<%BANNER%>

Evaluation of Bacteriocins in Xanthomonas perforans for Use in Biological Control of Xanthomonas euvesicatoria


PAGE 1

1 EVALUATION OF BACTERIOCINS IN Xanthomonas perforans FOR USE IN BIOLOGICAL CONTROL OF Xanthomonas euvesicatoria By AARON PAUL HERT A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2 ACKNOWLEDGMENTS My sincere thank out to Dr. Jeffrey B. Jone s, who served as my supervising committee cochair for my M.S. and Ph.D. degrees. I apprec iate all of the time that he spent sharing his ideas. He was always happy to discuss new ideas and theories. I also thank Dr. Timor Momol, who served as co-chair for my PhD and had the fa ith in me to complete my degree. I owe Dr. Pamela D. Roberts, who served as a co-chair during my masters degree and a committee member for my PhD. Her support both technically and fina ncially allowed me to complete my degree. I truly appreciate all her continued support of my work. My th anks also goes to Dr. Steve Olson, Dr. Jim Preston, and Dr. Martin Handfield fo r serving on my committee and all the helpful discussions about idea s and experiments. I would like to especial ly acknowledge Dr. Mizuri Marutani whom has been instrumental in much of the molecular aspects of my research and has played an integral part in bacteriocin analysis and attenuation mutant s creation. Her input and knowledge was utilized on a daily basis, I dont know if I could have done it with out her. I would like to thank Jerry Mi nsavage for his technical s upport and donation of several mutants to aid in my masters and Ph.D. degrees. The faculty and staff of the Department of Plant Pathology were very supportive throughout my program, providing a ssistance whenever it was needed. I also would like to thank my fellow graduate students, who provided continual support and friendship. Last but not least, I thank my family. My mother, fath er, brother, sister, and my niece and nephew have been in my heart and mind with me every step of th e way. Id like to thank all of my family for being there for me since day one, with all the love and belief in me that I could ask for.

PAGE 3

3 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................2 LIST OF TABLES................................................................................................................. ..........6 LIST OF FIGURES................................................................................................................ .........7 ABSTRACT....................................................................................................................... ..............9 CHAPTER 1 INTRODUCTION................................................................................................................. .11 History, Etiology, and Strain Diversity...........................................................................11 Epidemiology..................................................................................................................1 3 Disease Control...............................................................................................................1 4 Bacteriocins................................................................................................................... ..15 The Type III Secretion System (T3SS)...........................................................................16 Biological Control...........................................................................................................17 2 CHARACTERIZATION OF GENETIC DETERMINANTS AND EVALUATION OF THEIR ROLE IN ANTAGONISM........................................................................................19 Materials and Methods.......................................................................................................... .22 Bacterial Strains, Plasmids and Culture Conditions........................................................22 DNA Manipulations........................................................................................................22 Construction of Bacteriocin Mutants...............................................................................23 Construction of T2SS Mutant..........................................................................................24 Bioinformatics Characterizati on of Bcn+ Cosmid Clones..............................................24 Subcloning of BcnB and BcnC........................................................................................24 Protease Activity Assay...................................................................................................25 BcnA Timing of Activation and Size Analysis...............................................................25 In vitro Antagonistic Assay.............................................................................................25 Evaluation of Immunity...................................................................................................27 In vitro assays. ..........................................................................................................27 In planta assays. .......................................................................................................27 Results........................................................................................................................ .............28 Sequence Analysis of Genes Involved in BcnA Activity................................................28 BcnA Activity Requires ORFA ORF2, ORF3 and ORF4..............................................29 Localization of BcnA Activation.....................................................................................29 Identification of the Immunity Gene...............................................................................29 Sequence Analysis of BcnB and BcnC...........................................................................30 Purification and Characteriz ation of BcnB and BcnC.....................................................31 Type II Secretion Mutant Lost Secr etion of Amylase and Bacteriocins.........................32

PAGE 4

4 Discussion..................................................................................................................... ..........32 3 ANALYSIS OF PATHOGENICITY MUTANTS OF XANTHOMONAS PERFORANS AND THEIR EFFECT ON BACTERIOCIN EXPRESSION................................................51 Materials and Methods.......................................................................................................... .54 Bacterial Strains, Plasmids and Culture Conditions........................................................54 Plant Material................................................................................................................. .54 Primer Design.................................................................................................................. 55 Generation of Mutants.....................................................................................................55 Growth Room Growth Curve Assays..............................................................................58 Greenhouse Disease Severity Assay...............................................................................59 Growth Room Antagonism Assay...................................................................................59 Results........................................................................................................................ .............61 Sequence Analysis of Attenuate d Mutant Candidate Genes...........................................61 Population Dynamics and Pathogenicity Assays............................................................62 Antagonism Assays.........................................................................................................63 Discussion..................................................................................................................... ..........63 4 EVALUATION OF XANTHOMONAS PERFORANS MUTANTS IN CONTROLLING X. EUVESICATORIA IN GREENHOUSE AND IN THE FIELD.........................................80 Materials and Methods.......................................................................................................... .83 Bacterial Strains, Plasmids and Culture Conditions........................................................83 Generation of the 91-118:: opgH bcnB Attenuation Mutant........................................84 Plant Materials................................................................................................................ .84 Growth Room Growth Curve Assays..............................................................................84 Greenhouse Disease Severity Assay...............................................................................85 Growth Room Antagonism Assay...................................................................................86 Internal antagonism assays. .....................................................................................86 Phyllosphere antagonism assays. .............................................................................86 Field Experiments............................................................................................................87 Field plot design. ......................................................................................................87 Bacterial strains, inoculum produc tion, inoculation and plant material. ................87 Incidence of strains in lesions. .................................................................................88 Incidence of phyllosphere populations. ....................................................................89 Results........................................................................................................................ .............89 91-118 and Mutants Reduce E3-1 in Gr owth Room and in the Greenhouse..................89 Field Study.................................................................................................................... ...90 Discussion...............................................................................................................................91 5 SUMMARY AND DISCUSSION.......................................................................................106 APPENDIX SEQUENCE AND ALIGNMENT OF BCNB AND PATHOGENICITY-RELATED GENES CHOSEN FOR DELETION...................................................................................110

PAGE 5

5 LIST OF REFERENCES............................................................................................................. 135 BIOGRAPHICAL SKETCH.......................................................................................................149

PAGE 6

6 LIST OF TABLES Table page 2-1. Bacterial strains and plas mids used in this study...................................................................35 2-2. PCR primer used in bcnA bcnB and bcnC analyses for genetic manipulation......................36 2-3. Characterization of bacterio cin ORFs associated with th e expression of BcnA, BcnB and BcnC....................................................................................................................... .....37 3-1. Bacterial strains and plas mids used in this study...................................................................68 3-2. PCR primers used in hpaA hpaB hpaC xopA xopD avrBs2 and gumD analyses for genetic manipulations........................................................................................................69 3-3. List of attenua tion candidate genes...................................................................................... .70 3-4. Homology of X. perforans genes............................................................................................71 3-5. In planta growth and aggressiveness of X. perforans strain 91-118 mutants........................72 3-6. Growth room in planta internal and phyllosphere antagonism experiments..........................73 4-1. Bacterial strains and plas mids used in this study...................................................................95 4-2. In planta growth and aggressiveness of X. perforans strain 91-118 mutants as...................96 4-3. Growth room in planta internal and phyllosphere antagonism experiments..........................97 4-4. Incidence and recovery of X. euvesicatoria strain E3-1 in the field.......................................98

PAGE 7

7 LIST OF FIGURES Figure page 2-1. BcnA diagram showing position of individual ORFs............................................................38 2-2. BcnB diagram showing position of individual ORFs............................................................39 2-3. BcnC diagram showing position of individual ORFs............................................................40 2-4. Antagonism assays of b acteriocin-like activity against X. euvesicatoria ...............................41 2-5. Antagonism activation assays of bacteriocin-like activity.....................................................42 2-6. Secretion assays of bacteriocin-like activity..........................................................................43 2-7. Immunity assays of b acteriocin-like activity..........................................................................44 2-8. In vitro populations of X. euvesicatoria strain 91-106 transconjugants................................45 2-9. In planta populations of X. euvesicatoria strain 91-106 transconjugants.............................46 2-10. Antagonism assays of bacteriocin-like activity....................................................................47 2-11. Protease activity of bacteriocin candidates..........................................................................48 2-12. Evaluation of type tw o secretion system mutant xpsD ........................................................49 2-13. BcnA Model for ORFs predic ted involved in BcnA activity...............................................50 3-1. Illustration of deletion constructions................................................................................... ..74 3-2. In planta growth of wild-type (wt) and mutant X. perforans strains.....................................75 3-3. Disease severity on Bonny Best l eaflets 2 weeks after dip inoculation................................76 3-4. Phenotype in leaves of Bs2 transg enic tomato VS36 and pepper (ECW-20R).....................77 3-5. Growth room internal antagonism assay................................................................................78 3-6. Growth room phyllosphere antagonism assay........................................................................79 4-1. In planta growth of wild -type and mutant X. perforans strains............................................99 4-2. Disease severity on Bonny Best l eaflets 2 weeks after dip inoculation..............................100 4-3. Growth room internal antagonism assay.............................................................................101 4-4. Growth room phyllosphere antagonism assay.....................................................................102

PAGE 8

8 4-5. Quincy 2004 field experiment............................................................................................. 103 4-6. Quincy 2005 field experiment............................................................................................. 104 4-7. Citra 2005 field experiment.............................................................................................. ...105 A-1. BcnB sequence. Nucleotide sequence of a 5968 bp Kpn I and Eco RI fragment containing bcnB ...............................................................................................................110 A-2. Nucleotide sequence of a 980 bp fragment containing the hpaA ORF...............................114 A-3. Nucleotide sequence of a 2345 bp fragment containing the hpaB ORF.............................115 A-4. Nucleotide sequence of a 1419 bp fragment containing the hpaC ORF............................117 A-5. Nucleotide sequence of a 1716 bp fragment containing the xopA ORF.............................119 A-6. Nucleotide sequence of a 2173 bp fragment containing the xopD ORF............................121 A-7. Nucleotide sequence of a 2735 bp fragment containing the avrBs2 ORF...........................124 A-8. Similarity of amino acid sequence of HpaA.......................................................................127 A-9. Similarity of amino acid sequence of HpaB........................................................................128 A-10. Similarity of amino acid sequence of HpaC......................................................................129 A-11. Similarity of amino acid sequence of AvrBs2...................................................................130 A-12. Similarity of amino acid sequence of XopA......................................................................132 A-13. Similarity of amino acid sequence of XopD......................................................................133

PAGE 9

9 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EVALUATION OF BACTERIOCINS IN Xanthomonas perforans FOR USE IN BIOLOGICAL CONTROL OF Xanthomonas euvesicatoria By Aaron Paul Hert May 2007 Chair: Jeffrey B. Jones Cochair: M. Timur Momol Major: Plant Pathology Xanthomonas perforans strain 91-118 produces at least three different bacteriocin-like compounds (BcnA, BcnB, Bc nC) antagonistic toward X. euvesicatoria strains. Previous research showed that deletion of one bacteriocin (BcnB) produced the highest leve l of antagonism toward sensitive X. euvesicatoria strains. One aspect for this st udy was to further characterize each bacteriocin by deletion mutagene sis to establish which open r eading frames (ORFs) were responsible for bacteriocin activity for each bact eriocin as well as determining their possible functions. BcnA has been shown to contain at least four essen tial genes for activity and a model has been created to suggest the role of each gene. BcnB and BcnC were both found to be proteinases (endoproteina se Arg-C and extracellular meta lloproteinase, respectively). A second aspect of this study was to deve lop a viable biocontro l strategy by creating pathogenicity-attenuated mutants such that thes e attenuated mutants on the plant surface would suppress bacteriocin-sensitive strains. Several candidate genes were chosen based on mutant phenotypes in either X perforans (OpgHXcv) or the closely related X euvesicatoria strain 85-10 (hpaA, hpaB, hpaC, xopA, xopD, avrBs2 and gumD) Each candidate gene was amplified and PCR-assisted deletion mutagenesis was performed for final marker exchange into wild-type (wt)

PAGE 10

10 X perforans to create attenuation mutants. Each mu tant was tested for growth rate, disease severity and antagonism toward X. euvesicatoria -sensitive strains. Mutations in opgH and gumD gave the most significant reducti on in disease and growth rate wh ile maintaining the ability to reduce X. euvesicatoria populations. One attenuated mutant, 91-118:: opgH was chosen for further investigation. Greenhouse and field experiments were conducted using 91-118:: opgH bcnB to determine its ability to reduce X. euvesicatoria populations. Greenhouse a nd field experiments indicate 91-118:: opgH bcnB significantly reduced X. euvesicatoria populations. In the field, weekly application of 91-118:: opgH bcnB consistently reduced X. euvesicatoria populations as compared to the standard c ontrol (application of copper + manzate and actigard every two weeks). 91-118:: opgH bcnB applied every two weeks also significantly reduced X. euvesicatoria populations in one season, but were not si gnificantly different from the grower standard control.

PAGE 11

11 CHAPTER 1 INTRODUCTION Bacterial spot of tomatoes and peppers is caused by the phytopathogenic bacterium Xanthomonas euvesicatoria X. vesicatoria and X. perforans (69). It is a worldwide disease and is a major problem in Florida, particularly during periods of high temperatures and high humidity. Pohronezny et al. (115) estimated as high as 50% loss of marketable fruit due to bacterial spot on tomatoes. The pathogen is a Gram-negative, rod-shaped bacterium which can readily be isolated from diseased tissue. It is motile, possessing a single polar flagellum, strictly aerobic, and measures 0.7 to 1.0 m by 2.0-2.4 m. On nutrient agar the bacterium produces a characteristic yellow water-insoluble pigment called xanthomonadin and an extracellula r polysaccharide (EPS) termed xanthan gum (149). Bacterial spot of tomato affects the aerial por tions of the plant, with symptoms consisting of numerous small (1 to 5 mm) ci rcular lesions on leaves, stems a nd fruit. Bacterial spot can be distinguished from fungal leaf spots by a greasy, water-soaked appearance on the abaxial side of leaves. Chlorosis and epinasty of leaves occurs eventually leading to complete necrosis of tomato leaflets (140). History, Etiology, and Strain Diversity Bacterial spot of tomato and pepper is one of the earliest recorded bacterial diseases. X. euvesicatoria was first described as bacterial canker in South Africa by Doidge in 1921 (33). That same year Gardner and Kendrick (52) in the United States discovered a similar organism and referred to it as bacterial spot. Doidge performed a comprehensiv e study on the etiology of bacterial spot, which she termed tomato ca nker, and identified the causal agent as Bacterium vesicatorium whereas in the United States, Gardner and Kendrick (52) originally named the

PAGE 12

12 organism Bacterium exitiosa Over time, as a result of comprehensive studies (51), the two bacteria were discovered to be very si milar and, in the mid 1920s, were designated Bacterium vesicatoria Since that time the bacterium was rena med several times, changing genera from Bacterium to Pseudomonas to Phytomonas and finally to Xanthomonas (71). Once transferred to Xanthomonas the bacterium underwent several species and pathovar changes. In 1980 it was transferred to X. campestris pv. vesicatoria ( X c vesicatoria ) (35). In the 1990s, major changes occurred in the taxonomy of the genus Xanthomonas which resulted in the renaming of many species. Extensive comp arison of strains using DNA-DNA hybridizations resulted in the identification of two groups, A and B (140, 157). Group A (tomato race 1) strains were transferred to X axonopodis and designated X axonopodis pv. vesicatoria while Group B strains (tomato race 2) were placed in X vesicatoria (157). In the 1990s a new group of strains was identified in Florida (72) that was phylogene tically most closely related to group A, but was phenotypically and genotypically distinct enough from group A th at it was designated as group C. Because group C is most closely relate d to group A based on DNA-DNA hybridization this new group was designated within X. axonopodis pv. vesicatoria In 2004, the most recent changes occurred in nomenclature (69). Group A (tomato race 1) and Group C (tomato race 3) strains were removed from X. axonopodis since they shared less than 70% DNA relatedness to other X. axonopodis strains, and were placed in X. euvesicatoria and X. perforans respectively. Several avirulence genes have been characterize d in xanthomonads associated with tomato. In 1993, Whalen et al. (166) found X. euvesicatoria, tomato race 1 (T1), st rains to carry the avirulence gene avrRxv, which induces an incompatible react ion that activates localized cell death also known as a hypersensitive res ponse (HR) on the genotype H7998 carrying the corresponding resistance gene Rxv ; X. perforans tomato race 3 (T3), strains were determined to

