GU295003.1 CIP309 Colombia Potato UW80, S206 2 II/2 EF647735.1 CIP418 Indonesia Peanut MOH6 1 II/3 GU295005.1 Col8 Colombia Desmodium II/4PB EU795341.1 Col41 Colombia Musa sp. II/4PB EU795348.1 DAR64836 Australia Musa sp. 1 II/6 DQ011551.1 E2 China Eggplant 4 I/15 FJ561157.1 E69 China Eggplant 3 I/34 FJ561092.1 EU2 China Eucalyptus 3 I/44 FJ561152.1 GMI1000 F Guyana Tomato JS771, RUN54 3 I/12 or 18 EF192968.1 ICMP6782 Brazil Musa sp. II/6 DQ011553.1 ICMP7963 Kenya Potato RUN55 II/7 AF295263.1 ICMP9600 Brazil Musa sp. II/6 DQ011554.1 IPO1609 Netherlands Potato RUN1 II/1 and 2 EF371814.1 ISBSF1900 Brazil Musa sp. RUN301 II/6 EF371839.1 J25 Kenya Potato N2 III/22 AF295279.1 JT516 Reunion Is. Potato RUN160 II/1 and 2 AF295258.1 JT523 Reunion Is. Potato RUN333 I/13 to 18 AF295252.1
63 Table 3.2 Continued JT525 Reunion Is. Pelargonium asperum RUN60 III/19 to 23 DQ657650.1 JT528 Reunion Is. Potato 1 III/19 AF295273.1 K60 USA Tomato UW25 1 II/7 DQ657614.1 M2 China Mulberry 5 I/48 FJ561067.1 M3 China Mulberry 5 I/44 FJ561106.1 M4 China Mulberry 5 I/12 FJ561107.1 M6 China Mulberry 3 I/48 FJ561109.1 M7 China Mulberry 5 I/12 FJ561110.1 MAFF301558 Japan Potato RUN71 IV/8 or 10 DQ657634.1 MOLK2 Philippines Musa sp. II/3 EF371841.1 NCPPB332 Zimbabwe Potato RUN75 III/19 to 23 DQ657649.1 NCPPB3190 Malaysia Tomato RUN78 I/13 to 18 AF295253.1 NCPPB3987 Brazil Potato RUN81 II/ND AF295261.1 O3 China Olive tree 3 I/44 FJ561069.1 P11 China Peanut 3 I/17 FJ561068.1 P16 China Peanut 3 I/18 FJ561114.1 Pe1 China Pepper 3 I/14 FJ561154.1 Pe5 China Pepper 3 I/34 FJ561091.1 Po2 China Potato 2 II/1 FJ561158.1 Po14 China Potato 2 I/13 FJ561162.1 Po82 Mexico Potato 1 II/4 FJ561070.1 Po152 Mexico Potato 3 I/18 FJ561148.1 Po276 Australia Potato 2 II/1 FJ561082.1 PSI7 Indonesia Tomato RUN83 2 IV/8 or 10 EF371804.1 PSS81 Taiwan Tomato 3 I/14 FJ561066.1 PSS219 Taiwan Tomato 3 I/34 FJ561167.1 PSS358 Taiwan Tomato 3 I/15 FJ561065.1 R230 Indonesia Banana BDB IV/10 AF295280.1 R288 China Mulberry HT659 5 I/18 GQ907153.1 R292 China Mulberry RUN91 I/12 DQ657635.1 R058 ND ND ND ND DQ011543.1 Tb3 China Tobacco 3 I/17 FJ561128.1 Tm2 China Tomato 3 I/14 FJ561134.1 Tm3 China Tomato 3 I/18 FJ561135.1 Tm11 China Tomato 3 I/13 FJ561150.1 Tm13 China Tomato 3 I/17 FJ561133.1 Tm82 China Tomato 4 I/16 FJ561094.1 UW9 Costa Rica Heliconia JT644 II/3 AF295257.1 UW21 Honduras Banana R371, CIP21 1 II/6 DQ011546.1 UW129 Peru Plantain 1 II/4 EF371811.1 UW160 Peru Plantain R282 1 II/4 GU295051.1 UW162 Peru Musa sp. JT648 II/4 AF295256.1 UW163 Peru Plantain 1 II/4 GU295052.1
64 Table 3.2 Continued UW167 Costa Rica Banana R283, CIP125 1 II/3 DQ011545.1 UW175 Colombia Plantain 1 II/4 DQ011547.1 UW181 Venezuela Plantain JT649, K261 1 II/6 GU295053.1 UW477 Peru Potato RUN110 II/ND DQ657604.1 UW551 Kenya Geranium 3 II/1 DQ657596.1 Z7 China Ginger 4 II/16 FJ561142.1 Zo4 Philippines Ginger 4 I/14 FJ561156.1 a http://www.ncbi.nlm.nih.gov/genbank/ b The p hylotype, sequevar, biovar, host, and origin information for the were obtained from previously published literature ( Poussier et al., 2000; Fegan and Prior, 2005; Fegan and Prior, 2006; Castillo and Greenberg, 2007; Wicker et al., 2007; Cardozo et al., 2009; Xu et al., 2009;, or if unknown= ND
65 Figure 3.1 A 50% majority rule consensus rooted tree cr eated by Bayesian analysis of the egl Weighted line indicates posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. Strains from Carolina are red, Georgia green, Florida blue, Caribbean p urple, and unknown location black.
66 F igure 3.2 A 50% majority rule consensus un rooted tree cr eated by Bayesian analysis of the egl Weighted line indicates posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony a nd likelihood.
67 Figure 3.3 The zone of inhibition, indicated by the red circle, produced by Ralstonia solanacearum strains RS 37 and RS 5. The lawn strain was sprayed 24hr aft er the lawn strain was sprayed.
68 Figure 3.4 Gel displaying bands created by using primers for multiplex Musa, lanes 2 4, 759/760 Ralstonia specific primers, lanes 5 7, and one primer set from the Musa multiplex PCR Musa06, lane 8. Lane 1 contains Ladder 10 100 bp markers. Strain RS37 was used in lanes 2, 5 and 8, RS5 in lanes 3 and 6, and 527 in lanes 4 and 7.
69 Figure 3.5 Comparison of d ried M usa Dwarf Nam wa root mass 30 days post inoculation of water control (WC), RS37 and Race 2 strains ; UW2, UW70, and UW170. Same letter over each bar indicates no significant difference according to Figure 3.6 Comparison of M us a Dwarf Nam wa root mass 30 days post inoculating plants with RS 5 ( Race 1 ) UW2 ( Race 2 ) and RS 37 Same letter over each bar indicates no significant difference according to 0 2 4 6 8 10 12 WC UW2 RS37 UW70 UW170 g a b b b b 0 1 2 3 4 5 6 7 8 9 10 UW2 RS37 RS5 WC g a b a b
70 CHAPTER 4 MULTIPLE LOCI VARIABLE NUMBER TAND EM REPEAT ANALYSIS A MONG RALSTONIA SOLANACEAR UM STRAINS FROM THE SOUTHEASTERN UNITED STATES Introduction Bacterial taxonomy has been an arduous task. Aspects of the current taxonomy for all life can be traced to Aristotle. Carl Linnaeu s in 1753 published Systema Naturae in which he described the binomial system of nomenclature used today. Over a century later Ferdinand Cohn, with the aid of the compound microscope, published the first bacterial classification system (Palleroni, 2003). Bacteria were classified based on phenotypic characteristics, which included morphology and their biochemical and DNA hybridization first was used to estimate relatedness a mong animals (Hoyer et al., 1964) and then among bacteria (Johnson and Ordal, 1968). This permitted bacterial classification to be based loosely on genotype, which provided some clarity over the relative confusion inherent in phenotypic classification. Soo n thereafter, bacterial diversity was studied using conserved regions of the bacterial genome such as rRNA (Pace and Campbell, 1971), which led to the pervasive sequencing of the 16s rRNA to infer phylogenetic relationships (Woese, 1987). The use of 16s rR NA brought additional order to bacterial classification. Bacterial nomenclature, like other biological systems, consists of a Latin binomial (genus and species) Bacterial s pecies are defined by members with 70% or greater si milarit y DNA DNA hybridization and sharing phenotypic characteristics (Konstantinidis et al., 2006; Staley, 2006) Phenotypic characteristics presently used in defining a bacterial species include: biochemical reactions, chemical composition, cellular structure s, and immunological featu res (Tortora et al., 1995)
71 Identifying a species and determining its limits presents the most challenging aspect of biological classification for any type of organism Currently, one of the main approaches to show intraspecies relatedness among eukaryotes, bacteria, or archaea is genetic typing (Bickford et al. 2007, Jacobs et al. 2004, Konstantinos et al, 2006, Li et al, 2009). Complete genome sequencing might be ideal for intraspecies comparisons, but the process can be costly and time consuming, especially if surveying many individuals. Researchers searched for a few molecular markers that would be cost effective and discriminative, but simple and reproducible. Multil ocus sequence typing (MLST) consists of sampling a panel of multiple housekeeping genes and analyzing them phylogenetically. Multilocus phylogenetic analyses have been shown to increase phylogenetic accuracy (Maiden, et al., 1998) (Hommais et al., 2005). Although useful in determining possible ancient relationships questions in bacteria, MLSTs are often unable to differentiate closely related bacterial strains (Li et al. 2 009). Simple sequence repeats (SSRs), also called microsatellites, have been used in multi cellular organisms to differentiate among populations of the same species. These repeating units of DNA have been used as molecular markers for various applications, including analysis of kinship and population structure. Previously, it was shown that the use of SSRs in unicellular organisms would be inconsistent due to the high mutation rate and horizontal gene transfer in E scherichia coli (Metzgar et al., 2001). Met zgar et al. (2001) showed inconsistency when comparing phylogenetic trees constructed with 25 long microsatellites to trees created with multilocus enzyme electrophoresis.
72 However, Diamant et al. (2004) showed that combining four SSRs for phylogenetic anal ysis created a tree that was in agreement with those constructed with the standard multilocus enzyme electrophoresis. They also noted that the flanking regions of the microsatellite were just as informative as the microsatellite itself. Over the years mult i locus SSR analysis has been increasingly recognized as the method of choice for genotyping a number of human pathogens including, Bacillus anthracis (Keim et al., 1999), Borrelia species (Farlow et al., 2002), Salmonella enterica (Lindstedt, 2005), Enter ococus faecium (Titze deAlmeida et al., 2004), E. faecalis (Top et al., 2004), E. coli (Keys et al., 2005 and van Belkum et al., 1998), Staphylococcus aureus (Francois et al., 2005), Vibrio cholera (Danin Poleg et al., 2007) and V. vulnificus (Broza et al ., 2007). Smith et al. (2007) showed the clades in the phylogeny of Corynespora cassiicola created by the analysis of SSRs correlated with the geographic region from which the pathogen was isolated. The taxonomic history of the bacterium known to cause bac terial wilt followed the same pattern of classification as previously described. It was first assigned to a taxonomic group based on phenotype and then later based on genotype. At first, phenotypic characters were used to place the bacterium in the genus Bacillus because it was rod shaped. Further investigation revealed that the bacterium contained polar flagella indicating that the bacterium was motile; therefore it was placed in the genus Pseudomonas It was given the specific epithet P. solanacearum to highlight its association with solanaceous plants (Kelman, 1953). The bacterium was classified further to subspecies based on the differential ability of strains to produce acid from several disaccharides and sugar alcohols (Hayward, 1964), as well as by host range
73 ( Buddenhagen et al., 1962 ). standard, pseudomonads were grouped as either fluorescent or non fluorescent based on genotypic characteristics (Palleroni, 2003). The pseudomonads were of ficially separated into different genera with the genus Burkholderia created to contain B. solanacearum among other species (Yabuuchi et al., 1992). Based on phenotypic characteristics, phylogenetic analysis of 16s rDNA and rRNA DNA hybridization, the path ogen was transferred to a new genus, Ralstonia where it resides today (Yabuuchi et al., 1995). However, DNA DNA hybridization revealed that R. solanacearum strains often have less than 70% homology, which is the threshold level expected within a bacterial species (Palleroni and Doudoroff, 1971, Fegan and Prior, 2005). R estriction fragment length polymorphism (RFLP) analysis of 62 strains determined that 28 unique groups exist within R. solanacearum and a similarity coefficient matrix showed that these groups formed two distinct major groups (Cook, 1989). Other DNA profiling methods used to group strains to the subspecies level included pulse field gel electrophoresis (Jaunet and Wang, 1999), amplified fragment length polymorphsim (AFLP) (Poussier et al. 2000), random amplification of polymorphic DNA (RAPD) (Jaunet and Wang, 1999), and PCR amplification of repetitive genetic elements, such as REP, ERIC, and BOX PCR (Thwaites et al., 2002, Norman et al., 2009). Phylogenetic analysis of several genes has r evealed that the bacterium has diverged into four different groups with distinct geographic origins. These four groups have been labeled phylotypes 1 through 4 (Table 1.1). The phylotype groups were confirmed using the following genes: 16s rRNA (Poussier e t al., 2000), internal transcribed spacer region (ITS) (Fegan et al., 1989), egl (Fegan, 1989), hrpB (Poussier
74 et al., 2000), mutS (Guidot et al., 2007, Wicker et al., 2007), and cytochrome b561 (Norman et al., 2009). Phylogenetic analysis of egl has been used to determine infrasubspecific groups, referred to as sequevars, which are similar in pathogenicity or originate from a common geographic region (Fegan and Prior, 2005). However, sequevar clustering differs depending on the gene used when tested with w orldwide representatives (Villa et al., 2005). A panel of five housekeeping genes and three virulence related genes were used in an MLST analysis of a worldwide sample of 58 R. solanacearum strains, which confirmed the existence of four phylotypes and fur ther subdivision in phylotype II (Castillo and Greenberg, 2006). Genomic comparison of six different strains representing the four phylotypes and the two divisions within phylotype II, confirmed the diversity that exists amid the Ralstonia complex (Remena nt et al., 2010). R solanacearum could be reclassified into at least three distinct taxonomic groups based on genomic differences that exist between phylotypes. Grouping based on average nucleotide identity values (ANI) revealed that phylotypes I and III should be considered one species and phylotypes II and IV should each be a distinct species. Topology of the phylogenetic tree computed from ANI values was consistent with trees previously computed with comparative genomic hybridization (CGH) microarray da ta and the mutS and hrpB sequences. To develop markers that could discriminate among closely related strains, microsatellites were identified in the R. solanacearum strain GMI1000, and PCR primers were developed that encompassed not only the SSR, but also the potentially variable flanking regions on either side of the SSR The DNA sequences were phylogenetically analyzed using parsimony, maximum likelihood and Bayesian analysis. These trees
75 were compared to each other using SSRs alone, SSRs with the flankin g regions also called microsatellite associated loci (MAL) and combining microsatellite associated loci (MALs.) These trees were then compared to trees created by egl and to previously created trees in other studies The goal of this work was to determine whether the trees created by SSR s we re similar to the egl trees, and whether these markers would be informative when comparing strains isolated in the southeastern U.S Materials and Methods Bacterial Cultures Strains were collected from researchers acro ss the southeastern U.S. (Table 1). Dave Norman's collection (which was given ID numbers in the 200s) included strains strains originated from north Florida. Strains from R 400s) were isolated in Georgia, and the 500 series of strains included those isolated from the Carolinas, which were from the collections of Dan Kluepfel and Asimina Mila. Strain AW1, isolated in Alabama, came from Mark Allen provided strains UW2, UW70, and UW170, and Anne Alvarez provided strain A3908. To confirm that the strains were indeed R solanacearum, they were streaked on semi selective SMSA (Englebrecht, 1994), then PCR ampli fied using the 759/760 Ralstonia specific primers (Opina et al., 1997). Genomic DNA Extraction and Multiplex PCR Genomic DNA was extracted and collected following methods outlined by Sambrook et al (1989) To confirm the presence of DNA, each DNA extract was tested by PCR using the 759/760 primers. The phylotype of all strains was identified using multiplex PCR as outlined by Fegan & Prior (2005). The PCR products were visualized
76 on a gel, and phylotype w as assigned according to band length. In each reaction RS5 phylotype II served as a positive control. Determining SSRs I used the two web driven software packages, http://insilico.ehu.es/microsatellite s (Bikandi et al. 2004 ) and http://minisatellites.u psud.fr/ASPSamp/base_ms/bact.php ( Liolios et al. 2008) to find novel SSRs in the GMI1000 genome. A homology search was performed using NCBI BLAST to determine if the SSRs were present in four additional genomes (or partial genomes) of R. solanacearum strains (i.e., UW551 race 3 biovar 2 IPO1609 race 3 biovar 2 and MolK race 2 biovar 3) Because previous studies showed that SSR fl anking regions were phylogenetically informative (Diamant et al, 2004), I located not only the microsatellites themselves, but also flanking regions outside the SSR that totaled 8 00 1,2 00 base pair s. These amplified DNA sequences were aligned manually usi ng Se Al (Rambaut, 2002). Primers were designed based on the conserv ed regions of the available R. solanacearum strains PCR amplification using these primers was first tested using genomic DNA from GMI1000. The SSRs were then tested on a panel of R solanacearum strains; GMI1000 race 1 biovar 3, RS5 race1 biovar1, and A3908 race2. PCR, Sequencing and Phylogenetic Analysis Three SSRs (including their flanking regions) and the egl gene were amplified using PCR for each strain collected. The PCR reactio n s consisted of GoTaq Flexi DNA Polymerase (Promega Corp., Madison, WI) with 5.0l of 5x buffer, 1.5l of MgCl 2 4.0l dNTP, each primer at 10pmol, 1 unit of Taq DNA polymerase, and 100ng of DNA. S terile deionized water was used to adjust t he total volume to 25l DNA cycle sequence amplification was conducted in a thermocycler (Bio Rad Laboratories,
77 Hercules, CA) using the following program: hot start at 95C for 5 min followed by 30 cycles of 95C for 45 sec, 68C for 30 sec and 72C for 60 sec, and fini shed with a 10 min extension at 72C. The annealing temperature was adjusted according to the com position of the oligonucleotide sequence s For egl amplification the primer pair JHFegl: GACGATGCATGCCGCTGGTCGC and JHRegl: CACGAACACCACGTTGCTCGCATTGG were des igned based on the egl sequences of GMI1000 (AL646053.1), UW551 (DQ657596.1), and Molk2 (CU694393.1) accessed from NCBI BLAST Sequevar was determined by phylogenetically analyzing the egl with the strains from this study with strains previously described (Table 3.2) The DNA sequences of the egl gene were used to select a panel of representative strains that underwent PCR and sequencing for the three SSRs and their flanking regions (Table 4.1). Closely related R. syzygii strains, blood disease bacterium R2 23 (DQ011538.1) and R. syzygii strain R058 (DQ011543.1) served as the outgroups for the egl tree The SSRs and flanking regions were amplified using the primers in Table 4.2. R. pickettii was used as the outgroup for the analysis performed with the SSRs. G enBank accession numbers are listed in Table 4.3. PCR products were submitted to Interdisciplinary Center for Biotechnology Research (ICBR) sequencing facility located sequencing Amplicons were sequenced in both the forward and reverse directions to create a consensus sequence that was used for phylogenetic analysis. Phylogenetic analysis was performed for the egl single SSRs, individual MALs, and the combination of 3 MALs. Seq uences were aligned using Muscle (Edgar, 2004), via http://www.ebi.ac.uk/Tools/muscle/index.html Se Al version 2.0a11 was used for
78 manual sequence manipulation; the program was available at http://tree.bio.ed.ac.uk/software/seal/ (Rambaut, 1996). Phylogenetic analysis was performed using maximum parsimony, maximum likelihood and Bayesian methods. Maximum parsimony was performed using PAUP* version 4.0b10 (Sworfford, 2003) and performed on the Fisher Cluster at The University of Florida Genetics Institute Gainesville, FL The parsimony dendrograms were created by performing tree bisection/reconnection (TBR) branch swapping and heuristic sear ches with random stepwise addition. Branch support for the parsimony trees was estimated by nonparametric bootstrapping (n=1,000 replicates) using TBR swapping (n= 1,000 replicates ; Felsenstein, 1985). RAxML VI HPC was used to run maximum likelihood and max imum likelihood bootstrap analysis (Stamatakis et al. 2005) RAxML VI HPC was also performed on the Fisher Cluster For the bootstrap analysis, 5 000 runs were used to generate 5 000 trees. Treefig was used to create the consensus tree s ( Rambaut et al. 2 006). MrBayes 3.1 was used to perform Bayesian analysis (Huelsenbeck et al. 2001) on the Fisher cluster The likelihood settings in Bayesian analysis were adjusted for each data set to the model that was suggested by (AIC) Modeltest 3.7 provided the AIC model selection strategy, and the best fit model for each data set is found in Table 4.4 (Posada and Crandall 1998). The model selected for egl was described in Chapter 3. The settings were adjusted for 1,000,000 generation s being sampled at every 500 generations. The burn in values for each analysis are listed in Table 4.4 and the consensus tree of the post erior probabilities was created by PAUP* 4.0b10.
79 Results Determining the SSRs R solanacearum SSRs were determined by using the completely sequenced genome of GMI1000 and the algorithms found at http://insilico.ehu.es/microsatellites (Bikandi et al. 2004 ) and http://minisatellites.u psud.fr/ASPSamp/base_ms/bact.php ( Liolios et al. 2008) The first website identified 13 SSRs and the second 63. The results from each website were compared to e ach other and based on length and 0% mismatch, 11 SSRs were selected for further development. The websites suggest primers for the 11 microsatellites, which were tested on GMI1000. Of the 11 primer sets only six produced PCR products. The SSRs that corresp onded to the six primer pairs were compared to complete or partially sequenced genomes. More specific primers were designed based on homologous regions when comparing the genomes. Only four of the six SSRs consistently produced the correct PCR product when amplified from GMI1000, RS5 and A3908. When sequenced, one of the SSRs produced sequences that differed dramatically in the forward and reverse sequencing direction. This could represent the amplification of two radically different alleles or non specific binding during PCR. Consequently, this SSR was discarded from further development. Three SSRs and their flanking regions, designated as SSR 1, SSR 3, and SSR 9,were used for phylogenetic analysis. Comparison of SSRs All of the microsatellites consisted of repeats that were 6 nucleotides or longer in length and were located on the chromosome ( Figures 4.1 4.3). SSR 1 of GMI1000 consists of TTGCGA repeated 6 times and is part of a probable dioxygenase related to 2 nitropropane dioxygenase and probable hema gglutinin related autotransporter
80 protein. The SSR 3 of GMI1000 consisted of GATCACCAG repeated three times and was part of a putative sugar transporter transmembrane protein. SSR 9 of GMI1000 was CAGCTGGAG repeated 3.5 times and was part of a probable for mate dehydrogenase iron sulfur subunit. SSR 1 of GMI1000 were not homologous to the SSR 1 amplified from UW551, IPO1609, and Molk2, and majority of the microsatillite is missing from these strains ( Figure 4.1) SSR9 of GMI100 appears to be homologous to the SSR 9 amplified from UW551, IPO1609, and Molk2 ( Figure 4.3). SSR 3 from GMI1000 was 100% identical to UW551 and IPO1609 ( Figure 4.2). All of the SSRs of UW551 and IPO1609 were 100% identical to each other. Molk2 was identical to UW551 and IPO1609 for SSR 1 and 9, but appears to be missing the first 9 nucleotides in SSR 3. Comparing the flanking regions of the microsatellites, SSR 1 and 3 from the 4 strains appear to be homologous, while the flanking regions of SSR 9 are almost 100% identical. Using the pri mers that were designed from the homologous flanking regions, the SSRs of strains listed in Table 4.1 were amplified and sequenced. The flanking regions of each microsatellite were removed before analysis of the SSRs alone. Not all of the strains produced bands strong enough to be amplified and sequenced. SSR 9 contained the greatest sequence diversity followed by SSR 1 ( Figures A.1 A.2). The SSR 3 was identical for all the strains except for Molk2 (not shown). Phylogenetic Analysis of the SSRs Maximum par simony, maximum likelihood, and Bayesian methods were used to phylogenetically analyze the SSRs with and without their flanking regions ( Figures 4.4 4.7, A.3 and A.4). The most informative microsatellites were SSR 1, which grouped most of the phylotype I s trains separate from the phylotype II ( Figures 4.4 4.5). SSR 9 was able to group most of the phylotype I strains together except for 227, 346, 364, and
81 367. However, SSR 3 was uninformative ( Figures A.3 and A.4). Strain 232, isolated from tomato in Guadelo upe was classified as phylotype II by multiplex PCR; however, it grouped with phylotype I strains in both SSR 1 and SSR 9 trees (with or wit hout their flanking regions; Figure s. 4.4 4.8 and 4.12). Included in this clade were other strains, which were isol ated from tomato in Guadeloupe that were classified as phylotype I based on multiplex PCR and clustering on the egl trees. Phylogenetic analysis of the egl also placed strain 232 in a group with other phylotype I strains. Multiplex PCR was performed twice and each time this strain displayed bands that were consistent with those of phylotype II (not shown). Analysis of MALs The flanking regions of microsatellites from bacteria were reported to be as informative as the SSR by itself when phylogenetically an alyzed (Diamant et al., 2004). Thus, the primers designed for sequencing the SSRs included up to 420bp on either side of the SSR. The MALs were analyzed in the same manner as the SSRs alone. Strains whose genomes were recently sequenced (i.e. CFBP2957, PSI 07, and CMR15) were included with these analyses (Fig ures 4.8 4.13). The MALs were considerably more informative than the SSRs alone, especially for SSR 3. For all three MALs, the phylotype I strains grouped together, and within phylotype II, the two sequ evar 1 strains formed a clade. CMR15, phylotype III, was more closely related to phylotype I, than phylotype II or IV were to phylotype I for trees created by MALs 1 and 3 (Fig ures 4.9 and 4.11). Remenant et al. (2010) suggested that the close relationship of phylotype I and III based on DNA homology is sufficient to classifify them in a single, distinct genus; and that phylotype II and IV should each be placed in a distinct genus (2010).