PAGE 13

13 carry avrXv3, which induces an HR in H7981 that contains the resistance gene Xv3 (4, 100). In 2000, a new avirulence gene avrXv4 was described in X. perforans strains based on reactions on the tomato genotype, LA716 ( Lycopersicum pinnellii ) which carries the Xv4 resistance gene (3, 4). Therefore, X. perforans strains carrying this new avirulence gene ( avrXv4 ) but lacking a functional avrXv3 have been designated as tomato race 4. Epidemiology Leben (84) introduced the concept of a resi dent phase where the pathogen is able to replicate in the phyllosphere (leaf surf ace) without causing visible symptoms. Xanthomonas euvesicatoria has been shown to colonize pepper and to mato leaves epiphytically (93, 137). Long-term survival occurs in crop residue and volunteer plants (70, 140) Seed contamination was proposed as an important mechanism for transmission by Bashan et al. (6); however, Jones et al (70) concluded that survival on seed occurs at extremel y low levels and may be less important in the epidemiology of the disease than other inoculum sources when the pathogen is endemic. In soils artificially infested with X.c. vesicatoria survival is poor with the bacterium being detected for only 16 days (6). The bacterium gains entry into the plant wh en: (I) conditions are favorable for disease development (57), or (II) a threshold of epiphytic populations is reached (94) or (III) if a plant is compromised by wounding (156). Xanthomonas euvesicatoria enters the plant in many ways. Routine farming operations damage the plant cau sing wounds that act as entry points (6, 116). Epidermal abrasions, leaf hair breakage and water congestion of the interce llular spaces increase entry of the bacterium up to 100-fold over healt hy plants (156). The ba cterium can also enter through natural openings such as stomates and hydathodes (96, 123, 145). High humidity is conducive to bacterial ingress and survival; high relative humidity has been shown to increase

PAGE 14

14 infection of X. euvesicatoria by 10to 100-fold on tomato leaves compared to low humidity (151). Dissemination of the bacterium is a major fact or in the epidemiology of the disease. Overhead irrigation enhances dissemination compar ed to furrow irrigation (160). Dissemination also occurs in the form of aerosols and wind-blow n rain (95, 162). Infected seed serves as a mechanism for dissemination (116). Farming practi ces, such as thinning, tying, and mechanized spraying also serve as fact ors in dissemination (116). Disease Control Bacterial spot of tomato is difficult to control when high te mperatures and high moisture exist. Bactericides, such as fixed coppers a nd streptomycin, have provided the major means of control (90, 142). Streptomycin-resi stant mutants were rapidly sel ected on streptomycin-treated plants (142). As a result of rapid selection for streptom ycin-resistant mutants, copper compounds have been used almost exclusively. However, Marc o and Stall in the 1980s (90) showed that many X. euvesicatoria strains were tolerant to copper and determined that c opper resistance is mediated by genes located on a self-transmissible plasmi d (143). Adding mancozeb, a fungicide, to copper sprays was shown to improve control effi ciency (25) and was s hown by Marco and Stall (90) to control copper-tolerant strains. Howeve r, they also showed th at this treatment is insufficient when conditions favorable for disease development exist. Because of the presence of copper-tolerant stra ins, other control strategies need to be considered. Identification of resistance genes and introgressio n into commercial genotypes has been a focus of breeding programs (135). There are currently no commercially available tomato varieties resistant to all r aces of bacterial spot.

PAGE 15

15 Bacteriocins Bacteriocins are proteins or peptides with antibacterial properties, in most instances targeting related bacteria belonging to the same species or genus (135). Bacteriocins of Grampositive bacteria, such as lactobacilli, typically are small peptides (40, 92). Bacteriocins from Gram-negative bacteria are often larger proteins among which the colicins from Escherichia coli represent the best-known examples (14, 66, 83, 91). There is considerable structural diversity among them which is reflected in widely di fferent modes of action, including membrane disruption, non-specific degradati on of nucleic acids and inhibition of peptidoglycan synthesis, and proteases. Reports on the production of bacteriocinlike compounds by phytopathogenic bacteria are scarce. The observed report of a phytopat hogenic bacterium produc ing bacteriocin-like compounds was reported by Okabe in the early 1950s (110). He reported that strains of Pseudomonas ( Ralstonia ) solanacearum were inhibitory ex clusively to other P. solanacearum strains. Since then, these t ypes of compounds have been repo rted for several other genera: Agrobacterium (77), Clavibacter (36), Erwinia (22, 37, 80) and Pseudomonas (30, 48, 82, 139, 152). There have been reports of xanthomonads producing bacteriocins as well (45, 155, 169). In 1991, X. perforans were first identified in Florida. In fields where both X. euvesicatoria and X. perforans were present, the X. perforans strains became predominant (72). In vitro assays have shown that X. perforans strains inhibit growth of X. euvesicatoria strains (39). Jones et al. (71) characterized this rela tionship in greenhouse experiment s tested on three genotypes including a T3 resistant genot ype. Under field conditions X. perforans strains had a competitive advantage over X. euvesicatoria strains (71, 117). Tudor (153 ) identified at least three antagonistic compounds in X. perforans strains that closely resembled bacteriocins. These compounds were determined to have na rrow inhibition spectra (restricted to Xanthomonas

PAGE 16

16 strains) and fit the definition of a bacteriocin described by Reeves (125) based on the following criteria: (I) the presence of a biologically active protein moiety, (II) inducibility with mitomycin C and (III) non-self inhibition. All three bacter iocin-like groups (BcnA, BcnB and BcnC) were unique in activity and specificity against X. euvesicatoria strains (155). The Type III Secretion System (T3SS) Pathogenicity of bacterial spot of tomato is determined by the type III secretion system (T3SS), which is highly conser ved in most Gram-negative bact erial pathogens of plants and animals (149). The T3SS is composed of a secre tion apparatus and an array of diverse proteins, known as effectors, that are in jected into plant cells via th e secretion apparatus (103). The T3SS is encoded by a gene cluster termed the hrp (h ypersensitive r esponse and p athogenicity) cluster (1). One group of genes located within the hrp cluster encodes for a secretion apparatus known as the hrp pilus. The hrp pilus serves as a secretion apparatus for the translocation of T3SS effector proteins (59, 131) Hrp pilus mutants no longer cause disease in susceptible plants and are unable to induce resi stance in resistant host and non-host plants (12, 16). Inside the host cell, type III effectors have specific functions and interact with specific targets in the host (17, 58, 63, 106); however, the functions of many effectors are unknown and their deletion produces no detectible phenotype. Some plants have developed resistance to th ese invading pathogenic bacteria via resistance genes (R genes) which recognize sp ecific effector proteins called avirulence (avr) genes. When the pathogenic bacterium injects an avirulen ce gene into a resistant plant carrying the corresponding R gene, an incompatible reaction, or HR, occurs which localizes the invading bacteria and limits secondary inf ection of surrounding cells (103).

PAGE 17

17 Biological Control Biological control is another important approa ch for control of the disease. Pathogen resistance to fungicides has prom pted interest in development of biocontrol agents (19), to provide additional tools for disease management. Unlike biocontrol of in sects, biocontrol of plant diseases is a relatively new field. In the last 25 years biocontrol ha s become an established sub-discipline in plant pathology (113). Agrobacterium radiobacter strain K84, registered with the United States Environmental Protection Ag ency (EPA) for control of crown gall in 1979 (EPA registration number 11,4201), and was the firs t commercially availa ble biological control agent against a bacterial plant disease (77). Sinc e then, a total of 14 bacteria and 12 fungi have been registered with EPA for control of plant diseases (47, 112) Several promising biological c ontrol approaches that include antagonistic microorganisms, natural fungicides and induced re sistance are available for use in disease control today (38). However, achieving success using biocontrol agen ts for many bacterial diseases has been difficult. Some success has been achieved in this area through empirical selection of biocontrol agents, as indicated by the commercializati on of the products Agriphage (a mixture of bacteriophages for control of bacterial spot of to mato (46)), Galltrol, for control of crown gall, and BlightBan A506, for control of fire blight a nd frost injury (86). For bacterial spot of tomato, field experiments have been conducted utilizing a non-pathogeni c bacteriocin-producing X. perforans strain to control disease incited by X. euvesicatoria strains (88). The nonpathogenic strain was able to reduce X. euvesicatoriaincited disease incidence and severity when applied prophylactically; however, the dise ase was still above acceptable levels (forty percent) (88). For Ralstonia solanacearum efforts to obtain a biological control strategy utilizing bacteriocin-producing non-pathogenic hrpmutants gave low to moderate levels of control of

PAGE 18

18 wildtype (wt) R. solanacearum (152). However, control using a partially pathogenic hrp mutant ( hrcV-), which is capable of higher levels of coloni zation of the root and stem tissue, achieved better control levels (49). Res earch into colonization has b een conducted to understand the possible relationship between inva sion efficiency of the biocont rol agent and its ability for disease control. Etchebar et al. (41) suggested that there wa s a positive correlation between colonization of the xylem by the hrp mutant and the level of control of the wt R. solanacearum The goal of this study was to evaluate a new biological contro l strategy utilizing pathogenicity-attenuated, bacteriocin-producing X. perforans strains for control of bacteriocinsensitive strains of X. euvesicatoria The objectives of this study we re: (I) to further characterize the bacteriocins associated with X. perforans (II) to identify and indivi dually delete genes that contribute to pathogenicity to create less virulent mutants of X. perforans and (III) to determine the ability of these pathogenicity-at tenuated mutant strains to antagonize X. euvesicatoria in vitro in planta and under field conditions.

PAGE 19

19 CHAPTER 2 CHARACTERIZATION OF GENETIC DETE RMINANTS AND EVALUATION OF THEIR ROLE IN ANTAGONISM Bacterial spot of tomato is incited by four Xanthomonas species: X. euvesicatoria, X. vesicatoria X. perforans and X. gardneri The first three bacterial species were previously known as tomato races 1 (T1), 2 (T2) and 3 (T3) respectively, based on their reaction on three tomato genotypes: Hawaii 7998 (H7998), Hawa ii 7981 (H7981) and Bonny Best (71, 72, 139). X. gardneri has only been found in former Yugosl avia, Costa Rica and Brazil (10, 123, 143) In 1991, X. perforans was first identified in Florid a (72). In fields where both X. euvesicatoria and X. perforans were present, X. perforans became predominant (72). This phenomenon was due to bact eriocin-like activity of X. perforans strains (152). Bacteriocins are substances, usually proteinaceous, that are inhibito ry or harmful toward closely related bacteria (124). Bacteriocins of Gram-negative bacteria re present a diverse group of proteins in terms of size, microbial target, mode of action and immunity mechanism. They are high molecular weight proteins that gain entry into susceptibl e cells by binding to surface receptors. Their mode of action varies from degradation of cellula r DNA, to disruption of cleavage of 16S RNA, inhibition of synthesis of the peptidoglycan a nd pore formation in the cytoplasmic membrane (26). The most extensively studied bacter iocins are the colicins produced by Escherichia coli (14, 66, 83, 91, 120, 121, 126, 163). A model system has been developed for proteinaceous bacteriocin production consisting of three components: the toxi n, the immunity gene and a mechanism for delivery (126). Several known bacter iocins are transcribed in an inactive form (pre-bacteriocin), which, upon secretion, is proce ssed to its active form (eg. Colicin V; (169)) (111).

PAGE 20

20 Reports on the production of bacteriocinlike compounds by phytopathogenic bacteria are limited. In the 1950s Okabe et al. (110) published the first artic le on phytopathogenic bacteria, where strains of Pseudomonas ( Ralstonia ) solanacearum were inhibitory only to other P. solanacearum strains. Since the original descript ion, production of such compounds has been reported for several other genera: Agrobacterium (77), Clavibacter (36), Erwinia (22, 38, 80) Pseudomonas (30, 82, 138, 151) and Ralstonia (41, 48). There have been a few reports of xanthomonads producing bacteriocins (45, 154, 168). In order to further characterize the bacteriocin-like activity of X. perforans a genomic library was screened to localize bacteriocin ac tivity (154). Three groups of clones were identified that showed unique bacteriocin activ ity and all three bacter iocin-like groups (BcnA, BcnB and BcnC) were unique in activity and specificity based on X. euvesicatoriasensitive strains (154). None of the clones conferred immunity to the other bacteriocins. BcnA was localized to an 8.0-kb fragment containing seven open r eading frames (ORF) identified in the sequenced region. The larg est ORF (ORFA), approximately 3.6-kb, is required for BcnA+ activity. The ORFA protein contains 1012 amino acids with a theoretical molecular weight of approximately 111-kDa. BcnA+ activity was detected in unconcentrated, cell-free extracts of strains expressing ORFA. In some bacteria, an immunity function is necessary in order to avoid self-inhibition of the producing strain (111, 169) The putative immunity function of BcnA was mapped to a 4.5-kb Bam HI/ Eco RI fragment downstream of ORFA (154). Southern hybridization analysis using an ORFA-specific probe indicated that among bacterial spot strains tested, only X. perforans strains hybridized. Hybridization of the probe to a chromosomal location suggests that BcnA+ is in the chromosomal DNA. Homology searches using the deduced amino acid sequence of the ORF revealed significan t homology to only two

PAGE 21

21 known proteins, WapA and Rhs. Both of these pr oteins contain multiple copies of an almost identical ligand-binding motif, t hought to be involved in carbohydr ate binding. Seven copies of a similar motif were found in ORFA. Xanthomonas campestris pv. glycines ( X.c. glycines ) is one of the few xanthomonads that produces multiple bacteriocin-like compounds with activity against selected xanthomonads (45). According to Tudor (153) at least one of the X.c. glycines bacteriocin-like compounds is very similar in activity to BcnA. In X.c. glycines bacteriocin-like compounds were heat sensitive and trypsin resistant (45), suggestiv e of the involvement of a high molecular weight protein. Relatively little is known about BcnB and Bc nC. BcnB and BcnC were previously subcloned to 8.9-kb and 5.1-kb fragments, respectiv ely (154). Both were sequenced (60, 154). No immunity factor was associated with Bc nB or BcnC activity. It is unknown how a heterologous strain that expresses either bacteriocin is not inhibited. Although the exact ORF involved in BcnC expression wa s not identified, one ORF with in this fragment showed significant homology to extracellu lar metalloproteases secreted by Aeromonas hydrophila Armillaria mellea Pleurotus ostreatus Grifola frondosa Aspergillus fumigatus and Penicillum citrinum (153). Enzymes produced by bacteria may mimic the action of bacteriocins. The zooA gene of Streptococcus zooepidemicus which encodes a bacteriocin-like inhibitory substance, contains a region with significant homol ogy to several known endopeptidases (137). It was reported that some pr oteinases and bacteriocins ar e secreted via the type two secretion pathway (T2SS) (2, 149, 160). In order to determine the involv ement of the T2SS in X. perforans bacteriocin production, disruption mutants were created in the clos ely related bacteria, X. euvesicatoria then transformed with a cosmid expres sing either BcnA, BcnB, or BcnC. The T2SS is composed of 12 proteins (Xps) for transl ocating extracellular pr oteins across the outer

PAGE 22

22 membrane in Xanthomonas campestris pv. campestris ( X.c. campestris ) (23). X.c. campestris secretes several hydrolytic enzymes, including -amylase, protease, pectate lyase and cellulase by the type 2 secretion pathway (23). The Xp sD T2SS protein, an outer membrane lipoprotein, is required for protein secr etion via the T2SS (23). An XpsD mutant from Xanthomonas oryzae pv. oryzae also lost T2SS function (50). Thus, xpsD was chosen for disruption to create a T2SS mutant. In this study we further ch aracterized three bacteriocins of X perforans to determine their activity, delivery mechanisms and immunity. The goals of this study were (I) to further ch aracterize the role of each bacteriocin-like compound secreted by X. perforans ; (II) to determine the de livery mechanism of each bacteriocin and; (III) to understand the po ssible functions of each bacteriocin. Materials and Methods Bacterial Strains, Plasmids and Culture Conditions Strains of X. euvesicatoria and X. perforans (Table 2-1) were grown on nutrient agar (NA) medium (Difco Laborat ories, Detroit, MI) at 28 C. Strains of Escherichia coli (Table 2-1) were grown on Luria-Bertani (LB) medium at 37 C (97). All strains were stored in 20% glycerol in sterile tap water at C. Antibiotics were used to maintain selection for resistance markers at the following concen trations: tetrac ycline (Tc) 12.5 g/mL; rifampicin (Rif) 100 g/mL; spectinomycin (Sp) 50 g/mL; kanamycin (Km) 25 g/mL; chloramphenicol (Cm) 34 g/mL; streptomycin (Sm) 200 g/mL and nalidixic acid (Nal) 50 g/mL. DNA Manipulations Standard techniques for molecular cloni ng were conducted as described by Sambrook et al (133). Restriction endonucleas e digestions were performed according to manufacturers specifications. All enzymes were obtained from Promega (Madison, WI) or Biolab (Ipswich,

PAGE 23

23 MA). All DNA extractions were done as described by Sambrook et al. (135) T4 DNA ligase (M180A) was used according to manufacturers sp ecifications (Promega). Constructs were transformed into competent Escherichia coli DH5 cells prepared as described in Sambrook et al. (135) and stored at C until transformations. Construction of Bacteriocin Mutants According to an ORF search, there are 5 ge nes in the BcnA fragment, designated ORFA, ORF2, ORF3, ORF4, and ORF5. To characterize the function of BcnA genes ORFA, ORF2, and ORF3 were disrupted either by deleti on or transposon mutagenesis to create 91-118:: ORFA, 91-118:: ORF2 and 91-118:: ORF3. ORF4 was previously disrupted (153). The 91-118:: ORFA mutant was cons tructed by deleting an EcoR V and Bgl II fragment. ORF2 and ORF3 mutants were created by using surr ounding sequences up a nd downstream of the target ORFA. For ORFA, PCR wa s performed with primers A5 and A3 (Table 2-2), then the resulting PCR product was inserted upstream of ORF3 subcloned in the phagemid vector pBluescript II KS (pBS) (Stratagene, La Jolla, CA) (Figure 2-1a). Final ORF2 was moved into suicide vector pOK1 us ing restriction enzymes Apa I and Spe I (Figure 2-1b). To make 91-118:: ORF3, PCR was performed with primers OR F2F and ORF3R, then the resulting PCR product was inserted upstream of OR F4 in pBS:ORF4 to create pBS: ORF3. The fragment containing ORF3 was subcloned into suicide vector pOK1 with Apa I and Sal I creating pOK1: ORF3. The final plasmid constructs were mated into 91-118 to make each mutant via suicide vector-assisted mutagenesis as describe d previously (74). Candidates were screened using PCR primers designed to amplify flanking regions of the cross-over region (Table 2-2). Each mutant was tested for bacteriocin activity against X. euvesicatoria strain 91-106.

PAGE 24

24 BcnB was disrupted by adding an insertion st op codon (TAA) using Quick-change XL Site Directed Mutagenesis kit (Stratagene). Construction of T2SS Mutant Mutations were created to determine if BcnA, B and C are secreted by T2SS. In order to clone xpsD gene, primers xpsDF and xpsDR were de signed using 85-10 genome sequence (149). A 2,229 bp xpsD gene was amplified and subcloned into pGEM vector (Promega). A chloramphenicol-resistance cassette fr om pRKP10 (123) was inserted in a kpn I site to disrupt xpsD This disrupted xpsD gene was subcloned into suicide vector pOK1 with Apa I and Xba I. The final plasmid constructs were mated into 91-106 to make 91-106:: xpsD via suicide vectorassisted mutagenesis as describe d previously (74). Candidates we re screened using PCR primers designed to amplify flanking regions of the cross-over region (Table 2-2). Bioinformatics Characterizati on of Bcn+ Cosmid Clones Each ORF was analyzed for sequence homology (BlastP, http://www.ncbi.nlm.nih.gov/BLAST/ ), signal peptides and hydrophobicity (SOSUI, http://sosui.proteome.bio.tuat.ac.jp/sosuiframe0.html ), and localization (PSORT, http://www.psort.org/ ). Subcloning of BcnB and BcnC The BcnB+ clone pXV6.0 (6.0-kb BcnB+ fragment) was subcloned using a Kpn I/ Eco RI fragment from pXV442 (BcnB+). The 6.0 kb fragment was sequenced (Appendix A-1). For sequence analysis (Figure 2-2), the Kpn I/ Eco RI region was cloned into pBS using Kpn I/ Eco RI enzymes and was sequenced using T7 and SP6 primers from pGEM and custom designed oligonucleotides generated by the DNA Sequencing Core Laboratory of the Interdisciplinary Center for Biotechnology Research, University of Florida, Gainesville, FL (UF-ICBR). DNA sequencing was performed by the DNA Sequenci ng Core Laboratory of UF-ICBR using the

PAGE 25

25 Applied Biosystems model 373 system (Foster Cit y, CA). Further subcloning of this fragment was performed using PCR primers (Table 2-2). The BcnC pXV1.7 clone (1.7-kb BcnC+ Sal I/ Eco RI fragment) was directionally cloned for expression in the plus and minus direction for expression of an extracellular metalloprotease (plus direction) or a glycine-rich pr otein (minus direction) (Figure 2-3). Protease Activity Assay Proteolytic activity was measured by a diffusion assay in agar plates containing skim milk (casein) as substrate as described previously (3 4). Five microliters of each bacterial suspension were applied onto the surface of plates contai ning 20 mL of 0.5% (wt/vol) skim milk, 2% (wt/vol) agar and 50mM Tris hydrochloride (pH 8.0) and allowed to incubate for 24 h at 28 C. Zones of clearing around the bacteria due to the degradation of the substrate were measured. BcnA Timing of Activati on and Size Analysis Bacteriocin activity was assessed to determ ine which fraction(s) contain active BcnA protein. Supernatants and cells were collected from 18 h nutrient broth (NB) cultures. Cells were collected via centrifugation, suspended w ith phosphate buffered saline (PBS) (135) and sonicated using a dig ital Sonifier unit model S-150D (B ranson Ultrasonics Corporation, Danbury, CT). Fractions were sonicated for 30 s two times on ice. Supernatant and sonicated cell fractions were assessed for bacterioci n activity by plate assay described below. Once fractions were prepared, size analysis was performed by sepa rating cell fractions by Microcon protein filtration system (Millipore, Bill erica, MA) with filter cut-offs of 50 kDA (YM-50) and 100 kDa (YM-100). In vitro Antagonistic Assay Each mutant was evaluated for its relative bacteriocin activity produced toward a sensitive X. euvesicatoria strain based on an in vitro zone of inhibition assay (153). Strains to be tested

PAGE 26

26 were shaken at 28 C overnight in NB. The cells were pe lleted and resuspended in sterile tap water. Resuspended cells we re then standardized to A600= 0.3 which is approximately 5 108 CFU/mL. A 25 l sample was spotted on a Petri plate (100 15mm) containing 20 mL NA (five samples per plate) and grown for 18 h at 28 C. After 18 h a suspension (5 107 CFU/mL) of X. euvesicatoria strain 91-106 (sensitive indi cator strain) was sprayed ove r the plate using a Sigma aerosol spray unit (Sigma Chemical, St. Loui s, MO). After 24-48 h incubation, zones of inhibition around the test st rain were measured. A second technique involved incubation of the plates for 24 h, killing the test strains by inverting glass plates over 2-3 mL of chlorofo rm until all of the chloroform was evaporated, aerating the plates for 1 h and overlaying the ag ar surface with 3.5 mL of 0.3% water agar (50 C) which contained 200 l of a 5 107 CFU/mL cell suspension of the indicator strain ( X. euvesicatoria strain 91-106). A clear zone of inhib ition around test coloni es after 24 48 h was considered indicative of antagonism and scored as bacteriocin-like (BcnA+) activity. Cell-free extracts were screened for BcnA activ ity by growing the test cultures for 18 h in NB followed by centrifugation to pellet cells. The supernatant was then sterilized using a low protein binding Microcon filter (Amicon, Beverly, MA) with a 0.22 m pore size and analyzed for antagonism by the well diffusion assay method (146). Five millimeter diameter wells were cut into 20 mL NA plates. Wells were filled with 100 L of test filtrates and left for 18 h to allow diffusion of the liquid into the medium. Plates were then overlaid with 3.5 mL of soft agar containing 200 l of 5 108 CFU/mL cell suspension of the in dicator strain 91-106. Plates were examined after 24 h at 28 C. Each test was replicated three times.

PAGE 27

27 Evaluation of Immunity In vitro assays. An immunity in vitro assay was conducted to determine the ORF responsible for BcnA immunit y. All ORFs downstream of OR FA, individually (ORF2, ORF3, ORF4, ORF5) and in combination (ORF2 to 3 a nd ORF2 to 5). were subcloned into pLAFR119 and then mated into 91-106 to create 91-106 + pL:ORFA, 91-106 + pL:ORF2, 91-106 + pL:ORF3, 91-106 + pL:ORF4, and 91-118 + pL:ORF5. Next they were evaluated for immunity to 91-118:: bcnBC which expresses BcnA. The te st strains were shaken at 28 C overnight in NB. The cells were pelleted and resuspended in sterile tap water. Antagonism assays were conducted using 91-118:: bcnBC as the producing strain and the mutants with deletion in all ORFs downstream of ORFA as the test strain. ORF5 was further tested to confirm its im munity. 91-106 containing either pLAFR119 (pL) or pLAFR119:ORF5 (pL:ORF5) were evaluated for sensitivity to 91-118:: bcnBC (Table 2-1). Strains were shaken at 28 C overnight in NB tubes for 18 h. Cells were then washed, resuspended in sterile tap wate r and standardized to produce 5 x 106 CFU/mL of 91118:: bcnBC then incubated at 28 C. After 6 h incubation, 5 105 CFU/mL of 91-106 strains (with pL or pL:ORF5) were added to the flasks. Samples were assayed at 24 h intervals for 96 h. Each experiment was conducted three times. Popul ation data were transformed to logarithmic values and standard errors were determined. In planta assays. The X. perforans and sensitive test X. euvesicatoria and transconjgant strains (91-106 + pL, 91-106 + pL:ORF2, 91106 + pL:ORF3, 91-106 + pL:ORF4 and 91-106 + pL:ORF5) were inoculated at 5 x 107 CFU/mL and 5 x 106 CFU/mL, respectively. The 91-118 strains were inoculated into leaflets by infiltra tion 18 h prior to inoculation with the sensitive strain. Six-week-old seedlings of the tomato cultigen Florida 47 were inoculated (15 leaflets

PAGE 28

28 each plant) using a hypodermic syringe as desc ribed previously (68). Following inoculation, plants were incubated at 24 C to 28 C. In order to determine popul ations of the sensitive test strain and transconjugants in leaflets, 1-cm2 leaf disks were remove d from inoculated areas, macerated in 1 mL sterile tap water and diluti on plated onto NA amende d with the appropriate antibiotic. Samples were assayed at 24 h intervals from 48 to 96 h. Each experiment was conducted three times. Populati on data were transformed to logarithmic values and standard errors were determined. Results Sequence Analysis of Genes Involved in BcnA Activity Previously a 12.1 kb fragment (pXV12.1) was shown to contain five ORFs (ORFA, ORF2, ORF3, ORF4 and ORF5) potentially important for expression of and for immunity to BcnA (56). Each putative ORF product was evaluated for pr esence of a signal peptide and localization (Table 2-3). Based upon sequence analysis ORFA is predicted to be a water soluble protein, with a hydrophobicity value of -0.56. It has a pred icted location in the ba cterial cytoplasm (0.56) with no predicted signal peptides. The putative ORF2 product has an N-terminal signal peptide, a hydrophobicity value of 0.28 and is predicted to be localized to the bacterial outer membrane (0.926) or the bacterial periplasmic space (0.175). The ORF3 product has an N-terminal signal peptide, a hydrophobicity value of 0.20, and an es timated localization to either the bacterial periplasmic space (0.939) or the bacterial outer membrane (0.326). The ORF4 has an N-terminal signal peptide, a hydrophobicity value of -0.36, and two transmembrane helices, from AA148 to 170 (VTAVAPPPTPTFQPAILTLGAVL) and from AA 176 to 198 (PAAVSWVSPIMGSIVLAPVLYFA). The ORF4 produc t is predicted to be located in the bacterial inner membrane (0.187). The ORF5 pr oduct has no signal peptide, a hydrophobicity

PAGE 29

29 value of 0.166, and is predicted to be a water so luble protein located in the bacterial inner membrane (0.109). BcnA Activity Requires OR FA, ORF2, ORF3 and ORF4 To further analyze the function of BcnA ORFA, ORF2, ORF3 and ORF4 were individually disrupted (91-118:: ORFA, 91-118:: ORF2, 91-118:: ORF3, 91-118:: ORF4). Each mutant was tested for bacteriocin activity against X. euvesicatoria strain 91-106. Three mutants, 91-118:: ORFA, 91-118:: ORF2 and 91-118:: ORF4, lost inhibition activity against the T1 strain (Figure 2-4). 91-118:: ORF3 had inhibition activity, but it was reduced compared to 91-118. Localization of BcnA Activation Bacteriocins are produced either in an active or inactive (pre-bacteriocin) form, which is activated during its secretion. In order to de termine the location of Bc nA activation, different cell fractions of a 24 h broth culture of 91-118 were tested for bacterioci n activity. Bacteriocin activity was only found in the supern atant (Figure 2-5). No activity was observed in supernatant from the less than 50 kDa fraction; however, supernatant from 50 to 100 kDa and above 100 kDa had inhibitory activity (Figure 2-6) confirming previous results (154 ). In order to determine if bacteriocin activity was associated with the cell fraction of 91-118, cells were disrupted by sonication, intact cells removed by filtration through a 0.22 m f ilter and then the bacteriocin activity was checked by plate assay. The cell fr action did not have activity (Figure 2-5). Identification of the Immunity Gene A 4.5-kb fragment downstream of ORFA was pr eviously found to contain the immunity gene (154). This fragment cont ains ORF2, ORF3, ORF4 and ORF5 (Figure 2-1). In order to identify which gene was responsible for immunity X. euvesicatoria strains were created that expressed each gene under a l ac promoter in pLAFR119 (Table 2-1). The positive control

PAGE 30

30 91-106 + pXV12.1 and 91-106 + pL:ORF5 were not sensitive to 91-118:: bcnBC while 91-106 + pL:ORF2, 91-106 + pL:ORF3 and 91-106 + pL:O RF4 were sensitive (Figure 2-7). In an in vivo experiment, bacterial populations of 91-106 + pL:ORF5 and 91-106 + pLAFR119 (empty vector) strain s co-inoculated with 91-118:: bcnBC were determined. 91-106 + pL:ORF5 reached concentrations of 105 to 106 CFU/cm2 tomato leaf tissue in the presence of BcnA, whereas 91-106 + pLAFR119 was reduced to thousand-fold less at 100 CFU per mL after 9 h (Figure 2-8). Similarly, greenhouse e xperiments showed that 91-106 + pL:ORF5 was able to establish population an average of 1.5 log higher than 91-106 + pLAFR119, when co-inoculated with 91-118:: bcnBC in leaf tissue (Figure 2-9). In addition, 91-118:: ORFA, 91-118:: ORF2, 91-118:: ORF3 and 91-118:: ORF4 mutants all maintained insensitivity to 91-118:: bcnBC in plate antagonism assays. These results clear ly demonstrate that OR F5 confers immunity. Sequence Analysis of BcnB and BcnC It has been shown previously (154) that plasmid pLAFR3 carrying 5.8-kb (pLB5.8) and 5.1-kb (pLC5.1) DNA fragments, BcnB and BcnC, re spectively, conferred b acteriocin activity to sensitive X. euvesicatoria strains. In order to id entify genes involved in BcnB and BcnC activity, subclones of different regions of those DNA frag ments were created in pLAFR119 (Figures 2-2 and 2-3). Each subclone was expressed in X. euvesicatoria strains ME90 or 91-106 and the ability to produce inhibition was tested on NA medi a using strain 91-106 as an indicator. BcnB and BcnC subcloned to 3.0-kb and 1.7-kb DNA fragments, respectively, carried on plasmid pLAFR119, were the smallest fragme nts that conferred bacteriocin activity (Figure 2-2 and 2-3). The nucleotide sequence revealed two complete ORFs named bcnB and bcnC BcnB shows homology to endoprotease Arg-C with a predicted amino acid size and molecular mass of 466 aa and 48, 487 MW. BcnB has no N-terminal signal peptid e. BcnB is predicted to be located in the bacterial outer membrane (0.933) or periplasm (0.258). BcnC shows homology to extracellular

PAGE 31

31 metalloprotease with a predicted amino acid si ze and molecular mass of 401aa and 42, 471 MW, respectively. BcnC has no N terminal signal peptid e. BcnC is predicted to be located in the bacterial inner membrane (0.351). The intr oduction of a stop codon (TAA, Table 2-2 in bold ) at the 5 end just downstream of the ATG start c odon disrupted BcnB activity when expressed in ME90 (ME90 + pL3.0mut) compared to the control (ME90 + pL3.0) (Figures 2-2 and 2-10). Plasmid pLAFR119 has only lac promoter. Dire ctional cloning of Bc nB and C genes was performed in pLAFR119. Plasmids pLB5.8 a nd pLC5.1 actively expressed BcnB and BcnC, respectively, without aid of the pL lac promoter (Figure 2-2 and 2-3), suggesting that their native promoters are functional. BcnB was subcl oned to a 3.0 kb fragment. An ORF with endoprotease Arg-C homology was determ ined to be responsible for BcnB+ activity based on analysis with of a stop codon (TAA) insertion in the forward direction (Fi gure 2-2). For BcnC, a 1.7 kb fragment of BcnC was directionally subc loned in pLAFR119 in both directions. The reverse direction BcnC (pL1.7CR) gave very s light bacteriocin activity compared to under direction of the lac pr omoter (Figure 2-3). Purification and Characteri zation of BcnB and BcnC Purification was conducted to ev aluate BcnB and BcnC activit y. Bacterial supernatant was concentrated with Microcon YM-100 (Millipore, Billerica, MA) or TC A precipitation (135). These concentrated samples were run on an SDS-PAGE gel and detected using Coomasie Brilliant Blue 250 (Pierce Biotec hnology, Rockford, IL) or Silver staining (BioRad Laboratories; Hercules, CA). No bands were detected. ME90 expressing ORFA was protease nega tive; however, ME90 expressing BcnB and BcnC produced clearing zones (2.0 cm and 3.1 cm respectively) typica l of protease activity (Figure 2-11). Size exclusion analysis was conducted using T3 strain 91-118:: ORFA