82 Trees created using MAL 9 were less informative than those created w ith SSR 1 and 3 ( Figures 4.12 4.13 and A.5 A.6). MAL 9 grouped the sequevar 1 strains in one clade and some of the phylotype II strains formed another clade. The lack of support for the tree could be due to the diversity of the microsatellite that makes up MAL 9. The product size of MAL 9 is smaller than the other MALs due to the lack of homology in the flanking regions. Comparing the trees created from the different methods, strains CFBP2957, phylotype IIA with CMR15, phylotype III, and PSI07, phylotype IV grouped with some of the Carolina strains ( Figures A.5 A.6). The maximum parsimony tree placed the UW551, whose sequence came from GenBank, with the other sequevar 1 strains. However the UW551 MAL 9 sequenced in the lab did not group with any group. Both of the UW551 sequences were identical ( Figure A.6). Combination of the Three MALs The three SSRs and their flanking regions were combined and phylogenetically analyzed using the same methods as before. Combining all 3 MALs appears to give the trees great er resolution because the posterior probability and bootstrap values were greater than those from trees generated by single SSRs with or without their flanking regions ( Figures 4.14 4.15). The phylotype I and sequevar 1 strains grouped together as they did previously with the egl SSRs, and MALs. The phylotype III strain was more closely related to the phylotype I strains and phylotype IV was more closely related to phylotype II, as mentioned above. Phylotype II grouped into 4 distinct groups designated A D ( Figure 4.15). Group A contained strains isolated from the southeastern U.S. Group B contained the sequevar 1 and race 2 strains. Group C contained strains that originated from the Caribbean, including RS37, which was described in the
83 previous chapter. La stly, group D contained strains isolated in Florida from pothos and anthurium cuttings that originated from Costa Rica (Norman et al., 2009). The trees created using the 3 MALs were very similar to trees generated with the egl gene. The 4 groups (A D) not ed in the combined MALs tree were also found in the egl trees. Strains belonging to group A strains are the same as sequevar 7, group B strains are similar to sequevar 1, group C strains belong to sequevar 5/6 and an undefined sequevar, and group D were found within sequevar 4. Discus sion We have demonstrated that sequenced loci containing SSRs can be used to determine infrasubspe cific groups. This is the first time that SSRs have been used to distinguish subspecies groups in plant pathogenic bacteria. The flanking regions of the SSRs were more informative than the SSRs by themselves. This study corroborates the importance of comb ining multiple SSRs for phylogenetic analysis (Diamant et al., 2004). In this study, the MALs phylogenetic trees showed greater resolution and higher confidence values than trees generated with the microsatellites alone. To reach consensus among phylogenet ic hypotheses it is useful to employ multiple types of phylogenetic analysis. Here we used maximum parsimony, maximum likelihood, and Bayesian analysis. Similar to the trees created by egl and the other genes used for phylogenetic analysis, SSRs were able to group the strains based on geographic region. However, like the previously published trees, trees created by SSRs were unable to group strains based on host. All strains that originated from southeastern U.S. were found in group A or sequevar 7. Two genotypes of R. solanacearum were identified in the southeastern U.S. that differed by a truncated non functional avirulence gene and the wild type gene
84 (Robertson et al., 2004). The Carolina strains, which had the truncated gene, have a broader host range and can cause wilt in tobacco, whereas the Georgia and Florida strains cause a hypersensitive reaction (HR) in tobacco. Tobacco is grown more frequently in the Carolinas than Florida, thus these strains might be under different selection pressures. A geo graphic division among these strains was not detected using either the egl or the SSR trees. Even though the SSRs used in this study confirmed the clades created by egl other microsatellites could be more informative for resolving geographic variation tha n the SSRs used in this study. These microsatellites were discovered using GMI1000 as a reference; hence, it is possible that microsatellites identified in another strain could exhibit more diversity within a sequevar. Microsatellites are believed to give organisms the ability to adapt rapidly to new environments (Li et al. 2004). Highly variable SSRs are correlated with the high frequency appearance of new alleles within species and populations (Weber and Wong, 1993). It is therefore possible that the SSRs would differ among populations; the Carolina strains, for example, experiencing selection pressures that differ from the Florida strains. Using SSRs we demonstrated which strains were native to the U.S. and identified exotic strains that were isolated in the U.S. According to this study and the previous chapter, Florida appears to have multiple populations of exotic strains. Based on phylogenetic analysis some of these exotic strains appear to have originated in the Caribbean or Latin America, whereas oth ers, those of phylotype I, likely came from Asia (Hong et al., 2008, Norman et al., 2009, Ji et al., 2007). Two predominant groups of exotic strains were isolated in Florida, group C/sequevar 5/6 and an undefined sequevar and group D/sequevar 4. Strains be longing to group C were isolated in the waterways,
85 greenhouses, and fields in northern Florida (Hong et al., 2008). Group D strains were isolated from pothos and anthurium cuttings originating from Costa Rica (Norman et al., 2009). Microsatellites proved t o be useful in separating populations within the Ralstonia species complex, and could be used to confirm sequevar status as well. The accepted model to determine sequevar is sequencing and analysis the egl Yet, confirmation of sequevar has not been determ ined when comparing trees created by other genes. A few of these genes have shown promise in the preliminary stages, but upon using strains representing different regions of the world the sequevar grouping was inconsistent with the topology of the egl tree s (Villa et al., 2005). Clades created by cytochrome b561 or mutS were similar to those produced by egl however a worldwide 2009, and Wicker et al, 2007). Careful con sideration should be made when the phylogeny of an organism is based on a single gene. DeLong and Pace (2001) stated on 16s rDNA may reflect one gene traces the evolution of the gene and not the organism. Comparison of the complete genome would be the most accurate way to distinguish amo ng closely related bacteria. However, due to time and costs, this method currently is not feasible for a large survey of related organisms. The analysis of microsatellites along with their flanking regions provides a survey of different areas of the chromo some (Table 4.2.), while being less expensive than complete genome sequencing.
86 The use of MALs could provide support for dividing R. solanacearum into three distinct species. If MALs are effective in determining phylotype and sequevar with strains origina ting worldwide, then a large survey could provide support for the division of the genus. Multiplex PCR could assign phylotype to multiple strains more quickly than phylogenetic analysis. However, in this study multiplex PCR assigned strain 232 to phylotype II, while it was classified as phylotype I based on phylogenetic analysis of the egl and the SSRs. Using phylogenetic analysis strain 232 clustered other strains isolated from the same host from the same geographic region The methods used in this study fo r phylogenetic analysis are the current recommended methods. Many previously published papers regarding the phylogeny of R. solanacearum used weaker methods of phylogenetic inference such as Neighbor Joining and UPGMA. Therefore, their results could be dr awn into question. These weaker methods are often used because they produce a single tree and give the researchers quick trees for publication. These methods were shown to be less effective and sometimes positively misleading under certain realistic condit ions (Felsenstein, 1978; Kuhner, and Felsenstein, 1994; Hillis, et al., 1994).. This could explain the inconsistencies found when comparing trees previously published for R. solanacearum (Taghavi et al., 1996, Poussier et al., 2000, Villa et al., 2005, and Li et al., 2009). Discrepancies between current and past phylogenetic reconstructions (e.g., their ability to tease apart fine scale differences among populations) could be due to either insufficient data or incomplete taxon sampling. Unknown diversity or lack of diversity could be seen as a result of unintentional ly bias ed sampling One way to overcome this error is to include multiple taxa and to compare analyses from different parts of the
87 genome. Phylogenetic analysis using maximum parsimony, maximum likelihood and Bayesian methods are currently used to reconstruct consensus phylogenetic trees that are reproducible. In this study we found small inconsistencies with a tree created using the parsimony method when creating the trees for MAL 9 (A.6). However being able to compare the topology with the other 2 methods and relying on the confidence values, the placement of this strain was determined to be incorrect based on maximum parsimony ( Figu re 4.9 ). This study showed that MALs could be used for phylogenetic analysis to demonstrate diversity within a bacterial species. Additional studies should be done to determine whether methods used in this study are applicable to other phytobacteria. Furth er studies are needed to determine if these markers can be used to show diversity in R. solanacearum strains originating from different parts of the world. This would indicate that MALs can determine the phylotype divisions and confirm sequevar groupings a s created by e g l Plants resistant to bacterial wilt are geographically constrained. Hopefully with the use of MALs, local populations can be identified and aid in our understanding of how the bacterium is able to overcome resistance. Further research is a lso needed to determine the frequency with which these microsatellites mutate, and how consistent those mutations are over time. In studies investigating the diversity of eukaryotes, both molecular and morphological data have been combined for phylogenetic analysis (Bybee et al., 2008; Goloboff et al., 2009; Ogden and Whiting, 2003). It could be possible to combine phenotypic characteristics along with molecular data to show a greater diversity in the infrasubspecific groups. Classical methods for grouping subdivisions of R.
88 solanacearum such as HR on tobacco, pathogenicity tests, and utilization of carbohydrates could be used as characters when comparing strains. Simultaneous analysis of both molecular and phenotypic data could possibly reveal where or whe n divergence took place for host specialization. Currently, bacterial species are defined as combine these characteristics in a phylogenetic analysis to determine clear divisions between closely related bacterial populations.