PAGE 32

32 supernatants. Protease activity was observed from total cell or filtrates of less than 50-kDa with only minor activity above 50-kDa. Type II Secretion Mutant Lost Secret ion of Amylase and Bacteriocins The role of the T2SS on delivery was determin ed for each bacteriocin by plate inhibition assay. Confirmation of deletion of xpsD was performed via PCR and analysis of starch hydrolysis media in XpsD mu tants (Figure 2-12). The X. euvesicatoria T2SS xpsD mutant expressing clones, 91-106 + pXV12.1 91-106 + pL5.8 and 91-106 + pL5.1, were unable to produce a zone of inhibition in plate assays while wt 91-106 expression of each clone produced typical zones for each bacteriocin. Discussion In this study, bacteriocins of X. perforans were further characterized to determine their activity and possible functions. Disruption of ORFA, ORF2 and ORF4 abolished BcnA activity, suggesting that BcnA is part of a multiple component family of bacteriocins. ORF5 was shown to encode the immunity function for BcnA, making normally sensitive X. euvesicatoria strains insensitive. This information and the predicted localization in the inner and outer membrane of the cell suggests that th ese ORFs make up the necessary parts of a three component system (the toxin, immunity and a mechanism for delivery) of a typical Gram-negative bacteriocin outlined by Riley and Wertz (128). Because of size selec tion (>100 kDA), ORFA is suggested to be the toxin, ORF2, ORF3 and ORF4 prot eins are responsible for deliver y and/or possible processing of BcnA (ORFA product) and ORF5 is the immunity function. All bioinformatics results (SOSUI and PSORT) suggest ORFA is a soluble cytoplas mic protein. BcnA was only detected in supernatants and not in detectable levels in the cell fraction of BcnA producing X. perforans cells. These results suggest BcnA may be activ ated upon secretion. ORF3 is included in the model because a mild reduction in antagonism wa s associated with ORF3 disruption. Based on

PAGE 33

33 predicted localization to the peri plasm and outer membrane, perhap s ORF3 aids in transfer of pre-bcnA (once pre-bcnA is in the periplasm) to ORF2 on the outer membrane. ORF2, ORF3 and ORF4 appear to play secondary roles such as in transport, modifi cation or secretion of BcnA. The zone inhibiti on assay and growth rate in vitro and in planta experiments strongly suggest that ORF5 is responsible for the immunity f unction. SOSUI predicted ORF5 would localize to the bacterial inner membrane s upported by a positive hydrophobicity value (0.2). This may suggest ORF5 disrupts BcnA or prevents delivery of active BcnA from entry into the cell. This is similar to what has been found for the immunity function of Colicin V (ColV) of E. coli (169). Col V is one of many known multiple component bacteriocins previously described (8, 99, 169). ColV was used as a reference for basic components of a Gram-negative bacteriocin (42, 43, 55, 56, 64, 169). ColV immunity was previously s hown to prevent inserti on of ColV into the inner membrane of sensitive strains of E. coli (169). Based on localization analysis (PSORT) we predicted the localization of each ORF involved in BcnA production. Base d on this information and what is known for ColV, we have developed a basic model for BcnA (Figure 2-13). For the BcnA model, all predicted locations for ORFs involved in BcnA activity were based on predicted localization and deletion analysis. The model suggests four steps: (I): pre-BcnA de livery into the periplasm with help of or chaperoned by ORF4; (II) Processing of pre-BcnA a nd delivery of the active BcnA outside of the cell by ORF2 and ORF3; (III) Entry of activ e BcnA into cells (unknown); (IVa) BcnA suppressed by ORF5; and (IVb) BcnA inhibition (e ither in the periplasm or in the cytoplasm). Previously BcnB activity was localized to a 5.9 kb fragment (153). Only two ORFs were found within this fragment that contained homology to genes of known function. One was an

PAGE 34

34 amino acid transporter and the ot her an endoprotease Arg-C. Bo th genes were isolated and tested for bacteriocin activity. On ly fragments containing the intact endoprotease like gene were active. This ORF was confirmed using an inse rted STOP codon (TAA) at the 5 end of the fragment which in turn lost activity confirmi ng that the endoprotease was responsible for the bacteriocin-like activity. Endoprot ease Arg-C is a family of se rine endoproteases which cleaves carboxyl peptide bonds of arginine residues. The enzyme has also been shown to cleave Lys-Lys and Lys-Arg bonds (119). BcnC was previously locali zed to a 1.7 kb fragment (61). Two possible ORFs were located within this fragment one in the plus and one in the minus direction. Directional cloning analysis shows that the plus directional ORF was responsible for BcnC activ ity. This gene had high homology to an extracellular metalloprotease ge ne family. Metalloproteases are proteolytic enzymes which use a metal for their catalytic mechanism. Most metalloproteases are zincdependent, while some use cobalt (3). BcnB and BcnC were tested for protease ac tivity based on homology data. Our findings show that both BcnB and BcnC exhibited protea se activity as determined by casein degradation analysis; however, BcnB produced smaller protease zones than BcnC. The results of the protein size filtration data were consistent with the predicted size of BcnB (48 MW) and BcnC (42 MW). The ORFs responsible for BcnA activity were identified and their possi ble roles have been hypothesized. Further research is needed to determine their specifi c roles. Only one gene was determined to be necessary for expression of BcnB or BcnC. The protease assays have determined their roles as proteases ; however, further research is necessary to determine the target of these proteases within sensitive strains.

PAGE 35

35 Table 2-1. Bacterial strains and plasmids used in this study Strain or plasmid Relevant characte ristics Source or reference* Xanthomonas euvesicatoria E3-1 NalRSmR (61) ME-90 RifRKmR (154) 91-106 (154) 91-106:: xpsD XpsDCmR This study X. perforans 91-118 RifR (154) 91-118:: ORFA ORFARifR This study 91-118:: ORF2 ORF2RifR This study 91-118:: ORF3 ORF3RifR This study 91-118:: ORF4 ORF4RifR This study 91-118:: ORF5 ORF5RifR This study Escherichia coli DH5 FrecA BRL C2110 NalR BRL PIR Host for pOK1; SpR oriR6K RK2 replicon UB Plasmids pBluescript-KS+ Phagemid, pUC derivative; AmpR Stratagene pLAFR119 TcR rlx+ RK2 replicon BJS pRK2073 SpR tra+ mob+ (28) pRKP10 CmR cassette (123) pOK1 Suicide vector; SacB (63) pL5.8 pLAFR119 Kpn I/ Eco RI 5.8-kb BcnB+ fragment This study pL3.0 pLAFR119 Kpn I/ Xba I 3.0-kb BcnB+ fragment This study pL3.0mut pLAFR119 Kpn I/ Xba I 3.0-kb BcnBfragment This study with a TAA stop codon insertion pLKH pLAFR119 Kpn I/ Hin DIII 2.3-kb BcnBfragment This study pLHX pLAFR119 Xba I/ Hin DIII 2.4-kb BcnBfragment This study pL5.1 pLAFR119 HinD III/ Eco RI 5.1-kb BcnC+ fragment This study BRL, Bethesda Research Labor atories, Gaithersburg, MD; Stra tagene, Stratagene Inc., La Jolla, CA; BJS, B. J. Staskawicz, University of California, Berkeley, CA; UB, U. Bonas, MartinLuther-Universitt, Halle, Germany.

PAGE 36

36 Table 2-2. PCR primer used in bcnA bcnB and bcnC analyses for genetic manipulation plasmid Primer name restriction site Primer sequence length GC (%) Tm (C) pLAFR NP5 CCCTTCACCAAGTTCGACGACA 22 54.5 61 NP3 GCGGGTGCCGTGCTCGTGTT 20 70 65 BcnA A5 Xho I CCTCGAG ATGCGCCACCCGTCGG 16 75 60 A3 Xho I CCTCGAG CAGCAAAAGCTGATAGAGC 19 47.3 54 ORF5F Hin DIII GGGGAAGCTT CAGGGTGGCGGCAAGGGA 27 70.3 70.2 ORF5R Hin DIII GGGGAAGCTT GGGCTTCTCTGGAAGCGGAC 29 65.5 69.5 ORF3F Hin DIII CCCGAAGCTT CCGGTTGACCTCTATGTAGATGGATGC 36 55.6 68.8 ORF3R Hin DIII CCCGAAGCTT CCCAGTGCAAATGTAAGCCGCGAC 33 60.6 69.6 ORF2F Hin DIII GGGGAAGCTT ACACAGGACGGGACATGCACAG 31 61.3 68.9 ORF2R Hin DIII GGGGAAGCTT ACAACCTCCACATCTCGCACCG 31 61.3 68.9 ORF4F Hin DIII CCCAAGCTT GCCGGATGCGACATTGTTGCGC 31 61.3 70.2 ORF4R Hin DIII CCCAAGCTT GCTTGGTTCAAGCTCATCACC 30 53.3 66.5 BcnB B5' new Eco RI CGGAATTC CAATCGCAAGAACGCGATG 21 50 63.4 B32 Kpn I CGGGTACC CTGGCCGAAGTAGGTGGAAAT 21 52.3 69.3 BORF1F ATGGGCTTGTCGGCCACATAATCGTCACAA 30 50 68.1 BORF1R TTGTGACGATTATGTGGCCGACAAGCCCAT 30 50 68.1 BORF2F AACGAACGAAGGTTACACTGGCTCCACCAT 30 50 68.1 BORF2R ATGGTGGAGCCAGTGTAACCTTCGTTCGTT 30 50 68.1 BORF3F ATGAATCGCAAG TAA GCGATGTATCTGGCG 30 46.7 66.8 BORF3R CGCCAGATACATCGCTTACTTGCGATTCAT 30 46.7 66.8 BORF4F ATGGCTGCAAATTGATAATGCGCTCACGGT 30 46.7 66.8 BORF4R ACCGTGAGCGCATTATCAATTTGCAGCCAT 30 46.7 66.8 BORF5F CAGCCCGCGCGATTAGATGACCATTGCCAT 30 56.7 70.9 BORF5R ATGGCAATGGTCATCTAATCGCGCGGGCTG 30 56.7 70.9 BORF7F ATGGACGATCGCTAACCGTCGATCCGCTTC 30 56.7 70.9 BORF7R GAAGCGGATCGACGGTTAGCGATCGTCCAT 30 56.7 70.9 B5XhoI Xho I CCTCGAG ATGAATCGCAAGAACGCG 18 52.3 58 BcnC C5' Eco RI CGGAATTC CGTGAAGAACGTCTTCCTC 27 51.9 54 C3' Kpn I GGGGTACC CTTGTCGTCATC GTTCTGCGCCGGAGTGTT 37 50 54 C5XhoI Xho I CCTCGAG GTGAAGAACGTCTTCCTC 18 61.1 58 T2SS xpsDF ATGACGCCGCGCCTGTTTCC 20 65 58 xpsDR CCCTTCTCAAGTGGCTGCAT 20 60 58

PAGE 37

37 Table 2-3. Characterization of bacteriocin ORFs associated with the expression of BcnA, BcnB and BcnC AAaSize (kDa) Signal PeptidebLocalizationcLocation valuechydrophobicitybBcnAORFA1012111.0nonebacterial cytoplasm0.56-0.56 ORF212413.6MTLCIFLPLLCAKASAAPYVVMGNIVTROuter membrane or p eri p lasmic s p ace 0.926 & 0.175 0.28 ORF310011.0MRFYRISLLALIIFASPRASAOuter membrane or p eri p lasmic s p ace 0.326 & 0.939 0.20 ORF429031.9MNKCSDAYGIYLRTLFVFFMYTLFCTSASAQVIRYinner membrane0.187-0.36 ORF514515.9noneinner membrane0.1090.17 BcnB bcnB 46548.5none Outer membrane or periplasmic space 0.933 & 0.258 -0.07 BcnC bcnC 40042.5noneinner membrane0.351-1.44 a BcnA ORF predicted amino acid size (AA) were previously described (154). b Prediction of signal peptide was performed with SOSUI ( http://sosui.proteome.bio.tuat.ac.jp/sosuiframe0.html ). c Prediction of localization was performed using PSORTb v.2.0 ( http://www.psort.org/ ).

PAGE 38

38 1 4 5 6 7 8 9 10 ORFA 2 3 4 5 Bcn-A operon(Kb) A. B.ActivityImmunity ORF2-+ ORF3-+ RF4-+ ORF5-91-106 + pXV12.1++ 91-106 + ORF4-+ 91-106 + 4.5R-91-106 + ORF2-5R-91-106 + ORF2-5+ 91-106 + ORF2-3+91-106 + ORF291-106 + ORF391-106 + ORF491-106 + ORF5-+ orfA3 pOK1: orf2 orfA3 pOK1: orf2 H A EH X E ORFA H A EH X E ORFA pOK1: ORF2 Figure 2-1. BcnA diagram showing position of individual ORFs and positions and fragment constructs. The grey arrows indicate the directional expression of fragments under the lac promoter of pLAFR119. The circle on ME90 + ORF4 indicates the location of a transposon insertion. B. Diagram of deletion construct for ORF2.

PAGE 39

39 0.5 1 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 Amino acid transporter bcn-b (endoproteaseArg-C) (Kb) pL:XH pL:KH pL:3.0M + pL:3.0 + pL5.8 BCN plasmid pL:XH pL:KH pL:3.0M + pL:3.0 + pL5.8 BCN plasmid stop Figure 2-2. BcnB diagram showing position of i ndividual ORFs and direc tional expression (grey arrows) and activity (table) of fragment constructs under the lac promoter of pLAFR119. The x indicates an artificial stop cod on (TAA) insertion in frame of the bcnB sequence.

PAGE 40

40 0.5 1 1.5 2.0 2.5 3.0 3.5 4.0 4.5 (Kb) + pL1.7 pL1.7CR pL2.5 + pL5.1 BCN plasmid + pL1.7 pL1.7CR pL2.5 + pL5.1 BCN plasmid bcn-c (Extracellularmetalloprotease) glycine-rich protein Figure 2-3. BcnC diagram showing position of i ndividual ORFs and direc tional expression (grey arrows) and activity (table) of fragment constructs under the lac promoter of pLAFR119.

PAGE 41

41 ++ ORF3+++inhibition zone ORF4 ORF2 ORFA WT T3 Test Strain ++ ORF3+++inhibition zone ORF4 ORF2 ORFA WT T3 Test Strain Figure 2-4. Antagonism assays of ba cteriocin-like activity against X. euvesicatoria strain 91-106 were assessed using cell-free supernatan ts of individual ORF knockout mutants of 91-118.

PAGE 42

42 +Inhibition zone ORF4 ORFA WT T3 Test strain +Inhibition zone ORF4 ORFA WT T3 Test strain -Inhibition zone ORF4 ORFA WT T3 Test strain -Inhibition zone ORF4 ORFA WT T3 Test strain Supernatant Cell fraction Figure 2-5. Antagonism activation assays of bacteriocin-lik e activity against X. euvesicatoria strain 91-106. Supernatant from X. perforans was separated into supernatant and cell fractions.

PAGE 43

43 ++ ++ +++Inhibition Zone X > 100 50 < X < 100 X < 50 All Size range (kDa) ++ ++ +++Inhibition Zone X > 100 50 < X < 100 X < 50 All Size range (kDa) Figure 2-6. Secretion assays of bacteriocin-like activity against X. euvesicatoria strain 91-106. Supernatant from X. perforans was separated based on size exclusion technique into three fragments, less than 50, 50 to 100 and over 100 kDa.

PAGE 44

44 + +inhibition zone ORF2 to 5 ORF2 & 3 pXV12.1 pLAFR119 challenge strain + +inhibition zone ORF2 to 5 ORF2 & 3 pXV12.1 pLAFR119 challenge strain + + +inhibition zone pL:ORF 5 pL:ORF4 pL:ORF3 pL:ORF2 challenge strain + + +inhibition zone pL:ORF 5 pL:ORF4 pL:ORF3 pL:ORF2 challenge strain Figure 2-7. Immunity assays of bacteriocin-like activity against X. euvesicatoria strain 91-106 using 91-118:: bcnBC as the producing strain. Bact eriocin sensitivity candidates were screened to identify ORFs expressing immunity.

PAGE 45

45 0 2 4 6 8 0369 Time after inoculation (h)Log10 CFU/m L Figure 2-8. In vitro populations of X. euvesicatoria strain 91-106 transconjugants with [pLAFR119 ( ) and pLAFR:ORF5 ( x )] and without [pLAFR119 ( ), pLAFR119:ORF5 ( )] co-inoculation of 91-118:: bcnBC Error bars indicate the standard error.

PAGE 46

46 0 1 2 3 4 487296 Time after inoculation (h) Log10 CFU/cm2 of leaf tissue Figure 2-9. In planta populations of X. euvesicatoria strain 91-106 containing plasmid pLAFR119 ( ) or pLAFR119:ORF5 ( ), respectively, when co-inoculated with 91-118:: bcnBC (BcnA expression only) in leaflets of tomato cultigen Bonny Best. Error bars indicate the standard error.

PAGE 47

47 +ME90 + pL3.0+ -inhibition zone ME90 + pL3.0mut ME90 + pLKH ME90 + pL5.8 ME90 + pLAFR119 Test Strain +ME90 + pL3.0+ -inhibition zone ME90 + pL3.0mut ME90 + pLKH ME90 + pL5.8 ME90 + pLAFR119 Test Strain Figure 2-10. Antagonism assays of bacteriocin-like activity against X. euvesicatoria strain 91-106 were assessed using cell-free supe rnatants of individual ORF knockout mutants of 91-118.