89 Table 4.1 Strains sequenced and used in this study Strains a Origin Host Origin Other Name Year Biovar b Phylotype c SSR sequenced d 203 FL Tomato P504 NA e 1 2 1 204 FL Tomato P505 NA e 1 2 1 205 FL Tomato P507 1996 1 2 1 206 FL Tomato P530 1997 1 2 1 208 FL Tomato P533 1997 1 2 1, 9 209 FL Tomato P534 1997 1 2 1 210 FL Tomato P535 1997 1 2 1 211 FL Tomato P536 1997 1 2 1 212 FL Potato P541 1997 1 2 1, 9 213 FL Potato P543 1997 1 2 1, 9 214 FL Potato P550 1997 1 2 1, 9 215 FL Potato P553 1997 1 2 1, 9 216 FL NA e P557 1997 1 2 1, 9 217 FL Pothos P564 1996 1 2 1, 9 218 FL Pothos P573 1999 1 2 1, 9 219 FL Tomato P576 1999 1 2 1, 9 220 FL Tomato P594 2000 1 2 1, 9 221 FL NA e 136 NA e 1 2 3, 9 222 FL Heliconia 158 NA e 1 2 1, 3, 9 223 FL Pothos 485 NA e 1 2 9 224 NA e Pothos NA NA e 1 2 1, 3, 9 225 NA e Pothos 517 NA e 1 2 1, 3, 9 226 NA e Pothos 521 NA e 1 2 1, 3, 9 227 NA e NA 528 NA e 1 1 1, 3, 9 228 FL Potato 544 1994 1 2 1, 3, 9 229 NA e Pothos 545 NA e 1 2 1, 3, 9 230 NA e Pothos 548 NA e 1 2 1, 3, 9 231 NA e Tomato NA e NA e 1 2 1, 3
90 Table 4.1 Continued 232 Guadeloupe Tomato 506 1985 1 2 1, 9 234 Martinique Tomato 610 1987 1 2 1, 3, 9 235 NA e Anthurium 618 NA e 1 2 1, 3, 9 236 NA e Anthurium 621 NA e 1 2 1, 3, 9 237 USA Tomato 660 NA e 1 2 3, 9 238 NA Pothos 673 NA e 1 2 1, 3, 9 239 NC NA e 688 NA e 1 2 1, 3, 9 240 NC Potato 689 NA e 1 2 1, 3, 9 241 NC Tomato 690 NA e 1 2 1, 3, 9 242 NC Tomato 691 NA e 1 2 1 301 FL Soil RS14 2001 1 2 1 302 FL Tomato RS18 1991 1 2 1 304 FL Geranium RS38 2001 1 2 1 305 FL Geranium RS39 2001 1 2 1 306 FL Geranium RS40 2001 1 2 1 307 FL Pond Water RS51 2001 1 2 1 329 FL Hydrangea RS118 2005 1 2 9 330 FL Pepper RS121 2005 3 1 3 331 FL Pepper RS122 2005 3 1 9 346 Peru Tomato TD406 NA e 3 1 9 364 Costa Rica Pepper UW255 1972 3 1 9 366 China Peanut UW368 NA e 1 1 9 367 China Pepper UW371 NA e 3 1 9 401 GA Tomato Rso81 5 1981 1 2 1, 3 402 GA Tobacco Rso96 41 1996 1 2 1 403 GA Tomato Rso81 2 1981 1 2 1 404 GA Tomato Rso86 2 1986 1 2 1 405 GA Coffee Weed Rso80 1 1980 1 2 1 406 GA Tomato Rso87 105 1987 1 2 1 407 GA Potato Rso84 1 1984 1 2 1
91 Table 4.1 Continued 501 NC Tobacco K60 NA e 1 2 1, 3, 9 502 NC Tobacco NC116 NA e 1 2 1, 3, 9 503 GA Tobacco GA122 NA e 1 2 1, 3, 9 504 GA Tobacco GA142 NA e 1 2 1, 3, 9 505 SC Tobacco SC06 NA e 1 2 1, 3, 9 506 SC Tobacco SC121 NA e 1 2 1, 3, 9 507 SC Tobacco SC108 NA e 1 2 1, 3, 9 508 NC Tobacco L 17 NA e 1 2 1, 3, 9 509 NC Tobacco L 18 NA e 1 2 1, 9 510 NC Tobacco L 19 NA e 1 2 1, 9 511 NC Tobacco J 12a NA e 1 2 9 512 NC Tobacco T 24 NA e 1 2 1, 9 513 NC Tobacco T 30 NA e 1 2 1, 3, 9 514 NC Tobacco T 32 NA e 1 2 1, 3, 9 515 NC Tobacco T 33 NA e 1 2 1, 3, 9 516 NC Tobacco V 167a NA e 1 2 1, 3, 9 517 NC Tobacco V 167b NA e 1 2 1, 3, 9 518 NC Tobacco V 168 NA e 1 2 1, 3, 9 519 NC Tobacco V 170 NA e 1 2 1, 3, 9 520 NC Tobacco 40a NA e 1 2 1, 3, 9 521 NC Tobacco K66 NA e 1 2 1, 3, 9 522 NC Tobacco K68 NA e 1 2 1, 3, 9 523 NC Tobacco K69 NA e 1 2 1, 3, 9 524 NC Tobacco K71 NA e 1 2 1, 3, 9 525 NC Tomato K74 NA e 1 2 1, 3, 9 526 NC Tomato K136 NA e 1 2 1, 3, 9 527 HI Helicona A3908 NA e 1 2 1, 3 RS5 FL Tomato NA e 1999 1 2 1, 3, 9 RS37 FL Geranium NA e 2001 1 2 1, 3, 9 GPERS001 Guadeloupe Tomato NA e 2009 1 I 1, 9 GPERS002 Guadeloupe Tomato NA e 2009 1 I 1, 9 GPERS004 Guadeloupe Tomato NA e 2009 1 I 1, 9
92 Table 4.1 Continued AW1 AL Tomato NA e NA e 1 II 1, 9 UW2 NA e NA e NA e NA e 1 II 1 UW70 Colombia Plantain NA e NA e 1 II 1 UW170 Colombia Heliconia NA e NA e 1 II 1 UW551 Kenya Geranium NA e 2003 3 II 1, 3, 9 GMI1000 French Guyana Tomato JS771, RUN54 NA e 3 I 1, 3, 9 a RS5 and RS37, 400 from Ron 500 from Dan Kluepfel and Asimina Mila except for 527 which and GPERS were isolated by Patrice Champoiseau. b Biovar was determined as outlined by Hayward, 1991 c Phylotype was determined by multiplex PCR and confirmed by phylogenetic analysis of the egl as out lined by Fegan and Prior 2005 d Three different simple sequence repeats (SSRs) were identified using GMI1000 and http://insilico.ehu.es/microsatellites and http://minisatellites.u psud.fr/ASPSamp/base_ms/bact.php All of the SSRs were sequenced. e NA=information was not available.
93 Table 4.2 List of strains obtained from GenB ank a used for phylogenetic study b Strain Origin Host Other No Biovar Phylotype GenBank acc. no. Starting nucleotide on the chromosome c SSR1 SSR3 SSR9 GMI1000 French Guyana Tomato JS771, RUN54 3 I AL646052.1 132,470 590,335 2,573,692 UW551 Kenya Geranium 3 IIB GQ907150.1 97,625 89066 92,556 IPO1609 Netherlands Potato RUN1 3 II B CU914168.1 197,514 637,913 894,776 CFBP2957 d French West Indies Tomato ND IIA FP885897.1 3,292,177 2,868,703 1,086,444 CMR15 d Cameroon Tomato ND III FP885895.1 3,464,998 3,006,982 1,061,305 PSI07 d Indonesia Tomato ND IV FP885906.2 3,397,158 2,936,982 1,105,420 Molk2 Philippines Musa sp. 1 IIB CU861906.1 165,121 602,150 13,516 Ralstonia pickettii 12J d Michigan Water ND ND CP001068.1 3,809,633 470,824 2,692,895 a http: //www.ncbi.nlm.nih.gov/genbank/ b Phylotype, origin a nd date for the strains other than those characterized in this study are from the literature. ND = not determined c Start and stop were determined when the simple sequence repeat (SSR) was compared and a ligned to the given strain. d These strains were added at the end of the study thus not included in the SSR by itself analysis.
94 Table 4.3 Simple sequence repeats (SSRs) primers designed from the conserved regions o f GMI1000, UW551, Molk2, and IPO1690. SSR Primer Product Size 1 GMMS1F1 GCCAAGGTCCGGCACTGGCTCGACAAGG 832bp GMMS1R3 TCCAGCGTCGATTCCGACACCGACA 3 GMMS3F1 GGTGGCGAATCAGGGACYCG 815bp GMMS3R6 CCTTGAACAGCGGGAAGTAGGTC 9 GMMS9F TCGTATCCATGGCTAGTACGC 210bp GMMS9R GATGTCATTGCCGACATCCT Table 4.4 Parameters adjusted for maximum likelihood settings for Bayesian method for phylogenetic analysis and forming posterior probability trees of simple sequence repeats (SSRs). SSR Model a Burnin b 1 JC 15,000 3 GTR +I+G 20,000 9 TIM 13,000 SSR with flanking regions 1 GTR +I+G 30,000 3 GTR +I+G 23,500 9 TVM+I+G 43,500 1+3+9 GTR +I+G 19,000 a Models were using Modeltest 3.7. b Burnin was calculated by plotting the number of generation versus the Lnl values and determining when the graph becomes stationary.
95 Figure 4.1 Simple sequence repeat 1 as found on Ralstonia solanacearum strain GMI1000. All R. solanacearum strains were obtained from NCBI BLAST. The repeat was highlighted in orange by the author. Figure 4.2 Simple sequence repeat 3 as found on Ralstonia solanacearum strain GMI1000. All R. solanacearum strains were obtained from NCBI BLAST. The repeat was highlighted in orange by the author.
96 Figure 4.3 Simple sequence repeat 9 as found on Ralstonia solanacearum strain GMI1000. All R. solanacearum strains were obtained from NCBI BLAST. The repeat was highlighted in orange by the author.
97 Figure 4 4 A majority rule consensus rooted tree cr eated by Bayesian analysis of SSR 1 Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. Strains from Carolina are red, Georgia green, Florida blue, Caribbean purple and unknown location black. R. solanacearum was isolated from potato (circle), tobacco (triangle), tomato (diamond), pothos/anthurium (square), and diverse location (s tar). Strains were not marked if location and host were unknown. The word GenBank.
98 Figure 4.5 A n unrooted majority rule consensus tree cr eated by Bayesian analysis of SSR 1 Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. The word GenBank.
99 Figure 4.6 A majority rule consensus rooted tree created by Bayesian analysis of SSR 9. Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. Strains from Carolina are red, Georgia green, Florida blue, Carib bean purple, and unknown location black. R. solanacearum was isolated from potato (circle), tobacco (triangle), tomato (diamond), pothos/anthurium (square), and diverse location (star). Strains were not marked if location and host were unknown. from GenBank.
100 F igure 4.7 A n unrooted majority rule consensus tree cr eated by Bayesian analysis of SSR 9 Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. The word GenBank.
101 Figure 4 8 A majority rule consensus rooted tree cr eated by Bayesian analysis of MAL 1 Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. Strains from Carolina are red, Georgia green, Florida blue, Caribbean purple, and unknown location black. R. solanacearum was isolated from p otato (circle), tobacco (triangle), tomato (diamond), pothos/anthurium (square), and diverse location (star). Strains were not marked if location and host were unknown. from GenBank.
102 F igure 4.9 A n unrooted majority rule consensus tree cr eated by Bayesian analysis of MAL 1 Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. The word GenBank.
103 Figure 4 10 A majority rule consensus rooted tree cr eated by Bayesian analysis of MAL 3 Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or g reater for parsimony and likelihood. Strains from Carolina are red, Georgia green, Florida blue, Caribbean purple, and unknown location black. R. solanacearum was isolated from potato (circle), tobacco (triangle), tomato (diamond), pothos/anthurium (square ), and diverse location (star). Strains were not marked if location and host were unknown. from GenBank.
104 F igure 4.11 A n unrooted majority rule consensus tree cr eated by Bayesian anal ysis of MAL 3 Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. The word GenBank.
105 Figure 4 12 A majority rule consensus rooted tree cr eated by Bayesian analysis of MAL 9 Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. Strains from Carolina are red, Georgia g reen, Florida blue, Caribbean purple, and unknown location black. R. solanacearum was isolated from potato (circle), tobacco (triangle), tomato (diamond), pothos/anthurium (square), and diverse location (star). Strains were not marked if location and host were unknown. from GenBank.
106 F igure 4.13 A n unrooted majority rule consensus tree cr eated by Bayesian analysis of MAL 9 Weighted line indicates a posterior probability values at 95 o r greater and bootstrap value of 75% or greater for parsimony and likelihood. The word GenBank.