PAGE 48

48 + + +++ +++Protease activity X > 100 50 < X < 100 X < 50 All Size Range (kDa) + + +++ +++Protease activity X > 100 50 < X < 100 X < 50 All Size Range (kDa) +ME90 + B+++ +++Protease activity ME90 + C ME90 + A ME90 + pLAFR WT T3 Strain +ME90 + B+++ +++Protease activity ME90 + C ME90 + A ME90 + pLAFR WT T3 Strain Figure 2-11. Protease activit y of bacteriocin candidates. A. Protease assay where 5 L of 5 x 108 CFU/mL of each strain was plated onto 0.5% skim milk agar and incubated for 24 h at 28 C. B. Fractions from X. perforans were separated based on size exclusion technique into three fr agments, less than 50, 50 to 100 and over 100 kDa and analyzed for protease activity. A. B.

PAGE 49

49 -+ Protease activity 91-106:: xpsD 91-106 Test strain -+ Protease activity 91-106:: xpsD 91-106 Test strain ++ ++ +++BcnAactivity 91-106:: xspD + pXV12.1 91-106 + pXV12.1 91-106 + pL 91-106 Test strain ++ ++ +++BcnAactivity 91-106:: xspD + pXV12.1 91-106 + pXV12.1 91-106 + pL 91-106 Test strain Figure 2-12. Evaluation of type two secretion system mutant xpsD A. Proteinase assay of 91-106 and 91-106:: xpsD B. Antagonism assays of BcnA using cell-free supernatants using X. euvesicatoria strain 91-106 as an indicator strain. A. B.

PAGE 50

50 BcnA BcnA Pre BcnA BcnA Orf4Outside Periplasm InsideOrf5 Orf2 Orf3 Orf3 Step 2 Step 1 Step 3 Step 4aStep 4b Figure 2-13. BcnA Model for ORFs predicted involved in BcnA activity based on predicted localization and deletion analys is. Step 1: pre-BcnA de livery into the periplasm by ORF4. Step 2: Processing of pre-BcnA and delivery of the active BcnA outside of the cell by ORF2 and ORF3. Step 3: Entry of ac tive BcnA into cells (unknown). Step 4 a: BcnA suppressed by ORF5. Step 4b: BcnA i nhibition (either in the periplasm or in the cytoplasm). ?

PAGE 51

51 CHAPTER 3 ANALYSIS OF PATHOGENICITY MUTANTS OF XANTHOMONAS PERFORANS AND THEIR EFFECT ON BACTERIOCIN EXPRESSION Fresh-market tomato production accounts for mo re than 50% of the harvested acres and 63% of the national yield in the southeastern U.S., with approximately 40,000 acres in Florida (29). Bacterial spot of tomato, incited by Xanthomonas euvesicatoria is a devastating disease of tomato in Florida, the Caribbean and worl dwide (11, 108, 114, 118, 132). There are at least three important management strategies to reduc e severity and incidence of bacterial tomato diseases: reducing initial inocul um, minimizing plant susceptibili ty (natural resistance) and chemical control (copper-based chemicals and antibiotics). Although ch emical control using copper bactericides is routinely used to contro l bacterial spot, these efforts are often futile because of the presence of copper-tolerant strain s of the bacterium (90). Adding mancozeb, a fungicide, to copper sprays was shown to impr ove control efficiency and was shown by Marco and Stall (90) to control copper-tole rant strains. However, they also showed that this treatment is insufficient when conditions favorable for disease development exist. Recently, new strategies have emerged that co uld be utilized as alternative management practices. These include using bacteriophages specifi c to the target bacteriu m (5), application of plant activators that induce systemic acquired resistance (SAR) in the plant to bacterial pathogens (89) and application of bacterial bi ological control agents (49, 67). These new strategies focus on reducing initia l epiphytic or internal populati ons of the potential pathogen. Several studies on xanthomonads have shown a relationship between epiphytic populations of the pathogen and disease se verity (73, 81). Lindemann et al. (86) demonstrated a strong correlation between a threshold le vel of epiphytic populations of Pseudomonas syringae and the occurance of disease in the field. Reducing ep iphytic populations may pl ay a key role in a successful biological control strategy.

PAGE 52

52 Recently Hert et al. (61) demonstrated that populations of X. euvesicatoria strains were reduced significantly by bacteriocin-producing X. perforans strains in the field. These results are supportive of the potential of bacteriocins for co ntrolling bacteriocin-sensitive strains and may provide a new approach to bi ological disease control. Non-pathogenic Hrp mutants of X. perforans provided significant levels of control (88, 102) against bact eriocin sensitive strains; however, there were still unacceptable levels of disease. In other studies, Hrp mutants were created in a bacteriocin expressing Ralstonia solanacearum strain for control of wild-type (wt) R. solanacearum strains (41, 49, 152). These studies revealed that slightly pa thogenic hrp mutants that were able to achieve higher levels of root and stem tissue co lonization provided higher levels of control than other hrp mutants with minimal colonization capabilities (49, 152). Etchebar et al. (41) suggested that there was a relations hip between the degree of colonization of the xylem by the mutant and the level of control of the wt R. solanacearum X. perforans also causes disease, therefore if X. perforans is to be used as a biological control agent, it is necessary to reduce the virulence of X. perforans to levels that are not dele terious to the plant. There has been progress in identifying ge nes involved in bact erial virulence of Xanthomonas species (3, 4, 44, 58, 64, 165). Identification of these genes as potential targets to create strains attenuated in vi rulence has become much easier as a result of genome sequencing of xanthomonads. Thieme et al. (150) sequenced the X. euvesicatoria genome and estimated that 480 putative pathogenicity factors and associated genes are found in the genome. These factors were placed into six categories: (I) secretion sy stems, (II) flagellum, (III) secreted proteins (via type III secretion system (T3SS)), (IV) det oxification, (V) surface structure and adhesion and (VI) quorum sensing. Based on previous studies, several candidate genes ( opgH avrBs2, hpaA

PAGE 53

53 hpaB hpaC xopA xopD and gumD ) appeared interesting for use in this study (18, 17, 21, 63, 76, 101, 106). One known X. perforans pathogenicity-attenuating mutant served as our model candidate, opgHXcv. The o pgH mutant was previously shown to have an attenuated virulence phenotype (101), exhibiting reduced disease severity and gr owth curve in susceptible tissue and triggering delayed hypersensitive reaction (HR) in a resistan t plant genotype. A second candidate gene, the effector avrBs2 was chosen as another candidate as a result of its role in virulence of X. euvesicatoria in pepper (76). It has been shown that avrBs2 mutants are less virulent on susceptible hosts (53, 76, 146). Genes hpaA hpaB and hpaC were shown to be involved in virulence. These genes play a role in pathogenicity as an effector ( hpaA ) and as chaperones ( hpaB and hpaC ) of effectors of the T3SS (18, 17, 63). HpaA appears to func tion as an effector molecule in X. euvesicatoria since disruption of hpaA eliminates disease symptoms in tomato and pepper plants without affecting the ability to elicit a hypersens itive response. HpaB and HpaC were shown to form an oligomeric protein complex and interact with tw o classes of effectors (class A containing XopJ and XopF1 and class B containing AvrBs3 and XopC) and HrcV of the T3SS (18). Two genes that code for pr oteins that are designated Xanthomonas outer proteins (Xop) were also chosen as potential candi dates that may affect virulence. XopA and XopD are secreted by the T3SS and thus represent putative effect or proteins. XopA is necessary for both in planta growth and full virulence (106). GumD of the gum operon is involved in xanthan gum bi osynthesis (7, 75). Xanthan gum is a high molecular weight extr acellular heteropolyme r produced by xanthomonads and has been

PAGE 54

54 implicated as a virulence factor based on deletion analysis in X c pv. campestris. Xanthan gum minus gumD mutants exhibited dramatic de lay in disease symptoms (21). In this study, mutations were created in the candidate genes described above to determine their role in pathogenicity and possible use as pathogenically-attenuated bacteriocin-expressing biological control agents. Materials and Methods Bacterial Strains, Plasmids and Culture Conditions Strains of X perforans and X euvesicatoria were grown on nutrient agar (NA) medium (Difco Laboratories, Detroit, MI) at 28 C (Table 3-1). Strains of Escherichia coli were grown on Luria-Bertani (LB) medium at 37 C (Table 3-1) (97). All strains were stored in 20% glycerol in sterile tap water at C. Bacterial cultures for plant inocul ations were grown in nutrient broth (NB) (Difco Laboratories, Detr oit, MI) for 18 h at 28C with shaking (100 rpm). Cells were pelleted by centrifugation (4,000 g 15 min) and resuspended in sterile tap water. Bacterial suspensions were standardized to an optical density at 600 nm (OD600) = 0.3 (5 108 CFU/mL) with a Spectronic 20 spectrophotometer (S pectronic UNICAM, Rochester, NY) and subsequently diluted in sterile tap water to appr opriate cell densities for individual experiments. Antibiotics were used to maintain selec tion for resistance markers at the following concentrations: tetracycline (Tc) 12.5 g/mL; rifampicin (Rif) 100 g/mL; spectinomycin (Sp) 50 g/mL; kanamycin (Km) 50 g/mL; chloramphenicol (Cm) 34 g/mL; streptomycin (Sm) 200 g/mL; and nalidixic acid (Nal) 50 g/mL. Plant Material Seeds of tomato ( Lycopersicon esculentum ) cv. Bonny Best were planted in Plugmix (W. R. Grace & Co., Cambridge, MA). After 2 weeks, the emerged seedlings were transplanted to

PAGE 55

55 Metromix 300 (W. R. Grace & Co.) in 10-cm pl astic pots. Seedlings were grown in the greenhouse at temperatures ranging from 25 to 35C. Primer Design Candidate genes (Table 3-2) were amplifie d using primers designed as follows. BLAST search analysis was conducted to locate conser ved regions flanking the candidate gene sequence by scanning genomic sequences of closely related Xanthomonas species (i.e. X. euvesicatoria strain 85-10 ( NC_007508.1 ), X. oryzae pv. oryzicola ( AY875714.3 ) X. oryzae pv. oryzae strain KACC10331 ( NC_006834.1 ), X. axonopodis pv. citri strain 306 ( NC_003919.1 ) and X. campestris pv. campestris strain ATCC 33913 ( NC_003902.1 ). Primers were designed to conserved regions for amplificati on of the corresponding regions in X. perforans strain 91-118 (Table 3-2). Candidates were confirmed by sequ ence analysis. Sequencing of the clones was conducted at the ICBR sequencing facility (Univers ity of Florida, Gainesville, FL) with the Applied Biosystems model 373 system (Foster City, CA). Generation of Mutants Candidate genes for creating attenuated mutants of X. perforans ( avrBs2, hpaA, hpaB, hpaC, xopA and xopD ) were disrupted using either restrict ion digestion or PCR-assisted deletion mutagenesis. All candidate genes were amplif ied and deleted using PCR primers described in Table 3-2. Each attenuated can didate was individually cloned into pGEM (Promega, Madison, WI) or pTOPO (Invitrogen, Carisbad, CA) then an internal fragment was deleted using restriction digestion or PCR assi sted deletion mutagenesis and re placed with a chloramphenicol resistance cassette. An avrBs2 mutant was created as follo ws. A fragment containing avrBs2 was amplified by PCR using primers avrbs2F and avrBs2R and cloned into pTOPO to create pTOPO: avrBs2 For disruption of avrBs2 two primer pairs, avrBs2F, avrBs2DD2F, avrBs2DD2R and avrBs2R

PAGE 56

56 were utilized to amplify portions of the 5 and 3 ends separately to delete a 174-bp fragment not amplified in either set of prim ers (Figure 3-1A). Both frag ments (5 and 3 products) were cloned into pTOPO creating pTOPO:avrBs2F and pTOPO:avrBs2DR. Primers avrBs2DD2F and avrBs2DD2R contain artificial Sal I restriction sites. The pTOPO:avrBs2-3 was then digested with Sal I (artificially created by avrBs2DD2 R) and pTOPO restriction site Spe I, then ligated into pTOPO:avrBs2DF, as illustra ted in Figure 3-1A, to create pTOPO: avrBs2 Next, the artificial Sal I restriction enzyme sites from avrBs2DD2F and avrBs2DD2R were utilized to insert a Cm resistance gene cassette (124) to create pTOPO: avrBs2 Finally, using pTOPOs MCS restriction sites Bam HI and Apa I, the deleted avrBs2 was ligated into pOK1 to create pOK1: avrBs2 The final construct was then mated in to 91-118 as described later in this section. The hpaA gene was amplified by PCR using primers hpaAF and hpaAR, containing artificial Hin DIII restriction sites and cloned into pGEM to create pGEM: hpaA For disruption of hpaA divergent PCR primers hpaADF and hpaADR we re utilized to delete a 446-bp internal fragment of hpaA (Figure 3-1B). Artificial Sma I restriction enzyme sites were added by primers hpaADF and hpaADR, then utilized to insert a Cm resistance gene cassette (124) to create pGEM: hpaA Finally, using pGEMs MCS restriction sites Spe I and Apa I, the deleted hpaA was ligated into the pOK1 to create pOK1: hpaA The final construct was then mated into 91118 as described later in this section. For deletion of hpaB a fragment containing hpaB was amplified by PCR using primers hpaBF and hpaBR, containing artificial Hin DIII restriction sites (not used in this study) and cloned into pGEM to create pGEM: hpaB For disruption of hpaB, a Cm resistance gene cassette (124) was inserted into a native Kpn I restriction site within hpaB to create pGEM: hpaB

PAGE 57

57 Finally, using pGEMs MCS restriction sites Spe I and Apa I, the deleted hpaB was ligated into the pOK1 to create pOK1: hpaB The final construct was then mated into 91-118 as described later in this section. A fragment containing hpaC was amplified by PCR using primers hpaCF and hpaCR and cloned into pGEM to create pGEM: hpaC For disruption of hpaC, divergent PCR primers hpaCDF and hpaCDR were utilized to de lete a 58-bp intern al fragment of hpaC Artificial Hin DIII restriction enzyme sites were added by hpaC DF and hpaCDR and utili zed to insert a Cm resistance gene cassette (124) to create pGEM: hpaC Finally, using pGEMs MCS restriction sites Spe I and Apa I, the deleted hpaC was ligated into the pOK1 to create pOK1: hpaC The final construct was then mated into 91118 as described later in this section. A fragment containing xopA was amplified by PCR using primers XopAF and XopAR, containing artificial Hin DIII restriction sites, and clone d into pGEM to create pGEM: xopA For disruption of xopA two Ava I restriction sites, located within pGEM: xopA were utilized. This Ava I deletion completely deleted xopA leaving flanking DNA for marker exchange. Finally, using pGEMs MCS restriction sites Spe I and Apa I, the deleted xopA was ligated into the pOK1 to create pOK1: xopA The final construct was then mated into 91-118 as described later in this section. A fragment containing xopD was amplified by PCR using primers xopDF and xopDR and cloned into pGEM to create pGEM: xopD For disruption of xopD divergent PCR primers xopDDF and xopDDR were utilized to de lete a 134-bp internal fragment of xopD Artificial Hin DIII restriction enzyme sites were added by x opDDF and xopDDR were utilized to insert a Cm resistance gene cassette (124) to create pGEM: xopD. Finally, using pGEMs MCS

PAGE 58

58 restriction sites Spe I and Apa I, the deleted xopD was ligated into the pOK1 to create pOK1: xopD The final construct was then mated into 91-118 as described later in this section. A fragment containing gumD was amplified by PCR using primers GumDF and GumDR and cloned into pGEM to create pGEM: gumD Two Nco I restriction sites within pGEM: gumD were used for disruption of gumD This Nco I deletion removed a 400bp segment of the gumD gene leaving 5 and 3 porti ons of flanking DNA for marker ex change. Finally, using pGEMs MCS restriction sites Spe I and Apa I, the deleted gumD was ligated into the pOK1 to create pOK1: gumD The final construct was then mated into 91-118 as described later in this section. All candidates were confirmed by PCR using primers pOK1F and pOK1 R (Table 3-2) for sequencing of each candidate deletion at the DNA Sequencing Core Labor atory, as mentioned previously. Once sequence analysis confirmed dele tion of each candidate, su icide vector assisted mutagenesis was performed as described previous ly (74). Candidates were screened using PCR primers designed to amplify flanking regions of the cross-over region (Table 3-2). Deletion mutant 91-118:: avrBs2 was also confirmed in resistant cultivars of tomato (transgenic VS36 containing 35S::B s2 (148)) and pepper (ECW-20R ) for loss of HR. Leaves were infiltrated with a bacterial suspension of 5 x 108 CFU/mL using a hypodermic syringe as described previously (68) and scored for presen ce or absence of a hypersensitive response (HR) after 24 h and observed for 72 h. Growth Room Growth Curve Assays Growth room assays were conducted to compar e the growth curve of the deletion mutants with that of parent strain 91-118. Th e strains were inoculated at 3 105 CFU/mL into leaflets of 6-week-old seedlings of the tomato cultivar Bonny Best. Leaflets were infiltrated (15 leaflets per strain) using a hypodermic syringe and needle, as described previous ly (68). Following

PAGE 59

59 inoculation, plants were incubate d at 24C to 28C. Three samples were taken for each treatment every 24 h for 96 h. Populations were quantified by macerating 1-cm2 leaf disks in 1 mL sterile tap water and dilution plating on to NA medium amended with the appropriate antibiotic. Plates were incubated at 28C and colonies were count ed after 48 to 72 h. Population data were log transformed and standard errors were determin ed. The overall growth curve was determined by calculating the area under the population pr ogress curve (AUPPC). The AUPPC is a modification of the area under the disease prog ress curve (AUDPC) which has been used to analyze population progress ( 136): standardized AUPPC = [( xi + xi 1)/2]( ti ti 1), where x is population density in log10 CFU per cm2 and t is time in hours. The AUPPC values for the strains were compared by analysis of variance and subsequent separation of sample means by Waller-Duncan multiple range test using SAS version 9.0 (SAS Inc., Cary, NC). Each experiment was conducted three times. Greenhouse Disease Severity Assay Greenhouse pathogenicity assays were c onducted to compare symptom development incited by the mutants and wt 91-118. In each te st, four young Bonny Best plants (four-true-leaf stage) were inoculated with each strain by dippi ng into sterile tap wate r suspensions containing 3 106 CFU/mL of bacteria and 0.025% Silwet L-77 (Loveland Industries, Inc., Greeley, CO) for 15 s. Plants were maintained in the gree nhouse during the evaluation period. The plants were assessed for disease severity 14 to 21 days after inoculation. Disease assessments were made based on leaf and stem ratings compiled from three separate greenhouse inoculation tests. Growth Room Antagonism Assay Antagonism assays were performed to determine the effect of wt and mutant X. perforans strains on a sensitive X. euvesicatoria strain E3-1. Internal and external/leaf surface (phyllosphere) populations were evaluated usi ng two different antagonism assay techniques.

PAGE 60

60 Internal antagonism assay. Six-week-old seedlings of the tomato cultigen Florida 47 were inoculated with a 5 107 CFU/mL suspension of X. perforans (15 leaflets per strain) using a hypodermic syringe as described previously (6 8) followed 12 h later by injecting a 5 106 CFU/mL suspension of bacteriocin-sensitive X. euvesicatoria strain E3-1. Each treatment consisted of three replications. Followi ng inoculation, plants were incubated at 24 C to 28 C. In order to determine populations of the sensitive strain (E3-1SmNal) in leaflets, 1-cm2 leaf disks were removed from inoculated areas, macerated in 1 mL sterile tap water and dilution plated onto NA amended with Nal and Sm to qualify E3 -1 populations. Samples were assayed at 24 h intervals for 96 h. Each experiment was conduc ted three times. Population data were log transformed and standard errors were determin ed. AUPPC values (calculated as described above) were compared by analysis of variance and subsequent separation of sample means by Waller-Duncan multiple range test using SAS version 9.0 (SAS Inc., Cary, NC). Phyllosphere antagonism assay. Growth room phyllosphere antagonism assays were conducted to determine if the gene deletions affect ed the levels of antago nism toward external leaf populations of X. euvesicatoria strain E3-1 by comparing the antagonistic ability of the mutants with the parent strain 91-118. Six-week-old Bonny Best tomato seedlings were dipped into a suspension of the wt or mutant strain of X. perforans adjusted to 5 107 CFU/mL amended with Silwet L77 (0.025%). Seven days later the plants were sprayed with a 5 107 CFU/mL suspension of X. euvesicatoria strain E3-1. Following spray inoculation, plants were incubated at 24 C to 28 C. Leaf tissue was sampled every 24 h for 96 h to quantify E3-1 populations. Three leaflets were taken at each time point. Each leaflet was weighed, placed in a polyethylene bag containing 10 mL of sterile ta p water and shaken on a wrist action shaker (Burrel Co., Oakland, CA) for 20 min. The lea f-wash was then dilution plated on NANalSm to

PAGE 61

61 selectively determine the concentration of E3-1 colonies. Population data were analyzed following log10 transformation and st andard errors were determine d. AUPPC values (calculated as described above) were compared by analysis of variance and subsequent separation of sample means by Waller-Duncan multiple range test using SAS version 9.0 (SAS Inc., Cary, NC). Each experiment was conducted three times. Results Sequence Analysis of Attenuated Mutant Candidate Genes The eight candidate genes selected for disrup tion (Table 3-3), were amplified and cloned from X. perforans strain 91-118 and sequenced (Appendi x A-2, A-3, A-4, A-5, A-6 and A-7). Sequence analysis of each gene was c onducted to determine relatedness between X. perforans AA sequence and other proteins using BLAST sear ch protocol (Blastp) at the NCBI website ( http://www.ncbi.nlm.nih.gov/BLAST/ ). For hpaA hpaB hpaC and avrBs2 the nucleotide and deduced amino acid sequences of these genes ha d very high homology (75% to 100%) (Table 34) to the corresponding genes in other xanthomonads such as X. euvesicatoria strain 85-10 ( NC_007508.1 ), X. o. oryzicola ( AY875714.3 ) X. o. oryzae KACC10331 ( NC_006834.1 ) X. a. citri strain 306 ( NC_003919.1 ) X. c campestris strain ATCC 33913 ( NC_003902.1 ) and X. c. glycines ( AF499777.1 ) (Appendix A-8, A-9, A-10 and A-11 a nd Table 3-4). XopA and XopD had less AA homology than all other attenuated mutant candidate ge nes (Appendix A-12 and A13 and Table 3-4). XopA only had high homology to X. euvesicatoria strain 85-10 (100%) and much lower homology with X. c. glycines (47%). XopD only had high homology to X. euvesicatoria strain 85-10 (85%) and X. c campestris strain ATCC 33913 ( 74%). XopD was also found to have 86% nucleotide homology with P. syringae pv. eriobotryae gene psvA ( AB018553 ).

PAGE 62

62 Population Dynamics and Pathogenicity Assays In growth room experiments, there were three separate groups according to overall AUPPC values. The first group in cludes wt 91-118, 91-118:: hpaB and 91-118:: avrBs2 Although 91-118:: avrBs2 was not considered significan tly different overall, it was significantly different at the 96 and 120 h time points according to stan dard error. Populations of 91-118 and 91118 hpaB exhibited a normal growth curve over the 120 h sample period and, based on the AUPPC (Table 3-5). The second group included 91-118:: xopA and 91-118:: gumD These mutants were not significantly different from each other and grew 0.5 to 0.75 log10 CFU/mL lower than wt 91-118 throughout the e xperiment (Figure 3-2). 91-118:: hpaC was between groups according to overall significant difference and was not considered significantly different according to standard error at 48 and 72 h from 91-118:: xopA and 91-118:: gumD. The third group consisted of only 91-118:: opgH 91-118:: hpaC was considered significantly different from 91-118:: opgH at 24, 48 and 72 h according to standard error. 91-118:: opgH consistently grew 1 to 1.5 log10 CFU/mL lower than wt 91-118. 91-118: xopA and 91-118:: avrBs2 The largest reduction was observed for 91-118:: opgH which was consistently significantly lower populations over the experiment. Greenhouse disease severity expe riments were conducted to determine the effects of the mutations on the ability of 91-118 to cause disease in planta 91-118:: xopA and 91-118:: opgH mutants induced 1.5 to 3.2 times less dis ease than 91-118 (Figure 3-15 and Table 3-5). The largest reduction in di sease severity was observed in 91-118:: opgH (Table 3-5). The avrBs2 disruption mutant was confirmed on pepper and tomato genotypes expressing the Bs2 corresponding R gene. A resistance response (HR) was only observed in tomato leaves infiltrated with wt 91-118, infiltration with 91-118: avrBs2 did not give an HR (Figure 3-3).

PAGE 63

63 Antagonism Assays Growth room (internal and ex ternal) antagonism assays were conducted to determine the antagonistic ability of deletion-mutant candidates toward E3-1. Treatment with water prior to X. euvesicatoria resulted in a normal growth curve over the 96-h sampling period in both experiments based on the AUPPC (Figures 3-4 and 3-5). All attenuated mutant candidates significantly reduced X. euvesicatoria populations in the internal antagonism experiment (Figures 3-5). Wt 91-118 gave the most significant reduction in E3-1 populations. There were 2 groups of mutant s from this experiment, however all were not significantly different from one another. 91-118:: hpaB 91-118:: xopA and 91-118:: gumD gave the most reduction of the mutants tested, howe ver they were not significantly different from any the other strains overall. Looking at sta ndard error, however, they appeared to be significant from hpaC at 96 h. The second group of mutants (91-118:: hpaC and 91-118:: avrBs2 ) was non-significantly different from the X. euvesicatoria 91-106 strain treatment. In the external antagonism assa y, all attenuation mutant candida tes tested were similar to wt 91-118 in antagonism, however th ey were significantly different overall (Table 3-6). At 48 and 72 h, however, 91-118:: opgH was not significantly different according to standard error. Overall, wt 91-118 and all 91118 mutants tested were signifi cantly effective at reducing populations of E3-1. Discussion The goal of this project was to identify ge nes mutants that woul d provide a dramatic reduction in disease symptoms while still maintain ing the significant expression levels of BcnA and BcnC. Several pathogenicity related genes ( hpaA hpaB hpaC xopA xopD avrBs2 gumD )

PAGE 64

64 were evaluated for their possibl e attenuating effects when in X. perforans 91-118:: opgH served as the model system. This opgH mutant had an attenuated phe notype in disease severity and growth curve experiments, as observed previ ously (101), and maintained its ability to reduce X. euvesicatoria significantly better than water and X. euvesicatoria controls. The opgH mutant was selected as a model system for these experiments because it was also chosen for further investigation in the form of fi eld experiments in Chapter 4. Growth curve analysis suggests mutants (91-118:: xopA 91-118:: gumD 91-118:: hpaC and 91-118:: opgH ) were effected in overall fitness w ithin the plant by exhibiting a reduced growth curve peak compared to wt 91-118. There are a few hypotheses that may explain why we observed this reduction. One explanation could be associated with the effect of the mutants on the overall fitness of the bacterium. A sec ond hypothesis may be that there is recognition of the pathogen by the plant due to the mutations cr eated. It does not appear that these mutants would eventually reach the levels based on the stationary phase of the curve at 96 to 120 h supporting the second hypothesis. Further resear ch is needed to solid ify which or if both hypotheses is correct. Of the mutants created in this study, 91-118:: gumD and 91-118:: opgH exhibited the overall characteristics we were looking. Both mutants caused significant reductions in growth curve and disease severity while maintaining rela tively high levels of antagonism in internal and external antagonism experiments. Xanthan gum biosynthesis appears to be important in pathogenicity for X. perforans as described previously in X. c. campestris (21). This gene may be of interest for further inve stigation in designing a pathogenicity-attenuat ed biological control agent.

PAGE 65

65 Mutant 91-118:: xopA incited intermediate levels of di sease severity and reduced growth curve compared to wt 91-118. Deletion of xopA affected both internal and external antagonism compared to wt X. perforans however, the bacteria still main tained relatively high levels of antagonism. Overall XopA is necessary for full virulence and in planta growth as previously described in X. euvesicatoria (106), however, disease levels caused by the xopA mutant were still too high (>25%) to be a viable pathogenicity-attenuated mutant for use as a biocontrol agent. 91-118:: hpaC and 91-118:: avrBs2 were significantly different in overall growth rate, however, they were the most affected in ove rall antagonism according to internal antagonism experiments. Both only reduced E3-1 leve ls similar to a, non-bacteriocin producing, X. euvesicatoria 91-106 strain and were not significantly di fferent from the other mutants tested or the 91-106 treatments (Table 3-6). Two candidates, 91-118:: hpaB and 91-118:: avrBs2 exhibited growth curve and disease severity with overall similarity to wt 91-118. These candidates could, how ever, be distinguished from wt, based on antagonism in internal antagonism assays. Although these genes were previously shown to be involved in virulence in X. euvesicatoria (18, 17, 53, 63, 76, 146), they do not appear to affect disease severity sufficien tly to be feasible for creating a pathogenicityattenuated biological control agen t. Furthermore, these mutants dramatically reduced bacteriocin antagonism; however, further research is needed to determine if this reduction is due to overall fitness of the bacteria or pa rtial recognition and suppression of the bacterium by the plant as hypothesized earlier. Two genes, opgH and gumD exhibited the desired attenuation and bacteriocin activity when mutated. Although the othe r mutants did not reduce pathogeni city sufficiently alone, they may provide a more dramatic effect when in combination with one another. HpaB and HpaC are

PAGE 66

66 chaperones for effectors to the T3SS channel (H paB and HpaC) (18, 17), and have been shown to directly interact forming an oligomeric pr otein complex and interact with two classes of effectors (class A containing XopJ and XopF1 and class B containing AvrBs3 and XopC) and HrcV of the T3SS (18). Base d on their known in teraction in X. euvesicatoria HpaB and HpaC may provide a greater reduction in disease when both are knocked out within the same bacterium (18). Although eight pathogenicity related genes were selected to study, there is an abundant source of genes involved in pathogenicity that coul d be exploited to create further pathogenicityattenuated mutants. Many xanthomonads and ot her pathogenic bacteria have been recently sequenced such as X. euvesicatoria strain 85-10 (150), X. axonopodis pv. citri strain 306 (27), Xanthomonas campestris pv. campestris (27), X. oryzae pv. oryzae (85), P. syringae pv. tomato (15) and R. solanacearum (133). These sequences will pr ovide essential information for understanding how bacteria develop a pathogenic relationship with the host. For instance, Thieme et al. has estimated that there are ~480 putative pathogenicity factors in X. euvesicatoria strain 85-10 (150). This represents a large pool of putative pathogeni city genes, and therefore, opportunity to utilize these genes to create pathogenicity-attenuated mutants. Although there are a number of reports about mutants with reduced pathogenicity in X. euvesicatoria (17, 18, 54, 76, 105, 106, 107, 130, 146, 171), there are only a few reports of attenuated phenotypes in X. perforans (101) This information was utilized to crea te a biological control strategy to allow a bacteriocin-producing X. perforans strain to effectively colonize the plant and deliver bacteriocins while causing little to no disease. Other secretion systems may prove to be important in pathogenicity as well. X. euvesicatoria has many substrates delivered via the type II secretion system such as cellulases,

PAGE 67

67 -glucosidases, pectate lyases, polygalacturonase s and xylanases are proposed to exhibit plant cell wall-degrading activity (150). Deletion of the type II secretion system may be too detrimental to the bacterium; however, deletion of a number of the delivered cell wall-degrading enzymes (eg. polygalacturnate lyase, -amylase, or endoglucconase) may provide a desirable pathogenicity-attenuated phe notype. A putative type IV secretion system in X. euvesicatoria has been shown to have homology to the Icm/Dot syst em of human pathogens (20). The essential role of Icm/Dot type IV secretion system of Legionella species may suggest a possible role in virulence in Xanthomonas as well (20, 150). The type IV pilus is also thought to be involved in movement by retraction to medi ate adhesion to plant tissue (107). Inactivation or over expression of the quorum sensing auto-inducers may also provide an alteration in pathogenicity. Several gene s encoding this system that are found in X. euvesicatoria ( rpfA to H ) (150). Diffusible signal fact ors (DSFs) have been shown to be involved in regulation of the synthesis of extracellu lar enzymes, exopolysaccharides and cyclic glucans (98, 163). The genes mentioned here along with many othe rs may provide optimal attenuation for our system. Another possibility may be to create deletions in multiple pathogenicity related genes (as discussed with hpaB and hpaC ) to determine their combined mutant phenotypes. This may be useful as a tweaking tool to create the optimal level of colo nization and infection of the host plant. One hurdle to overcome concerning ch aracterization of thes e genes is functional redundancy. Mutations in most effector gene s do not show significan t effects on bacterial virulence when deleted, presumably because of functional redundancy of some effectors (9, 24, 31, 78). Overall, there is a great deal of potenti al for utilization of these pathogenicity related genes for creation of a pathogenicity-a ttenuated biological control agent.

PAGE 68

68 Table 3-1. Bacterial strains and plasmids used in this study Strain or plasmid Relevant characte ristics Source or reference* Xanthomonas euvesicatoria E3-1 NalRSmR (61) 91-106 (155) ME-90 RifRKanR (155) X. perforans 91-118 RifR (155) 91-118:: opgH OpgHRifR (101) 91-118:: avrBs2 AvrBs2RifRCmR This study 91-118:: xopA XopARifR This study 91-118:: xopD XopDRifRCmR This study 91-118:: hpaA HpaARifRCmR This study 91-118:: hpaB HpaBRifRCmR This study 91-118:: hpaC HpaCRifRCmR This study 91-118:: gumD GumDRifR This study Escherichia coli DH5 Frec A BRL C2110 NalR BRL PIR Host for pOK1; SpR oriR6K K2 replicon UB Plasmids pBluescript-KS+ Phagemid, pUC derivative; AmpR Stratagene pLAFR3 TcR rlx+ RK2 replicon BJS pRK2013 KmR tra+ mob+ (28) pOK1 Suicide vector; SacB (63) pOK1: opgH OpgHSmR This study pOK1: avrBs2 AvrBs2SmRCmR This study pOK1: xopA XopASmR This study pOK1: xopD XopDSmRCmR This study pOK1: hpaA HpaASmRCmR This study pOK1: hpaB HpaBSmRCmR This study pOK1: hpaC HpaCSmRCmR This study pOK1: gumD GumDSmR This study BRL, Bethesda Research Labor atories, Gaithersburg, MD; Stra tagene, Stratagene Inc., La Jolla, CA; BJS, B. J. Staskawicz, University of California, Berkeley, CA; UB, U. Bonas, MartinLuther-Universitt, Halle, Germany.