107 Figure 4 14 A majority rule consensus rooted tree cr eated by Bayesian analysis of all thre e MALs Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimony and likelihood. Strains from Carolina are red, Georgia green, Florida blue, Caribbean purple, and unknown location black. R. solanacearum was isolated from potato (circle), tobacco (triangle), tomato (diamond), pothos/anthurium (square), and diverse location (star). Strains were not marked if location and host were uence was acquired from GenBank.
108 F igure 4.15 A n unrooted majority rule consensus tree cr eated by Bayesian analysis of all three MALs Weighted line indicates a posterior probability values at 95 or greater and bootstrap value of 75% or greater for parsimo ny and likelihood. Groups A indicates the sequence was acquired from GenBank.
109 CHAPTER 5 OVERALL SUMMARY AND DISCUSSION R alstonia solanacearum the causal agent of b acterial wilt can cause serious yield loss for many economically important crops. This study focused on investigating a method for controlling the disease, characterizing strains from the southeastern U.S., and determining if microsatellites could be used to differentiate infrasu bspecies groups. The first objective was to determine the effectiveness of the application of thymol and acibenzolar S methyl (ASM) for decreasing bacterial wilt disease incidence on tomato. The second objective was to distinguish subspecies populations of R. solanacearum strains from the southeastern U.S. by applying classical and current characterization techniques. The third objective was to determine the effectiveness of using microsatellites, microsatellite associated loci (MALs) and the combination of MALs for distinguishing subspecies groups within strains collected in the southeastern U.S. Management of Bacterial Wilt in Tomatoes with Thymol and Acibenzolar S Methyl In this study we showed that the application of thymol as a soil fumigant and folia r application of ASM was able to control bacterial wilt on moderately resistant tomato cultivars and to a lesser degree on susceptible tomatoes. This would provide another tool in the small arsenal of defense against the causal agent of bacterial wilt. Pre viously, the two products were applied separately in field conditions and both were able to decrease disease incidence and increase fruit yield (Ji et al., 2005; Pradhanang et al., 2005). We determined that the application of both products will not have a negative e ffect on tomato production. We were able to show that t he combination of both products numerically increased fruit yield and decreased disease incidence for the susceptible cultivar. In both trials moderately resistant plants that received the
110 co mbination of both chemicals had statistically significant lower disease incidence and increased fruit yield compared to the untreated control ( UTC ) and in one experiment compared to thymol alone Previous studies have shown that both products are effecti ve at decreasing the incidence of different plant diseases ( Ji et al. 2005 ; Pradhanang et al., 2005). Further research should be done to determine the effectiveness of the combination of both products against multiple plant pathogens. If these chemicals a re successful in controlling a wide spectrum of diseases, they could be another alternative to methyl bromide. The next step would be to determine the minimum inhibitory concentration of thymol in field conditions to determine the most effective concentrat ion that is economically beneficial to growers. Grafting is a fairly recent method that has shown promise for controlling the disease (Rivard and Louws, 2008). Further research is also needed to determine if grafting can be used as part of an integrated ma nagement strategy with thymol and ASM. Diversity Amongst Ralstonia solanacearum strains from the southeastern United States Currently one of the largest concerns pertaining to R solanacearum is importation of exotic strains, especially sequevar 1 ( race 3 biovar 2 ). In order to determine if exotic strains have been introduced into a country, we must first understand the characteristics of the native strains. The major incidence of bacterial wilt is restricted predominantly to the southeastern U.S. Strains w ere collected representing three different geographic regions, the Carolinas, Georgia, and Florida. The objective of the study was to determine the attributes of these strains, and then characterize an exotic strain.
111 Strains collected in this study were c haracterized based on pathogenicity tests, biovar, HR tests on tobacco, and phylotype using multiplex PCR. All the strains that were pathogenic were able to wilt tomato plants, and a few were able to wilt pepper. Most of the Carolina strains did not elicit a hypersensitive response (HR) on tobacco, while most of the Florida strains were HR positive. All identified strains native to the U.S. belonged to phylotype II. We determined sequevar by phylogenetic analysis of egl gene. Strains were assigned to sequev ar based on cladistic grouping with previously published sequences of strains (Fegan and Prior, 2005, Wicker et al., 2007 and Ji et al., 2008). Native U.S. strains grouped within sequevar 7. Strains that grouped within sequevars 4, 5/6 and an undefined seq uevar were identified as exotic strains. The designation of an undefined sequevar was due to a lack of previously identified strains within this clade. Previously, it was reported that the undefined sequevar was part of sequevar 5 (Ji et al., 2008) or 4NPB (Hong et al, 2008). However, by including more taxa than the previous two studies used in their analyses, these strains did not group with sequevar 5 or 4NPB strains. One strain that belonged to the undefined sequevar was further characterized. This stra in, RS37, was isolated in North Florida from the waterways, tomato fields, and different plant species (Hong et al., 2008). The typical sequevar 7 strains were not detected in any of the samplings. It was determined that RS37 was able to produce a bacterio cin against RS5, a sequevar 7 strain, previously isolated in North Florida. We further characterized the strain by performing pathogenicity tests on cucurbits and Musa species. This strain was not pathogenic on the cucurbit cultivars tested. We were able
112 t o show that inoculation of RS37 was deleterious to the growth of Musa plants. Plants inoculated with RS37 had less root mass and were shorter in height compared to the water control or race 1 inoculated plants. We also showed that symptoms can vary dependi ng on Musa genotype. Based on the results of this study, it was determined that exotic strains have become established in the U.S., and they have different genotypic and phenotypic characteristics than the native strains. We recommend further characteriza tion of RS37. Although we were able to show that RS37 produces a bacteriocin, the next step would be to determine if this strain is more competitive than RS5 and other sequevar 7 strains. Further characterization of the other exotic strains identified in t his study should be performed to determine if other important economic crops could be potential hosts. Multiple Loci Variable Number Tandem Repeat Analysis Among Ralstonia solanacearum Strains from the southeastern United States In this study we identifie d microsatellites with their flanking regions (MALs) that could be used for determining R. solanacearum infrasubspecies groups. These microsatellites were identified using GMI1000 and the draft sequences of other strains. Primers were designed for the ampl ification of these microsatellites and their flanking regions Three SSRs were identified that were consistently amplified in a panel of R. solanacearum strains. Using the strains in our collection these three SSRs were amplified and sequenced. Phylogeneti c analysis was performed on microsatellites alone, the MALs, and then the combination of the three MALs together. Trees produced by the analysis of combination of the three MALs were the more informative than the SSRs or the MALs alone trees. The trees pr oduced by the combination of the MALs displayed 4 groups within phylotype II. These groups correspond to clades (sequevars)
113 found in the egl phylogenetic trees. This confirms that these loci were able to determine sequevar. The trees produced by the egl an d these loci were able to differentiate between strains that were native to the U.S. and were exotic. Using the three methods for phylogenetic analysis, maximum parsimony, maximum likelihood and Bayesian, we had high confidence that groupings created in th e trees were correct. Further research in combining microsatellites with phenotypic characteristics could give more support to infrasubspecies groups. To date the quickest method for identifying sequevar 1 ( race 3 biovar 2 ) strains is by real time PCR ; however at times this method can be unreliable. As noted in this study some SSRs are unique to a specific population. It could be possible to identify SSRs or regions in the flanking regions of the SSRs that would be specific to sequevar 1 strains. This would provide another confirmation test when identifying specific groups of strains. Microsatellites were useful in determining infrasubspecies groups within R. solanacearum Analysis of these loci was able to determine with confidence the regional origin of the strains tested in our collection. This information could be vital for identifying the potential geographic origin when bacterial outbreaks occur. Further research should be done to determine if MALs could be located in other phytobacteria and used to identify subspecies groups. Hopefully by identifying these populations researchers will be able to distinguish epidemiological and ecological groupings, and thus be able to predict the characteristics of unknown strains. This information should be impor tant for plant breeders, pathologists and quarantine officials.
114 APPENDIX SEQUENCE S AND TREES Figure A.1. An example of the diversity of SSR 9 of R. solanacearum strains isolated. Strains without P9 at the end of its name indicate strains obtained from NCBI BLAST.
115 Figure A.2. An example of the diversity of SSR 1 of R. solanacearum strains isolated. Strains without P1 at the end of its name indicate strains obtained fro m NCBI BLAST.
116 Figure A 3. A majority rule consensus rooted tree cr eated by Bayesian analysis of SSR 3 Numbers indicated the confidences of correct position of clades based on creating the consensus tree. he sequence was acquired from GenBank.
117 F igure A 4. A majority rule consensus unrooted tree created by Bayesian analysis of SSR 3 Numbers indicated the confidences of correct position of clades based on creating the consen sus tree. indicates the sequence was acquired from GenBank.
118 Figure A.5. A majority rule consensus rooted tree cr eated by Bayesian analysis of MAL 9 Numbers indicated the confidences of correct position of clades base d on creating the consensus tree. sequence was acquired from GenBank.
119 Figure A 6. A majority rule consensus rooted tree cr eated by parsimony analysis of MAL 9 Numbers indicated the confidences of correct position of clades based on creating the consensus tree. sequence was acquired from GenBank.
120 LITERATURE CITED Aeschbach, R. Lliger J., Scott, B. C. Murcia, A ., Butler, J ., Halliwell, B. and Aruoma, O. 1994. Antioxidant actions of thymol, carvacrol, 6 gingerol, zingerone and hydroxy tyrosol. Food. Chem. Toxic. 32: 31 36. Alvarez, A. 2005. Diversity and diagnosis of Ralstonia solanacearum Bacterial Wilt Disea se and the Ralstonia solanacearum Species Complex. Ed. by Allen, C., Prior, P., and Hayward, A. C, The American Phytopathological Society, St. Paul, MN. pp. 437 4 47. Anith, K. N., M. T. Momol, J. W. Kloepper, J. J. Marois, S. M. Olson, and J. B. Jones. 200 4. Efficacy of plant growth promoting rhizobacteria, acibenzolar s methyl, and soil amendment for integrated management of bacterial wilt on tomato. Plant Diseas. 88: 669 6 73. Anuratha, C. S., Gnanamanickam, S. S. 1990. Biological control of bacterial wilt caused by Pseudomonas solanacearum in India with antagonisti c bacteria. Plant and soil 124: 109 1 16. Backeljau, T., De Bruyn, L., De Wolf, H., Jordaens, K., Van Dongen, S., and Winnepenninckx, B. 1996. Multiple UPGMA and Neighbor joining trees and the performance of some compute r packages. Mol. Biol. Evol 13: 309 3 13. Bayer, U. 1985. Lecture notes in earth sciences. Vol. 2. Pattern recognition problems in geology and paleontology. Springer Verlag, Berlin/Heidelberg. Induction of systemic resistance to Pythium damping off in cucumber plants by benzothiadiazole: ultrastructure and cytochemistry of the host response. The Plant J 14: 13 21 Bickford D Lohman D J Sohdi N S Ng P K L, Meier R Winker K Ingr am K K and Das I 2007 Cryptic species as a window on diversity and conservation. Trends Ecol Evol 22:148 1 55. Bikandi, J., San Milln, R., Rementeria, A., and Garaizar, J. 2004. In silico analysis of complete bacterial genomes: PCR, AFLP PCR, and endonuclease restriction. Bioinformatics 20:798 79 9. Boshou, L. 2005. A broad review and perspective on breeding for resistance to bacterial wilt. Bacterial Wilt Disease and the Ralstonia solanace arum Species Complex. Ed. by Allen, C., Prior, P., and Hayward, A. C, The American Phytopathological Society, St. Paul, MN. pp. 225 2 38. Bradley, D. E. 1967. Ultrastructure of Bacteriophages and Bacteriocins. Bacteriolgic. Rev. 31:230 314.