PAGE 69

69 Table 3-2. PCR primers used in hpaA hpaB hpaC xopA xopD avrBs2 and gumD analyses for genetic manipulations Gene primer name restriction site Primer sequence length GC (%) Tm (C) hpaA hpaAF Hin DIII AAGCTT GCTCAAGCTGGTGGTG 22 54.5 hpaAR Hin DIII AAGCTT ATCTAATCGTGCGCCTGC 24 50.0 56.0 hpaA2F ACGCAAACGAGCAGGAAG 18 55.6 hpaA2R AGCAGGATCAGTGGAAGCAG 20 55.0 55.0 hpaADF Sma I CCCGGG GTTTGGCTTCGATCTCTTCCTGCTC 31 61.3 hpaADR Sma I CCCGGG ATCTCCTGCTTCCACTGATCCTGCT 31 61.3 54.0 hpaB hpaBF Hin DIII GGGAAGCTT GCGACGCTGCGCGACA 26 65.4 hpaBR Hin DIII GGGGAATTC AAGGCTGCCATGGAGGAGGG 29 62.1 60.0 hpaB2F CAGAAGTCCATGAACAACAAGATCACG 27 44.4 hpaB2R ATCTCCCGCCAAACCTGTATCG 22 54.5 57 hpaC hpaCF GGCATCGAGATCGCCCAG 18 66.7 hpaCR CGCATACCGCAACCGCAG 18 66.7 58.5 hpaC2F GATAGCCAGCCACGCTTCCC 20 61.5 hpaC2R CGCACAGCTCGCGCTTCC 18 61.7 58.5 hpaCDF Hin DIII GGGAAGCTT GACAGATTGCGACCGAGTGGATAC 33 54.5 hpaCDR Hin DIII GGGAAGCTT GATACGTAAGGGTGGGTCGGTTTG 33 54.5 61& 67 xopA XopAF Hin DIII GGGAAGCTT TGCTGGAAGAGGAAAAGCG 28 53.6 XopAR Hin DIII GGGGAATTC AATCCGCGCGTGCGA 24 62.5 52& 60 XopA2F GAGAGGCTGAGGCTAGT 17 56.0 XopA2R TCATTGAATACGTCGCACC 22 57.0 57.0 XopA3F AAGTGGATAACGGCAGTGAG 23 56.0 XopA3R CGGAAAGCGACACAGCAG 18 57.2 54.5 xopD xopDF TGCTGCCTTTTTGATGGAC 19 47.4 xopDR TCCTGCCAACCCTACTTTAC 20 50.0 52.0 xopD2F TCCAAAAAGCAAGCCCAC 18 52.6 xopD2R GACGAGCAATGACCAATGAG 20 55.4 52.0 xopD3F GAGCCAACTTCAGAATGCG 19 55.2 xopD2R GACGAGCAATGACCAATGAG 20 55.4 55.0 xopDDF Hin DIII CCCAAGCTT CTGAAATCACTGCTTCACCCAGAC 33 51.5 xopDDR Hin DIII CCCAAGCTT GGTTCTTCCTATTCGTCCCTGTTC 33 51.5 59& 66 avrBs2D avrBs2F ATCGCCCGCATCGCCTTC 18 59.5 avrBs2R CACGCAGTCGCCTCCACC 18 61.7 60.0 avrBs2DD2F Sal I GTCGAC CTCGTAGGCATGATCGATGGAC 28 57.1 avrBs2DD2R Sal I GTCGAC GGAAACTACGTCAAGACCGACC 28 57.1 53& 59 gumD gumDF TCGTTCCTCTTCGTCGCAGC 20 60 gumDR TCCCGTATGTTTCGGGCTCCT 20 60 60

PAGE 70

70 Table 3-3. List of attenua tion candidate genes and thei r published, corresponding mutant phenotype Candidate Genes Organism Phenotype Reference opgHxcv X. perforans moderate reduction in in planta growth. Minsavage et al. 2003 hpaA X. euvesicatoria moderate reduction in in planta growth. Huguet et al. 1998 hpaB X. euvesicatoria moderate reduction in in planta growth. Bttner et al. 2004 hpaC X. euvesicatoria significant reduction in in planta growth. Bttner et al. 2005 xopA X. euvesicatoria moderate reduction in in planta growth. Nol et al. 2001 & Nol et al. 2002 xopD X. euvesicatoria no reduction in in planta growth. Nol et al. 2002 avrBs2 X. euvesicatoria Significant reduction in in planta growth. Kearney and Staskawicz 1990 gumD X. c. campestris delayed disease symptom development. Chou et al. 1997

PAGE 71

71 Table 3-4. Homology of X. perforans genes chosen to be deleted to create pathogenicityattenuated mutants to genes of ot her closely related xanthomonads % Homology (AA) Xanthomonads HpaAHpaB HpaC AvrBs2 XopA XopD X. euvesicatoria (85-10)a 98 100 99 98 100 85 X. oryzae. pv. oryzicola 88 94 88 88 X. oryzae pv. oryzae (KACC10331) 87 95 89 89 X. axonopodis pv. citri (306) 75 93 86 95 X. campestris pv. campestris (ATCC 33913) 75 85 55 76 74 X. campestris pv. glycines 75 93 87 47 a Strains are designated in parenthesis

PAGE 72

72 Table 3-5. In planta growth and aggressiveness of X. perforans strain 91-118 mutants as measured by area under the population progre ss curve (AUPPC) and percent disease severity, respectively, following inocul ation of Bonny Best tomato plants Strain X. perforans (91-118)118.1 ac38.7 a 91-118 opgH 98.5c 12.3c 91-118 xopA 105.5b 25.7 b 91-118 hpaB 116.4a 36.3a 91-118 hpaC 104.4bc33.7a 91-118: avrBs2 113.7a38.0a 91-118 gumD 106.4b 13.7 c Growth curve [AUPPC]aDisease severity [% disease severity]b a AUPPC from the growth curve of in ternal wt and mutant strains of X. perforans inoculated at 5 x 106 CFU/mL over a 120 h period. b Percent disease severity 14 days after dip inoculation of each bacterium at 5 x 106 CFU/mL amended with 0.025% Silwet L-77. c Values followed by the same letter are not significantly different based on Waller-Duncan multiple range test ( P = 0.05).

PAGE 73

73 Table 3-6. Growth room in planta internal and phyllosphere an tagonism experiments measuring X. euvesicatoria strain E3-1 populations when co-inoculated with water, X. euvesicatoria strain E3-1 or wt and mutants of X. perforans strain 91-118 measured as area under the population progress curve (AUPPC) Strain water control115.3 abX. euvesicatoria (91-106)101.7b 120.8 a X. perforans (91-118)67.0d 69.0 c 91-118 opgH 86.7c 79.1 b 91-118 xopA 87.2c 84.7 b 91-118 hpaB 89.2c91-118 hpaC 94.6bc91-118: avrBs2 95.7bc91-118 gumD 86.7c Internal antagonism [AUPPC]aPhyllosphere antagonism [AUPPC] a AUPPC from antagonism assay over a 96 h period based on recove red populations of X. euvesicatoria strain E3-1. b Values followed by the same letter are not significantly different based on Waller-Duncan multiple range test (P = 0.05).

PAGE 74

74 pGEM: hpaA hpaA Hin DIII Hin DIII Sal I Sal I pGEM: hpaA + CMR CmRSal I SalISpeI Sal I SpeI pTOPO:avrBs2D5 avrBs2DF Sal I SpeI pTOPO:avrBs2D3 avrBs2DF pTOPO:avBs2D5 + avrBs2D3 Sal I Sal I SpeI pTOPO: avrBs2 + CMR avrBs2DF SpeI pGEM: hpaA hpaA Hin DIII Sal ISalI Hin DIII avrBs2DF avrBs2DF CmR avrBs2DFA. B. Figure 3-1. Illustration of deletion constructions. A. Deletion strategy for avrBs2 B. Deletion strategy for hpaA, hpaC xopA and xopD pGEM: hpaA pGEM: hpaA CMR CMR avrBs2DR avrBs2DR avrBs2DR pTOPO:avrBs2DR pTOPO:avrBs2DF pTOPO:avrBs2DF + avrBs2DR pTOPO: avrBs2 + CMR

PAGE 75

75 2 3 4 5 6 7 8 024487296120 Time after inoculation (h)Log10 CFU/cm2 of leaf tissue Figure 3-2. In planta growth of wild-type (wt) and mutant X. perforans strains. Plants were inoculated at 5 x 105 CFU/mL of 91-118:: opgH ( ), 91-118:: hpaB ( ), 91-118:: hpaC ( ), 91-118:: xopA ( x), 91-118:: gumD ( ), 91-118:: avrBs2 ( ), wild-type 91-118 ( ) in tomato genotype Bonny Best Error bars indicate the standard error.

PAGE 76

76 Figure 3-3. Disease severity on Bonny Best leaf lets 2 weeks after dip inoculation (5 106 CFU/mL + 0.025% Silwet L-77) with X. perforans strains. Top: wt 91-118 (left), 91-118:: xopA (left center) and 91-118:: opgH, (right center) and 91-118:: hpaB (right). Bottom: 91-118:: hpaC (left), 91-118:: avrBs2 (center) and 91-118:: gumD (right).

PAGE 77

77 Pepper [ECW-20R] Tomato [VF36 + Bs2] HR HRHR 91-118:: avrBs2 91-118 91-118:: avrBs2 91-118 strain Pepper [ECW-20R] Tomato [VF36 + Bs2] HR HRHR 91-118:: avrBs2 91-118 91-118:: avrBs2 91-118 strain Figure 3-4. Phenotype in leav es of Bs2 transgenic tomato VS36 and pepper (ECW-20R) inoculated with 5 x 109 CFU/mL of X. perforans strains 91-118 (left) and 91-118: avrBs2 (right). Phenotypes we re recorded 24 h after inoculation. Browning of the tissue is associ ated with a hypersensitiv e response (resistance).

PAGE 78

78 2 3 4 5 6 7 8 024487296 Time after inoculation (h)Log10 CFU/cm2 of leaf tissue Figure 3-5. Growth room intern al antagonism assay measuring X. euvesicatoria strain E3-1 in leaflets. Plants were infiltrated with 5 x 107 CFU/mL of 91-118:: opgH ( ), 91-118:: hpaB ( ), 91-118:: hpaC ( ), 91-118:: xopA ( x), 91-118:: gumD ( ), 91-118:: avrBs2 ( ), wild-type 91-118 ( ), X. euvesicatoria strain 91-106 ( ) and water ( ), followed 18 h later by 5 x 106 CFU/mL of E3-1 in tomato genotype Bonny Best. Error bars indicate the standard error.

PAGE 79

79 2 3 4 5 6 7 8 9 024487296 Time after E3-1 inoculation (h)Log10 CFU/cm2 of leaf tissue Figure 3-6. Growth room phyllosphe re antagonism assay measuring X. euvesicatoria strain E3-1 in leaflets. Plants were dip i noculated with suspensions of 5 x 107 CFU/mL (amended with 0.025% Silwet L-77) of: 91-118:: opgH ( ), 91-118:: xopA ( x), wild-type 91-118 ( ) and water ( ), followed 7 d later by spray inoculation of 5 x 107 CFU/mL of E3-1 on tomato genotype Bonny Best. Error bars indicate the standard error.

PAGE 80

80 CHAPTER 4 EVALUATION OF XANTHOMONAS PERFORANS MUTANTS IN CONTROLLING X. EUVESICATORIA IN GREENHOUSE AND IN THE FIELD Bacterial spot of tomato is incited by four Xanthomonas species; X euvesicatoria X vesicatoria X perforans and X gardneri The first three bacterial species were previously known as tomato races T1, T2 and T3, respecti vely, based on their reaction on three tomato genotypes: Hawaii 7998 (H7998), Hawaii 7981 (H7981) and Bonny Best (71, 72, 141). X gardneri has only been found in Yugoslavia, Costa Rica and Brazil (10, 122, 144). Control of bacterial spot of tomato is difficult when high te mperatures and high moisture conditions exist. The disease has been demonstr ated to cause significant damage to the crop resulting in major losses. Pohronezny and Vo lin (118) estimated as high as 50% loss of marketable fruit due to bacterial spot on to matoes. There are currently no commercially available tomato varieties resistant to bacterial spot. Scott and Jones (1 35) identified significant resistance in H7998 in which X. euvesicatoria strains induce an HR. In 1993 Whalen et al. (166) determined that X. euvesicatoria strains carry the avirulence gene avrRxv which induces an hypersensitive response (HR) on the genotype H 7998 carrying the corresponding resistance gene Rxv ; however, X. perforans T3 strains carry avrXv3 which induces an HR in H7981 containing the resistance gene Xv3 (100). Astua-Monge (4) characterized avrXv3 and found it to elicit an HR in some tomato and pepper varie ties. In 2000 a new avirulence gene avrXv4 was described in X perforans strains based on an HR in tomato genotype LA716 ( Lycopersicum pinnellii ) carrying the Xv4 resistance gene (3, 4). X. perforans strains carrying this new avirulence gene ( avrXv4 ) and a non-functional AvrXv3 are de signated as tomato race 4. Bactericides, such as fixed coppers and stre ptomycin, have provided the primary means of chemical control (90, 142, 143); however, strept omycin-resistant mutants and copper-tolerant strains became prevalent (143). Marco and Stall (90) reported widespread emergence of copper-

PAGE 81

81 tolerant X. euvesicatoria strains and that addition of the f ungicide mancozeb to copper sprays improved disease control caused by copper-tolerant strains (25, 90). Chemical control alone is insufficient to control the disease under optimal conditions for the pathogen. Additionally, the use of copper compounds led to soil cont amination in some instances (79). Recently, there has been increased interest in integrated biological control strategies for bacterial diseases, which are difficult to control with conventional management practices. Some success has been achieved in th is area through empirical selec tion of biocontrol agents, as indicated by the commercializ ation of the products AgriphageTM, a mixture of bacteriophages for control of bacterial spot of tomato (46), GalltrolTM for control of crown gall, and BlightBanTM A506 for control of fire blight and frost in jury (89). However, achieving success using biocontrol agents for many bacterial diseases has b een difficult. This failure may in part be due to the very narrow focus on the almost exclusiv e use of nonpathogenic, saprophytic bacteria as biocontrol agents. While our understanding of the ecology of nonpathogenic saprophytes is increasing, our knowledge is limited to labor-int ensive protocols for identifying potential biocontrol agents. New integrated biological co ntrol strategies are currently being sought including the use of bacterioci ns, attenuated plant pathogens a nd/or bacteriophages (28, 32, 46, 61, 67, 102, 108, 168) as part of an integr ated biological c ontrol strategy. One recent approach for biologi cal control has been the use of bacteriocins (77, 152). Bacteriocins are substances produced by bacteria that are inhibitory or harmful toward only closely related bacteria (125). Bacteriocins and bacteriocin-like compounds encompass an array of structurally different s ubstances including enzyme inhi bition, nuclease activity and pore formation in cell membranes (125, 129, 128). Bacteriocins produced by Escherichia coli and several Gram-positive bacterial species have been extensively characterized (65, 125, 147). For

PAGE 82

82 bacterial spot of tomato at least three distinct bacteriocin-like activitie s (BcnA, BcnB and BcnC) were identified in X. perforans that are antagonistic toward X. euvesicatoria strains (154). Hert et al. previously (61) demonstr ated that two of these bacterio cins previously discovered in X. perforans (BcnA and BcnC) (154) could effectively control X euvesicatoria populations in greenhouse and field experiments. Although early attempts for biocontrol of pl ant diseases using bacteriocin-producing strains were made, few have been implemented (158, 159). For bacterial spot of tomato, Liu (88) conducted biological control studies ut ilizing a non-pathogenic bacteriocin-producing X. perforans strain to control disease incited by X. euvesicatoria strains. The non-pathogenic strain reduced bacterial spot disease incidence and se verity by 10 to 15 percent in the field when applied prophylactically when compared to X. euvesicatoria alone; however, these levels were still unacceptible levels of control (40% disease) (88). For Ralstonia solanacearum efforts to obtain a biological control strategy utilizing bacteriocin-producing non-pathogenic hrpmutants gave low to moderate levels of control of wild-type (wt) R. solanacearum (152). However, control using a partially pathogenic hrp mutant ( hrcV-), which is capable of higher levels of colo nization of the root and stem tissue, achieved better control (49). Research into colonization has been c onducted to understand the possible relationship between invasion efficiency of th e biocontrol agent and it s ability for disease control. Etchebar et al. (41) suggested that there was a positiv e correlation between colonization of the xylem by the hrp mutant and the level of control of the wt R. solanacearum Based on previous studies in which non-pa thogenic strains provided low levels of biological control (41, 49, 88), it was hypothesized that moderate invasion by the biological control agent using a par tially pathogenic bacterio cin-producing strain of X. perforans rather