121 Broza, Y. Y., Da nin Poleg, Y., Lerner, L., Broza, M., and Kashi, Y. 2007. Vibrio vulnificus typing based on simple sequence repeats: insights into the biotype 3 group. J. Clin. Microbiol 45: 2951 29 59. Buddenhagen, I., Sequeira, L., Kelman, A. 1962 Designation of races i n Pseudomonas solanacearum Phytopath 52: 726. Buddenhagen, I., and Kelman, A. 1964. Biological and physiological aspects of bacterial wilt caused by Pseudomonas solanacearum Annu. Rev. Phytopathol. 2:203 230. Burrill, T. J. 1890. Preliminary notes upon t he rotting of potatoes. Soc. Prom. Agr. Sci., Pro. 11th Ann. Meeting 8: 21 22. Bybee, S. M. Ogden, T. H., Branham, M. A., and Whiting, M. F. 2008. Molecules, morphology and fossils: a comprehensive approach to odonate phylogeny and the evolution of the odon ate wing. Cladistics 24: 477 514. Cailet, S., and Lacroix, M. 2006. Effect of Gamma Radiation and Oregano Essential Oil on Murein and ATP Concentration of Listeria monocytogenes J. of Food Protect. 69: 2961 29 69. Cardozo, C., Rodguez, P., and Marn, M. 200 9. Molecular characterization of Ralstonia solanacearum species complex in the banana growing regio n of Uraba. Agro Colombiana 27: 203 10. Caruso, P., Palomo, J. L., Bertolini, E., lvarez, B., Lpez, M., and Biosca E. 2005. Seasonal variation of Ralstonia solanacearum biovar 2 populations in a Spanish river: recovery of stressed cells at low temperatures. App l and Environment Microbiol 71: 140 14 8. Castillo, J. A. and Greenberg, J. T. 2006. Evolutionary dynamics of Ralstonia solanacearum. Appl Environ Microb iol. 73:1225 12 38. Champoiseau, P. G., Jones, J. B. and Allen, C. 2009. Ralstonia solanacearum race 3 biovar 2 causes tropical losses and temperate anxieties. PHP 2009 0313 01 RV. Chinnasri, B., Sipes, B. S. and Schmitt, D. P. 2003. Effects of acibenzola r s methyl application to Rotylenchulus reniformis and Meloidogyne javanica J. Nematol. 35: 110 11 4. Cook, D. R., and Sequeira, L. 1988. The use of restriction fragment length polymorphism (RFLP) analysis in taxonomy and diagnosis. ACIAR Bacterial Wilt New sletter. No. 4, 4. Danin Poleg, Y., Cohen, L. A., Gancz, H., Broza, Y. Y., Goldshmidt, H., Malul, E., Valinsky, L., Lerner, L., Broza, M., and Kashi, Y. 2007. Vibrio cholerae strain typing and phylogeny study based on simple sequence repeats. J. Clin. Micr obiol. 45:736 7 46.
122 DeLong, E. F., and Pace, N. R. 2001. Environmental diversity of bacteria and archaea. Syst. Biol. 50:470 47 8. Denny, T. 2006 Plant pathogenic Ralstonia species: Plant associated bacteria. S. S. Gnanamanickam, ed. Springer, Netherlands. p p: 573 644. Department for Environment, Food and Rural Affairs (DEFRA). 2003. Potato brown rot: monitoring programme. Retrieved Jun. 24, 2005, from http://www.defra.gov.uk/planth/pbr1.htm. Delespaul, Q., de Billerbeck, V. G., Roques, C. G., and Michel, O. 2000. The antifungal activity of essential oils as determined by different screening metho ds. J. Essent. Oil Res. 12:256 2 66. Diamant, E., Palti, Y., Gur Arie, R., Cohen, H., Hallerman, E., M., and Kashi, Y. 2004. Phylogeny and strain typing of Echerichia coli inferred from variation at mononucleotide repeat loci. Applied and Environmenl Microb. 70:2464 24 73. Edgar, R.C. 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res 32: 1792 1797. Elphinstone, J. G., St anford, H. M., and Stead., D. E. 1998. Survival and transmission of Ralstonia solanacearum in aquatic plants of Solanum dulcamara and associated surface w ater in England. Bull. OEPP 28: 93 94. Elphinstone, J. G. 2005. The current bacterial wilt situation: a global overview. Bacterial Wilt Disease and the Ralstonia solanacearum Species Complex. Ed. by Allen, C., Prior, P., and Hayward, A. C, The American Phytopathological Society, St. Paul, MN. pp. 9 28. Englebrecht, M. C. 1994. Modification of a semi selecti ve medium for the isolation and quantification of Pseudomonas solanacearum Bacterial Wilt Newsletter 10:2 5. Evans, J. D., and Martin, S. A. 2000. Effects of thymol on ruminal micro organisms. Curr. Microbiol. 41: 336 3 40. Farlow, J., Postic, D., Smith, K. L., Jay, Z., Baranton, G., and Keim, P. 2002. Strain typing of Borrelia burgdorferi Borrelia afzelii and Borrelia garinii by using multiple locus variable number tandem repeat an alysis. J. Clin. Microbiol. 40: 4612 46 18. Fegan, M., Holoway, G., Hayward, A. C., and Timmis, J. 1998. Development of a diagnostic test based on the polymerase chain reaction to identify strains of R. solanacearum exhibiting the biovar 2 genotype.: Bacterial Wilt Disease: Molecular and Ecological Aspects. P. Prior, C. Allen, and J. Elphinstone, eds. Springer Verlag, Berlin. pp. 34 43
123 Fegan, M., Taghavi, M., Sly, L. I., and Hayward, A. C. 1989. Phylogeny, diversity and molecular diagnostics of Ralstonia solanacearum : Bacterial Wilt Disease: Molecular and Ecological Aspects. P. Pr ior, C. Allen, and J. Elphinstone, eds. Springer Verlag, Berlin. pp. 19 33. Fegan, M., and Prior, P. 2005 How complex is the Ralstonia solanacearum species complex? Bacterial Wilt: The Disease and the Ralstonia solanacearum Species Complex. C. Allen, P. P rior, and A. C. Hayward, eds. American Phytopathological Society, St. Paul, MN. pp. 449 461 Fegan, M., and Prior, P. 2006. Bacterial wilts of banana caused by membe rs of the Ralstonia solanacearum species complex Australas. Plant Pathol. 35:93 101. Felse nstein, J. 1978. Cases in which parsimony or compatibility methods will be posit ively misleading. Syst Zool 27: 401 4 10. Felsenstein, J. 1985. Confidence limits on phylogenies: An approach using the bootstrap. Evolution 39:783 791. Francois, P., Huyghe, A., Charbonnier, Y., Bento, M., Herzig, S., Topolski, I., Fleury, B., Lew, D., Vaudaux, P., and Harbarth, S. 2005. Use of an automated multiple locus, variable number tandem repeat based method for rapid and high throughput genotyping of Staphylococcus aureus is olates. J. Clin. Microbiol. 43: 3346 33 55. French, E.B., L. Gutarra, P. Aley, and J. Elphinstone. 1995. Culture media for Ralstonia solanacearum isolation, identification, and maintenance. Fitopatologia 30:126 1 30. French, E. R., and Sequeira, L. 1968. Bacterial wilt or moko of pl antain in Peru. Fitopatologia 3: 27 38. Genin, S., and Boucher, C. 2002. Ralstonia solanacearum : secrets of a major pathogen unveiled by analysis of its ge nome. Mol. Plant Pathol. 3: 111 118. Gillings, M. R., Fahy, P., and Davies, C. 1993. Restriction analysis of an amplified polygalacturonase gene fragment differentiates strains of the phytopathogenic bacterium, Pseudomonas solanacearum Lett. in Appl. Microbiol. 17:44 4 8. Goloboff, P. A., Catalano, S. A., Marcos Mirande, J., Szumik, C. A., Salvador Arias, J., Kllersj, M. and Farris, J. S. 2009. Phylogenetic analysis of 73,060 taxa corroborates major eukaryotic groups. Cladistics 25: 211 30. Granada, G. A., and Sequeira, L. 1983. Survival of Pseudomo nas solanacearum in soil, rhizosphere, and plant roots. Can. J. Microbiol. 29:433 440. Guidot, A., Prior, P., Schoenfeld, J., Carrre, S., Genin, S., and Boucher, C. 2007. Genomic structure and phylogeny of the plant pathogen Ralstonia solanacearum inferre d from gene distribution analysis. J of Bacteriol. 189:377 3 87.
124 Hawks, B. 2002. Agricultural Bioterrorism Protection Act of 2002; Listing of Biological Agents and Toxins and Requirements and Procedures for Notification of Possession. Accessed April 13, 20 08 Internet http://www.fas.org/sgp/news/2002/08/fr081202.html Hayward, A. C. 1964. Characteristics of Pseudomonas solanacearum J of Applied Bacteriol 27: 265 2 77. Hayward, A. C. 1991. Bio logy and epidemiology of bacterial wilt caused by Pseudomonas solanacearum Annu. Rev. Phytopathol. 29:65 87. Hayward, A.C. 1995. Pseudomonas solanacearum In Pathogenesis and Host Specificity in Plant Diseases: Histopathological, Biochemical, Genetic and Molecular Bases (Vol. 1) (Singh, U.S. et al., eds), Elsevier Science. pp. 139 151 Hillis, D. M. Huelsenbeck, J. P., Cunningham, C. W. 1994. Application and accuracy of molec ular phylogenies. Science. 264: 671 27 7. Holt, J. G., Krieg, N. R., Sneath, P. H. A ., Staley, J. T., and Willims, S. T., (eds.). 1994. Bergey's Manual of Determinative Bacteriology, 9th edition. Baltimore: Williams and Wilkins. pp.11 14 Hommais, F., Pereira, S., Acquaviva, C., Escobar Pramo, P., and Denamur, E. 2005. Single nucleotide polymorphism phylotyping of Escherichia coli Applied and Environmen Microb 71:4784 92. Hong, J. C., Ji, P., Momol, M. T., Jones, J. B., Olson, S. M., Pradhanang, P., and Guven, K. 2004. Ralstonia solanacearum detection in tomato irriga tion ponds and weeds. First International Symposium on Tomato Diseases and 19th Annual Tomato Disease Workshop. Hong, J. C., Momol, M. T., Jones, J. B., Ji, P., Olson, S. M., Allen, C., Sanches Perez, A., Pradhanang, P., and Guyen, K. 2008. Detection of R alstonia solanacearum in Irrigation Ponds and Aquatic Weeds Associated with the Ponds in North Florida. Plant Dis. 92:1674 16 82. Hoyer, B. H., McCarthy, B. J., and Bolton, E. T. 1964. A molecular approach in the systematics of higher organisms. Science. 14 4:959 9 67. Hueck, C. J. 1998. Type III protein secretion systems in bacterial pathogens of animals and plants. Microboil. Mol. Biol. Rev. 62:379 433. Huelsenbeck, J. P., Ronquist, F., Nielsen, R., and Bollback, J. P. 2001. Bayesian inference of phylogeny a nd its impact on evolutionary biology. Science 294:2310 23 14 Jacobs, D. K. Haney, T. A., and Louie, K. D. 2004 Genes, diversty and geological process on the pacific coast. An nu. Rev. Earth Planet. Sci. 32: 601 6 52.
125 Janse, J. D. Araluppan, F. A. X., Schans, J., Wenneker, M., and Westerhuis, W. 1997. Experiences with bacterial brown rot Ralstonia solanacearum biovar 2, race 3 in the Netherlands. Bacterial Wilt disease. Molecular and ecological aspects. Prior, P., Allen, C., and Elphinstone, J. G. eds, Springe r Verlag, Heidelberg, Germany. pp.146 52. Janse, J. D., van den Beld, H. E., Elphinstone, J., Simpkins, S., Tjou Tam Sin, L. N. A., and van Vaerenbergh, J. 2005. Introduction to Europe of Ralstonia solanacearum biovar 2, race 3, in Pelargonium zonale cutti ngs from Kenya. Bacterial Wilt Disease and the Ralstonia solanacearum Species Complex. Ed. by Allen, C., Prior, P., and Hayward, A. C., The American Phytopathological Society, St. Paul, MN. pp. 81 94. Jaunet, T. X., and Wang, J. F. 1999. Variation in genot ype and aggressive ness diversity of Ralstonia solanacearum race 1 isolated from tomato in Taiwan. Phytopath. 89:320 327. Jeong, Y., Kim, J., Kang, Y., Lee, S., and Hwang, I. 2007. Genetic Diversity and Distribution of Korean Isolates of Ralstonia solanacearum Plant Dis. 91:1277 12 87. Ji, P., Allen, C., Sanchez Perez, A., Yao, J., Elphinstone, J. G., Jones, J. B., and Momol, M. T. 2007. New diversity of Ralstonaia solanacearum strains associated with vegetable and ornamental crops in Florida. Plan t Dis. 91:195 203. Ji, P., Momol, M. T., Olson, S. M., Pradhanang, P. M., and Jones, J. B. 2005. Evaluation of thymol as biofumigant for control of bacterial wilt of tomato under field conditions. Plant Dis. 89:497 500. Ji, P., Momol, M. T., Rich, J. R., Olson, S. M., and Jones, J. B. 2007 Development of an integrated approach for managing bacterial wilt and root knot on tomato under field conditions. Plant Dis. 91:1321 13 26. Johnson, J. L., and Ordal, E. J. 1968. Deoxyribonucleic acid homology in bacter ial taxonomy: effect on incubation temperature on reaction specificity. J. of Bacteriol. 95:893 900. Keim, P., Klevytska, A. M., Smith, K. L., Schupp, J. M., Okinaka, R., Jackson, P. J., and Hugh Jones, M. E. 1999. Multiple locus variable number tandem rep eat analysis reveals genetic relationships within Bacillus anthracis J. Bacteriol 182: 2928 29 36. Kelman, A. 1953. The Bacterial Wilt caused by Pseudomonas solanacearum A Literature Review and Bibliography. A N. C. State College Publication, Raleigh, N. C. Kelman, A. 1954. The relationship of pathogenicity in Pseudomonas solanacearum to colony appearance on a tet razolium medium. Phytopath. 44: 693 695.
126 Keys, C., Kemper, S ., and Keim, P. 2005. Highly diverse variable number tandem repeat loci in the E. coli O157: H7 and O55: H7 genomes for high resolution molecular typing. J. Appl. Microbiol. 98:928 9 40. Konstantinids, K. T., Ramette, A., Tiedje, J. M. 2006. The bacterial species definition in the genomic era. Phil. Trans. R. Soc. 361: 1929 19 40. Kovach, W. L 1993 MVSP Plus: multivariate statistical package, version 2.1. Kovach Computing Services, Pentraeth, Anglesey. Kuc, J. 2001. Concepts and directions of induced systemic resistance in plants and its application. Eur. J. Plant Pathol. 107:7 12. Kuhner, M. K., and Felsenstein, J. 1994 A simulation comparison of algorithms under equal and unequal evolutionary rates. Mol. Biol. Evol. 11:459 4 68. Lambert, R. J. W., Skandamis, P. N., Coote, P. J. and Nychas. G. J. E. 2001. A study of the minimum inhibitory con centration and mode of action of oregano essential oil, thymol and carvacrol. J. of App. Micro. 91:453 4 62. Lee, S., Tsao, R., Peterson, C. J., and Coats, J. R. 1997. Insecticidal activity of monoterpenoids to western corn rootworm (Coleoptera: Chrysomelida e), twospotted spider mite (Acari: Tetranychidae), and house fly (Diptera: Muscidae). J. Econ. Entomol. 90: 8 83 892. Lelliott, R. A., and Stead, D. E. 1987. Methods for the Diagnosis of Bacterial Diseases of Plants. Blackwell Scientific, Oxford, UK. Lewis I vey, M. L, McSpadden Gardner, B. B., Opina, N., and Miller, S. A. 2007. Diversity and geographic distribution of Ralstonia solanacearum from eggplant in the Philippines. Phytopathology 97:1467 1475. Li, W., Raoult, D., and Fournier, P. 2009. Bacterial stra in typing in the geno mic era. FEMS Microbiol Rev 33: 892 916. Li, X. and Hayward, A. C. 1993. The use of the Biolog identification system for the rapid identification of plant pathogenic Pseudomonads In G. L. Hartman & A. C. Hayward (Eds.), Bacterial Wilt. ACIAR Proceedings No. 45 Canberra, Australia: Australian Centre for International Agricultural Research. pp. 45 48. Li, Y., Korol, A. B., Fahima, T., and Nevo, E. 2004. Microsatellites within genes: structure, function, and evolution. Mol. Biol. Evol. 21: 999 1 0 07. Lindstedt, B.A. 2005. Multiple locus variable number tandem repeats analysis for genetic fingerprinting of pathogenic bacteria. Electrophoresis 26: 2567 25 82.
127 Liolios K, Mavrommatis K, Tavernarakis N, and Kyrpides, NC. 20 09 The Genomes On Line Dat abase (GOLD) in 2007: status of genomic and metagenomic projects and their associated metadata. NAR 36, D475 D479. Louws, F. J., Wilson, M. Campbell, H. L. Cuppels, D. A. Jones, J. B. Shoemaker, P. B. Sahin, G. and Miller, S. A. 2001. Field control of bacterial spot and speck of tomato using a plant activator. Plant Dis. 85: 481 4 88. Lui, H., Zhang, S., Schell, M. A., and Denny, T. P. 2005. Pyramiding unmarked deletions in Ralstonia solanacearum shows that secreted proteins in addition t o plant cell wall degrading enzymes contribute to virulence. Mol. P lant Microbe Interact. 18:1296 1 305. Liu, Y., Kanda, A., Yano, K., Kiba, A., Hikichi, Y., Aino, M., Kawaguchi, A., Mizoguchi, S., Nakaho, K., Shiomi, H., Takikawa, Y., and Ohnishi, K. 2009. Molecular typing of Japanese strains of Ralstonia solanacearum in relation to the ability to induce a hypersensitive reaction in tobacco. J. Gen Plant Pathol 75:369 3 80. Lu, Z. X., Gaudet, D. Puchalski, B ., Despins, T. Frick M. and Laroche A. 2006. Inducers of resistance reduce common bunt infection in wheat seedlings while differentially regulating defence gene expression. Physiol. and Molecul. Plant Pathol. 67:138 1 48. Maiden, M. C. J., Bygraves, J. A., Feil, E., Morelli. G., Russell, J. E., Urwin R., Zhang, Q., Zhou, J., Zurth, K., Caugant, D. A., Feavers, I. M., Achtman, M., and Spratt, B. B. 1998. Multilocus sequence typing: a portable approach to the identification of clones within populations of pathogenic microorganisms. Proc. Natl. Acad. Sc i. USA 95:3140 314 5. Martin, C., French, E. R., and Nydegger, U. 1982. Strains of Pseudomonas solanacearum affecting Solanaceae in the Americas. Plant Dis. 66:458 460. Martin, F. N. 2003. Development of alternative strategies for management of soilborne p atho gens currently controlled with methyl bromide. Annu. Rev. Phytopathol. 41:325 350. McCarter, S. M. 1991. Bacterial wilt. Compendium of tomato disease. J. B. Jones, J. P. Jones, R. E. Stall, and T. A. Zitter, eds. The American Phytopathological Societ y, St. Paul. pp 28 9. McGarvey, J. A., Denny, T. P., and Schell, M. A. 1999. Spatial temporal and quantitative analysis of growth and EPS I production by Ralstonia solanacearum in resistant and susceptible tomato cultivars. Phytopath 89: 1233 39. McLaughlin R. J., and Sequeira, L. 1989. Phenotypic diversity in strains of Pseudomonas solanacearum isolated from a single potato field in north eastern Florida. Plant Dis. 73:960 964.
128 Metzgar, D. Thomas, E., Davis, C., Field, D., and Wills, C. 2001. The microsate llites of Escherichia coli : rapidly evolving repetitive DNAs in a non pathogenic prokar yote. Mol. Microbiol. 39: 183 1 90. MIDI, Inc. 2001. Identification of bacteria by gas chromatography of cellular fatty acids. Retrieved June 17, 2005, from http://www.mid iinc.com/media/pdfs/TechNote_101.pdf. Nakaho, K., Hibino, H., and Miyagawa, H. 2000. Possible Mechanisms limiting movement of Ralstonia solanacearum in resistant tomato tissues. J of Phytopah. 148: 181 1 90. Noling, J. W. and Becker, J. O. 1994. The challenge of research and extension to define and implement alternatives to methyl bromide. J. Nematol. 26:573 586. Norelli, J. L., Jones, A. L. and Aldwinckle. H. S. 2003. Fire blight management in the twenty first century using new technologies that enh ance host resi stance in apple. Plant Dis. 87: 756 7 65. Norman, D. J., Zapata, M., Gabriel, D. W., Duan, Y. P., Yuen, J. M. F., Mangravita Novo, A., and Donahoo, R. S. 2009. Genetic diversity and host range variation of Ralstonia solanacearum entering North America. Phytopath. 99:1070 10 77. Obradovic, A., Jones, J. B., Momol, M. T., Olson, S. M., Jackson, L. E., Balogh, B., Guven, K., Iriate, F. B. 2005. Integration of biological control agents and systemic acquired resistance inducers against bacterial s pot on tomato. Plant Dis. 89: 712 71 6. sense and sensitivity. Cladistics. 19: 432 4 42. Oostendrop, M., Kunz, W., Dietrich, B., and Staub, T. 2001. Induced disease resistance in pla nts by chemicals. Eur. J. Plant Pathol. 107:19 28. Opina, N., Tavner, F., Hollway, G., Wang, J. F., Li, T. H., Maghirang, R., Fegan, M., Hayward, A. C., Krishnapillai, V., and Hong, W. F. 1997. A novel method for development of species and strain specific DNA probes and PCR primers for identifying Burkholderia solanacearum (formerly Pseudomonas solanacearum ). Asia Pa cific J.Mol.Biol.Biotechnol. 5 :19 30. Pace B., and Campbell, L. L. 1971. Homology of ribosomal ribonucleic acid of diverse bacterial species wi th Escherichia coli and Bacillus stearothermophilus J. of Bacteriol. 107:543 5 47. Palleroni, N. J. 2003. Prokaryote taxonomy of the 20 th century and the impact of studies on the genus Pseudomonas: a personal view. Microbiol. 149:1 7.
129 Palleron i, N. J., and Doudoroff., M. 1971 Phenotypic characterization and deoxyribonucleic acid homologies of Pseudomonas solanacearum J. of Bacteriol. 107:690 69 6. Pastrik, K. H. and Maiss, E. 2000. Detection of Ralstonia solanacearum in potato tubers by polyme rase chain reaction. Pht opathologische Zeitschrift 148: 619 6 26. Poussier, S., Trigalet Demery, D., Vandewalle, P., Goffinet, B., Luisetti, J., and Trigalet, A. 2000. Genetic diversity of Ralstonia solanacearum as assessed by PCR RFLP of the hrp gene region, AFLP and 16s rRNA sequence analysis, and identification of an African subdivision. Microbiol. 146:1679 16 92. Posada, D., and Crandall, K. A. 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14:817 818. Pradhanang, P. M. and Momol, M. T. 2001. Survival of Ralstonia solanacearum in soil under irrigated rice culture and aquatic weeds. J. Phytopathol. 149:707 711. Pradhanang, P. M., Momol, M. T., Olson, S. M., and Jones, J. B. 2003. Effects of plant essential oils on Ralsto nia solanacearum population density and bacterial wilt incidence in tomato. Plant Dis. 87:423 427. Pradhanang, P.M., Ji, P., Momol, M.T., Olson, S.M., Mayfield, J.L., and Jones.,J.B. 2005. Application of acibenzolar S methyl enhances host resistance in tom ato against Ralstonia solanacearum Plant Dis. 89: 989 993. Prior, P., and Fegan. M. 2005. Diversity and molecular detection of Ralstonia solanacearum race 2 strains by multiplex PCR In C. Allen, P. Prior, and A. C. Hayward (ed.), Bacterial wilt disease a nd the Ralstonia solanacearum species complex. APS Press, Madison, WI. p p. 405 4 14. Rambaut A. 1996. Se Al v2.0a11: se quence al ignment editor. Available from: http://evolve.zoo.ox.ac.uk/ Rambaut A. 2006. FigTree v1.3.1. Available from: http://evolve.zoo.ox.ac.uk/ Remenant, B., Coupat Goutaland, B., Guidot, A., Cellier, G., Wicker, E., Allen, C., Fegan, M., Pruvost, O., Elbaz, M., Calteau, A., Salvignol, G., Mornico, D., Mangenot, S., Barbe, V., Mdigue, C., and Prior, P. 2010. Genomes of three tomato pathogens within the Ralstonia solanacearum species complex reveal significant evolutionary divergence. BMC Genomics 11:379. Roberston, A. E., Fortnum, B. A., Wood, T. C., and Kluepfel, D. A. 2001. Diversity of Ralstonia sola nacearum in the southeastern United States. Beitr. Tabakforsch. Int. 19:323 331
130 Robertson, A. E., Wechter, W. P., Denny, T. P., Fortnum, B. A., Kluepfel, D. A. 2004. Relationship Between Avirulence Gene ( avrA ) Diversity in Ralstonia solanacearum and Bac ter ial Wilt Incidence. Mole. Plant Microbe Interact. 17:1376 13 84. Rojano, B., Saez, J ., Schinella, G., Quijano, J. Vlez, E. Gil, A. and Notario. R. 2008. Experimental and theoretical determination of the antioxidant properties of isoespintanol (2 isopro pyl 3,6 dimethoxy 5 methylphenol). J. of Mol. Struc. 877:1 6. Ronquist, F., and Huelsenbeck, J. P. 2003 MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 1572 157 4. Saddler, G.S. 2004. Management of bacterial wilt disease. Bacterial Wilt: The Disease and the Ralstonia solanacearum Species Complex. C. Allen, P. Prior, and A. C. Hayward, eds. American Phytopathological Society, St. Paul, MN. pp 121 32. Salanoubat,M., Genin,S., Artiguenave,F., Gouzy,J., Mangenot,S., Arlat,M., Billault,A., Brottier,P., Camus,J.C., Cattolico,L., Chandler,M., Choisne,N., Claudel Renard,C., Cunnac,S., Demange,N., Gaspin,C., Lavie,M., Moisan,A., Robert,C., Saurin,W., Schiex,T., Siguier,P., Thebault,P., Whalen,M., Wincker,P., Levy,M., Weissenb ach,J. and Boucher,C.A. 2002. Genome sequence of the plant pathogen Ralstonia solanacearum Nature 415: 497 502. Sambrook, J., Fristsch, E., and Maniatis, T. 1989. Molecular cloning: a laboratory manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Sanchez Perez, A., Mejia, L., Fegan, M., and Allen, C. 2008. Diversity and distribution of Ralstonia solanacearum strains in Guatemala and rare occurrence of tomato fruit infection. Plant Path. 57:320 3 3 1. Santos, B. M., Gilreath, J. P. Motis, T. N. Noling, J. W. Jones, J. P. and Norton, J. A 2006. Comparing methyl bromide alternatives for soilborne disease, nematode and weed management in fresh market tomato. Crop Protec. 25:690 6 95. Scheie, A. A. A. 1989. Modes of action of currently known chemical anti plaque agents other than chlorhexidine. J. Dental Res. 68:1609 1616. Schell, M. A. 2000. Control of virulence and pathogenicity genes of Ralstonia sol anacearum by an elaborate sensory ar ray. Annu Rev of Phytopath. 38: 263 2 92. tolerant, bacterial wilt tolerant tomato. HortScience 30:641 642. M., Kosalec I. J. Pieckov, E. and Pepeljnak S. 2006. Antifungal activity of thyme ( Thymus vulgaris L.) essential oil and thymol against moulds from damp dwellings. Letters in Appl. Microbiol. 44: 36 42.
131 Seal, S. 1998. Molecular methods for detection and discrimination of Ralstonia solanacearum. .: Bacterial Wilt Disease: Molecular and Ecological Aspects. P. Prior, C. Allen, and J. Elphinstone, eds. Sp ringer Verlag, Berlin. pp. 103 15. Shapira, R. and E. Mimran. 2007. Isolation and Characterization of Esch erichia coli Mutants Exhibiting Altered Response to Thymol. Microb. Drug Resist. 13:157 65. Smith, J. J., Offord, L. C., Kibata, G. N., Murimi, Z. K., Trigalet, A., and Saddler, G. S. 1998. The development of a biological control agent against Ralstonia so lanacearum Race 3 in Kenya. Bacterial wilt disease: molecular and ecological aspects. Ed. by Prior, P., Allen, C., and Elphinstone, J. Springer Verlag Berlin Heidelber New York. pp. 337 3 42. Smith, L. J., 2007. Host range, phylogenetic, and pathogenic dive rsity of Corynespora cassiicola (Berk. & Curt.) Wei. University of Florida, 2008, Gainesville, FL. http://purl.fcla.edu/fcla/etd/UFE0022669 Stamatakis, A., Ludwig, T. and Meier. H. 2005. Raxml iii: A fast program for maximum likelihood based inference of l arge phylogenetic trees. Bioinfor 21:456 63. Staley, J. T. 2006. The bacterial species delemma and the genomic phylogenetic species concept. Philos. Trans. R. Soc. B. 361: 1899 909. Stavisky, J., Funderburk, J., Brodbeck, B. V., Olson, S. M., and Anders en, P. C. 2002. Population dynamics of Frankliniella spp. and tomato spotted wilt incidence as influenced by cultural management tactics i n tomato. J. Econ. Entomol. 95: 1216 12 21. Stead, D. E. 1992. Grouping of plant pathogenic and some other Pseudomonas spp. by using cellular fatty acid profile s. Int. J. Syst. Bacteriol. 42: 281 295. Swofford, D. L. 2003. PAUP*: Phylogenetic Analysis Using Parsimony (* and Other Methods), version 4.0. Sinaure Associates, MA. Taghavi, M., Hayward, C., Sly, L. I., and Fegan, M. 1996. Analysis of the phylogenetic relationships of strains of Burkholderia solanacearum, Pseudomonas syzgii, and the Blood Disease Bacterium of banana based on 16s rRNA gene sequences. Int J Syst Bacteriol 46:10 1 5. Thwaites, R., Mansfield, J., Eden G reen, S., and Seal, S. 1999. RAPD and rep PCR based fingerprinting of vascular bacterial pathogens of Musa spp. Plant Pathol. 48:121 12 8. Titze de Almeida, R., Willems, R. J. L., Top, J., Pereira Rodrigues, I., Fonseca Ferreira, R., Boelens, H., Brandileon e, M. C. C., Zanella, R. C., Soares Felipe, M. S., and van Belkum, A. 2004. Multilocus variable number tandem repeat polymorphism among Brazilian Enterococcus faecalis strains. J. Clin. Microbiol. 42: 4879 48 81.
132 Top, J., Schouls, L. M., Bonten, M. J. M., a nd Willems, R. J. L. 2004. Multiple locus variable number tandem repeat analysis, a novel typing scheme to study the genetic relatedness and epidemiology of Enterococcus faecium isolates. J. Clin. Microbiol 42: 4503 4511. Tortora, G. J., Funke, B. R., and Case, C. L. 1995. Microbiology. An Introduction. Fifth Edition. The Benjamin/ Cummings Publishing, Co. Inc., Redwood City, CA. pp. 255 272. Toukam, G. M. S., Cellier, G., Wicker, W., Guilbaud, C., Kahane, R., Allen, C., and Prior P. 2009. Broad diversity o f Ralstonia solanacearum strains in Cameroon. Plant Dis. 29:1123 11 30. U.S. Department of Agriculture (USDA). 1976. United States Standards for Grades of Fresh Tomato. U.S. Dep. Agric., Agric. Marketing Serv. pp. 1 11. van Belkum, A., Scherer, S., van Alph en, L., and Verbrugh, H. 1998. Short sequence DNA repeats in prokaryotic genomes. Mi crob and Molec Bio Reviews. 62: 275 2 93. Villa, J. E., Tsuchiya, K., Horita, M., Natural, M., Opina, N., and Hyakumachi, M. 2005. Phylogenetic relationship of Ralstonia sola nacearum species complex strain from Asia and other continents based on 16s rRNA, endoglucanase and hrpB gene sequences. J Gen Plant Pathol. 71:39 46. Volcani, Z. and Palti, J. 1960. Pseudomonas solanacearum in Israel. Plant Dis. Reporter. 44: 448 44 9. Wal ker, T. and Collion, N. H. 1998. Priority setting at CIP for the 1998 2000 Medium Term Plan. International Potato Center, Lima, Peru. Walsh, S. E., Maillard, J. Y., Russell, A. D., Catrenich, C. E ., Charbonneau, D. L., and Bar tolo, R. G. 2003. Activity and mechanisms of action of selected biocidal agents on Gram positive and ne gative bacteria. J. Appl. Micro biol. 94:240 247. Wang, J., and Lin, C. 2005. Integrated management of tomato bacterial wilt. AVRDC The World Vegetable Center. pp. 3 15. Weber, J. L. and Wong, C. 1993. Muttaion of human short tandem repeats. Hum. Mol. Genet. 2:1123 112 8. Wicker, E., Grassart, L., Coranson Beaudu, R., Mian, D., Guilbaud, C., Fegan, M., and Prior, P. 2007. Ralstonia solanacearum strains from Martinique (French West Ind ies) exhibiting a new pathogenic potential. Appl. Environ. Microbiol. 73:6790 6801. Woese, C. R. 1987 Bacterial evolution. Microbiol. Rev. 51 : 221 2 7 1.
133 Xu, J., Pan, Z. C., Prior, P., Xu, J. S., Zhang, Z., Zhang, H., Zhang, L. Q., He, L. Y., and Feng, J. 20 09. Genetic diversity of Ralstonia solanacearum strains from China. Eur. J. Plant Pathol. 125:641 6 53. Yabuuchi, E., Kosako, Y., Oyaizu, H., Yano, I., Hotta, H., Hashimoto, Y., Ezaki, T., and Arakawa, M. 1992. Proposal of Burkholderia gen. nov. and transfer of seven species of the genus Pseudomonas homology group II to the new genus, with the type species Burkholderia cepacia (Palleroni and Holmes 1981) comb. nov. Microbiol and Immunol. 36:1251 12 75. Yabuuchi, E., Y. Kosak, I. Yano, H. Hotta, and Ni shiuchi. Y. 1995. Transfer of two Burkholderia and an Alcaligenes species to Ralstonia gen. nov. Proposal of Ralstonia picketti (Ralston, Palleroni and Doudoroff 1973) comb. nov., Ralstonia solanacearum (Smith 1896) comb. nov. and Ralstonia eutropha (Davi s 1969) comb. nov. Microbiol. Immunol 39:897 904.Zachos, D.G. 1957. The brown rot of potatoes in Greece. Annales de l'Institut Phytopathologique Benaki New Series 1:115 11 7. Zehr, E. I. 1969. Studies of the distribution and economic importance of Pseudomo nas solanacearum in certain crops in the Philippines. Philippine Agriculturist 53:218 2 23.
134 BIOGRAPHICAL SKETCH Jas on Hong was born in Logan, Utah in 1978, but grew up in Wooster, Ohio. For the last two years of high school he was dual enrolled at the local community college and graduated with an Associate of Science the same year he graduated from high s chool. In 2003, he graduated from The Ohio State University with a Bachelor of Science degree in microbiology. During the summers of 2002 and 2003 he had internships with Dr. Jackson at AgriPhi (now Omnilytics) and was exposed to plant pathology, bacteriophage and research. In 2005 he graduated with his Master of Science from the University of Florida, College of Agricultural and Life Sciences Departme nt of Plant Pathology His m R. solanacearum strains in irrigation ponds, and determining methods for decreasing the bacterial population in the water. In 2005, he continued his research on R. solanacearum and was admitted into a Ph.D. program at the Plant Pathology Department at University of Florida from which he received his Ph.D. 2010 His research project focused on controlling bacterial wilt in the field and determining the biodiversity of R. solanacearum stra ins in the southeastern U.S. while looking at simple sequence repeats and microsatellites to aid in determining diversity.