PAGE 83

83 than a non-pathogenic strain may increase the effici ency of control under field conditions. These mutants may colonize the leaf tissue internally more effectively than non-pathogenic strains and this could potentially result in more effective biological control. A known X. perforans mutant gene opgHXcv was selected to create the pathogenicity -attenuated mutant ( 101). OpgH mutants have a reduced disease severity and growth curve in susceptible tissue and delayed avirulent HR phenotype in resistant plant tissue. The objective of this study wa s to evaluate the ability of a pathogenically-attenuated bacteriocin-producing strain (91-118:: opgH bcnB ) to reduce the populations and of the disease caused by bacteriocin-sensitive strains of X. euvesicatoria Previous research showed (61) that deletion of BcnB produces lower recovery of sensitive X. euvesicatoria strains than wt X perforans when co-inoculated with X euvesicatoria Materials and Methods Bacterial Strains, Plasmids and Culture Conditions Strains of X. perforans and X. euvesicatoria were grown on nutrient agar (NA) medium (Difco Laboratories, Detroit, MI) at 28 C (Table 4-1). Strains of E. coli were grown on LuriaBertani (LB) medium at 37 C (97). All strains were stored in 20% glycerol in sterile tap water at C. Bacterial cultures for plant inoculations were grown in nutrient broth (NB) (Difco Laboratories, Detroit, MI) for 18 h at 28C w ith shaking (100 rpm). Cells were pelleted by centrifugation (4,000 g 15 min) and resuspended in steril e tap water. Bacterial suspensions were standardized to an op tical density at 600 nm (OD600) = 0.3 (5 108 CFU/mL) with a Spectronic 20 spectrophotometer (Spectronic UNICAM, Rochester, NY) and subsequently diluted in sterile tap water to appropriate cell densities for individual experiments. Antibiotics were used to maintain selection for resist ance markers at the following concentrations:

PAGE 84

84 tetracycline (Tc) 12.5 g/mL; rifampicin (Rif) 100 g/mL; spectinomycin (Sp) 50 g/mL; kanamycin (Km) 50 g/mL; chloramphenicol (Cm) 34 g/mL; streptomycin (Sm) 200 g/mL; and nalidixic acid (Nal) 50 g/mL. Generation of the 91-118:: opgH bcnB Attenuation Mutant Triparental matings were performed using E. coli DH5 containing pRK2013Km as the helper plasmid (Table 4-1), E. coli DH5 containing pXV442-255 (pXV442Tc with insertion of a KmR cassette for inactivation of Bc nB) as the donor and 91-118:: opgH as the recipient (Recipient received from Gerald Minsavage). Ma rker exchange was achieved using standard methods (134). The candidate colonies were screen ed for loss of BCN activity and confirmed for insertion by Southern hybridizat ion (using subclone BcnB as th e probe) and PCR (with primers BCN-1 and BCN-2) (61). Plant Materials Seeds of tomato (Lycopersicon esculentum) c v. Bonny Best were planted in Plugmix (W. R. Grace & Co., Cambridge, MA). After 2 weeks, the emerged seedlings were transferred to Metromix 300 (W. R. Grace & Co., Cambridge, MA ) in 10-cm plastic pots. Seedlings were grown in the greenhouse at temperatur es ranging from 25 to 35C. Growth Room Growth Curve Assays Growth room assays were conducted to compare the growth curves of 91-118:: opgH and 91-118:: opgH bcnB mutants with the wt parent strain 91-118. Strains were grown in NB for 18 h, harvested by centrifugation and resuspended in st erile tap water. Strains were inoculated at 3 105 CFU/mL into leaflets of 6-week-old tomato seedlings. Leaflets were infiltrated (15 leaflets per strain) using a hypodermic syringe and needle, as describe d previously (68). Following inoculation plants were kept at 24C to 28C. Three samples were taken for each

PAGE 85

85 treatment every 24 h for 5 d. Bacterial popul ations were quantified by macerating 1-cm2 leaf disks in 1 mL sterile tap water and diluti on plating onto NA medium amended with the appropriate antibiotic. Plates we re incubated at 28C and colonies were counted after 48 to 72 h. Population data were log10 transformed and standard errors were determined. The overall growth curve was determined by calculating the area u nder the population prog ress curve (AUPPC). The AUPPC is a modification of the area under the disease progress curve (AUDPC) which has been used to analyze disease prog ress (136): standardized AUPPC = [(xi + xi 1)/2](ti ti 1), where x is population density in log10 CFU per cm2 and t is time in hours. The AUPPC values for the strains were compared by analysis of variance and subsequent separation of sample means by Waller-Duncan multiple range test using SAS version 9.0 (SAS Inc., Cary, NC). Each experiment was conducted three times. Greenhouse Disease Severity Assay Greenhouse disease severity assays were conducted to determine the effect of X. perforans mutant and wt 91-118 strains on symptom developmen t. In each test, four young (four-true-leaf stage) plants were inoculated with each st rain by dipping into ster ile tap water containing bacterial suspensions of 91-118, 91-118:: bcnB 91-118:: opgH 91-118:: opgH bcnB and X. euvesicatoria strain 91-106 (3 106 CFU/mL of bacteria ame nded with 0.025% Silwet L-77 (Loveland Industries, Inc., Greeley, CO)) for 15 s. Plants were maintained in the greenhouse during the evaluation period. The pl ants were assessed for disease severity 14 to 21 days after inoculation. Disease assessments for wt and attenuated mutant candidates of X. perforans strains were made based on leaf and stem ratings comp iled from three separate greenhouse inoculation tests.

PAGE 86

86 Growth Room Antagonism Assay Antagonism assays were performed to determine the effect of wt and mutant X. perforans strains on the bacteriocin sensitive X. euvesicatoria strain, E3-1. Internal and external (phyllosphere) populations were separately eval uated using two antagonism assay techniques. Internal antagonism assays. Strains were grown in NB for 18 h, harvested by centrifugation and resuspended in sterile tap water. X. perforans and X. euvesicatoria strains were inoculated at 5 107 CFU/mL and 5 106 CFU/mL, respectively. Six-week-old seedlings of the tomato cultigen Florida 47 were inocul ated (15 leaflets per strain) using a hypodermic syringe as described previously (68). The mutant and wt X. perforans strains were inoculated into leaflets by infiltration 12 h prior to inoc ulation with the sensitive strain (E3-1). Each treatment consisted of three replications. Fo llowing inoculation, plants were incubated at 24 C to 28 C. In order to determine populations of the sensitive strain (E3-1SmNal) in leaflets, 1-cm2 leaf disks were removed from inoculated areas, macerated in 1 mL sterile tap water and dilution plated onto nutrient agar amended with the approp riate antibiotic. Samples were assayed at 24 h intervals for 96 h. Each experiment was conduc ted three times. Population data were log transformed and standard errors were determin ed. AUPPC values (calculated as described above) were compared by analysis of variance and subsequent separation of sample means by Waller-Duncan multiple range test using SAS version 9.0 (SAS Inc., Cary, NC). Each experiment was conducted three times. Phyllosphere antagonism assays. Growth room phyllosphere antagonism assays were conducted to determine if the gene deletions affect ed the levels of antago nism toward external leaf populations of X. euvesicatoria strain E3-1 by comparing th e antagonistic ability of the mutants with the parent strain 91-118. Strain s were grown in NB for 18 h, harvested by

PAGE 87

87 centrifugation and resuspended in sterile tap water. Six-week-old Bonny Best tomato seedlings were dipped into 5 107 CFU/mL suspension of the wt and mutant strains of X. perforans amended with Silwet L-77 (0.025%) 7 days pr ior to spray inoculation with a 5 107 CFU/mL suspension of X. euvesicatoria strain E3-1. Following spray i noculation, plants were incubated at 24 C to 28 C. Leaf tissue was sampled every 24 h for 96 h for quantification of E3-1 populations. Three leaflets were taken at each ti me point. Each leafle t was weighed, placed in a polyethylene bag containing 10 mL of sterile tap water shaken on a Wrist Action shaker (Burrel Co., Oakland, CA) and shaken vigorously for 20 mi n. The leaf-wash was then dilution plated on NANalSm to selectively determine th e bacterial population of E3 -1. Population data were analyzed following log10 transformation and standard errors were determined. AUPPC values (calculated as described above) were compar ed by analysis of variance and subsequent separation of sample means by Waller-Duncan mu ltiple range test using SAS version 9.0 (SAS Inc., Cary, NC). Each experiment was conducted three times. Field Experiments Field plot design. The field experiments were set up in a completely randomized block design consisting of four replicat ions. Raised beds were 0.91 m wide and were covered with black plastic mulch. Plots were arranged in paired beds that were 1.83 m from center to center and each set of paired beds was 7.32 meters apart. Plots within the paired beds were spaced 6.1 m apart. Each plot containing 20 plants were spaced 457 cm apart. Bacterial strains, inoculum producti on, inoculation and plant material. Field experiments were performed using X. perforans mutant 91-118:: opgH bcnB and X. euvesicatoria strain E31 to evaluate antagonism of those strains to the X. euvesicatoria strain E3-1 (Table 4-1). Strains were grown in NB for 24 h, harvested by centrif ugation and resuspended in sterile tap water.

PAGE 88

88 The bacterial suspensions were adjusted to 5 107 CFU/mL and the surfactant Silwet L-77 was added to a final concentration of 0.025%. Pl ants were dipped into suspensions of the 91-118:: opgH bcnB strain 24 h prior to spray inoculation with a 5 107 CFU/mL suspension of E3-1. Six-week-old seedlings of the tomato genotype Florida 47 (Asgrow, Oxnard, CA) were used in all experiments. Incidence of strains in lesions. In 2004, field experiments were conducted at the North Florida Research and Education Center (NFREC) in Quincy, FL to evaluate recovery of wt X. euvesicatoria strain E3-1 and 91-118:: opgH bcnB from symptomatic leaf tissue. The experiment consisted of six treatments: (1) uni noculated control; (2) E3 -1 + growers standard; (3) E3-1 alone; (4) 91-118:: opgH bcnB alone; (5) E3-1 and 91-118:: opgH bcnB (applied biweekly); and (6) E3-1 and 91-118:: opgH bcnB (applied weekly). Plants in the grower standard was treated on weekly rotations of acibenzolar-S-methyl (0.055 g/L) (Actigard 50WG; Syngenta Crop Protection Inc ., Greensborough, NC) or copper hydroxide (3.6 g/L) (Kocide 2000; Griffin Corp., Valdosta, GA) plus manco zeb (2.5 g/L) (Manzate 75DF; Griffin Corp., Valdosta, GA) every two weeks. Symptomatic leaf tissue was collected every 2 weeks beginning 35 days after transplanting. Ten to tw enty leaflets were randomly collected in each plot and bacteria were isolated from thirty le sions. Individual lesi ons were macerated in 75 L of sterile deionized water and the suspensi ons were streaked on NA amended with 134 g/mL of pentachloronitr obenzene (PCNB) (126) and 50 g/mL of cycloheximide to eliminate fungal contaminants from the samples. Individual coloni es were plated onto two media to differentiate X. perforans and X. euvesicatoria (NA amended with the appropriate antibiotics for 91-118:: opgH bcnBRif Km and E3-1Nal Sm). The overall strain incidence was expressed by calculating the area under the incidence progress curve (AUIPC). The AUIPC is a modification

PAGE 89

89 of AUDPC: standardized AUIPC = [(xi + xi 1)/2](ti ti -1), where x is the arcsin of the percent recovery (to normalize the data) and t is days after inoculation. The AUIPC values were compared by analysis of variance and subseque nt separation of sample means by Waller-Duncan multiple range test using SAS version 9.0 (SAS Inc., Cary, NC). Incidence of phyllosphere populations. In 2005, field experiment s were conducted at two locations (NFREC and the Citra Research Fa rm) to evaluate recovery of E3-1 and 91-118:: opgH bcnB from the surface of asymptomatic leaf tissue. Asymptomatic leaf tissue was sampled every 2 weeks beginning ~20 days afte r transplanting (DAT). Seven leaflets were collected from each plot. Each sample was weighed, placed into a polyethylene bag (Becton Dickinson, Rutherford, New Jersey) containing 5 to 10 mL of steril e tap water and shaken at 200 rpm for 30 to 45 minutes. Serial ten-fold dilu tions were made in sterile tap water. A 50 l aliquot of each dilution was plated two NA plates, one amended with 134 g/mL of PCNB and/or 50 g/mL of cycloheximide with addition of an tibiotics for selection of E3-1 (Sm and Nal) and the second for the 91-118:: opgH bcnB mutant (Rif and Km). After incubation at 28 C for 4-5 days, colonies typical of Xanthomonas were counted and populations were calculated. Data was analyzed for statistical significance by using the AUIPC. AUIPC values (calculated as described above) were compared by analysis of variance and subsequent separation of sample means by Waller-Duncan mu ltiple range test using SAS version 9.0 (SAS Inc., Cary, NC). Each experiment was conducted three times. Results 91-118 and Mutants Reduce E3-1 in Growth Room and in the Greenhouse Growth room experiments were conducted to determine the effects of the mutations to grow in planta The 91-118:: opgH bcnB mutant reached populations 1 to 1.5 log units lower

PAGE 90

90 than wt 91-118 (Figure 4-1). The 91-118 populat ion exhibited a normal gr owth curve during the 120-h sampling period and, based on the AUPPC re sults (Table 4-2), and had significantly higher populations over time than both 91-118:: opgH mutants. The 91-118:: opgH bcnB mutant was not significantly different from 91-118:: opgH suggesting that there was no pronounced effect on growth curve based on the bcnB deletion. Greenhouse disease severity expe riments were conducted to determine the effects of the mutations on the ability of 91-118 to cause disease in planta (Figure 4-2). Disease severity incited by 91-118:: opgH bcnB and 91-118:: opgH strains (12% and 14 %, respectively) was significantly lower than that of 91-118 (39 %) (Table 4-2). Both internal (Figure 4-3) and phyllosphere (Figure 4-4) antagonism assays under growth room conditions were performed to determ ine the antagonistic activity of 91-118 and 91-118:: opgH bcnB and 91-118:: opgH strains toward the E3-1 strain. For both assays mutants 91-118:: opgH and 91-118:: opgH bcnB were moderately antagonistic, whereas wt 91-118 and 91-118:: bcnB provided the greatest reduction in E3-1 populations. The water control treatment consistently had significantly higher population levels than all other treatments (Figures 4-3 and 4-4 and Tabl e 4-3). Treatment of a wt X. euvesicatoria strain (91-106) prior to X. euvesicatoria strain (E3-1NalSm) reduced populations by ~0.5 log10 CFU/mL in both assays compared to the water control. Differences in the levels of antagonism was observed between internal and phyllosphere antagonism assays. Field Study The ability of the attenuated mutant, 91-118:: opgH bcnB to reduce E3-1 populations was assessed in the field. Controls included non -inoculated control, a E3-1 alone control and 91-118 alone control plot. In 2004, symptomatic leaf tissue was sampled at the NFREC in

PAGE 91

91 Quincy, FL and X. euvesicatoria populations were signifi cantly reduced by both 91-118:: opgH bcnB control treatments (weekly and two w eek application) (Figure 4-5). E3-1 was recovered from less than 5 percent of the samples in 91-118:: opgH bcnB plots compared to 26 percent from E3-1 alone (Table 4-4). In 2005 at Quincy (Figure 4-6) and Citra (Figure 4-7) phyllosphere populations were sampled. In both locations E3-1 populati ons were significantly reduced by the 91-118:: opgH bcnB mutant when applied weekly througho ut the growing season (Figures 4-6 & 4-7). In Quincy, E3-1 was recovered from 30% of the samples in plots where E3-1 was applied alone (Table 4-4). In the treatment where 91-118:: opgH bcnB was applied every two weeks, the frequency of recovery of E3-1 popula tions was not significantly different from plots where E3-1 was applied alone. Weekly application of 91-118:: opgH bcnB however, significantly reduced recovery of E3-1 populations compared to plots where the E3-1 was applied alone (approximately 65 % reduction) (Table 4-4). In Citra (2005), both 91-118:: opgH bcnB weekly and biweekly treatments significantly reduced recovery of E3-1 populations (Figure 4-7). The AUEPC of 91-118:: opgH bcnB weekly and biweekly treatments had significan tly reduced E3-1 incidence compared to the grower standard (37%) and E3-1 al one (54%) plots (Table 4-4). Discussion In this study, we sought to crea te a pathogenically attenuated X. perforans mutant to: (I) express two of the three previously described bacteriocins (based on previous field analysis (61) and (II) maintain itself at a level to maintain antagonism toward X. euvesicatoria strains while causing minimal disease. We decided to exam ine the previously described osmorelgulated periplasmic glucan gene, opgHXcv (101). In greenhouse experi ments the growth curve of both

PAGE 92

92 attenuated mutants 91-118:: opgH and 91-118:: opgH bcnB exhibited reduced growth curves compared to 91-118. The 91-118:: opgH bcnB mutant also caused significantly less disease than wt which was similar to a previously study by Minsavage et al. (101). The 91-118:: opgH still maintained significant levels of antagoni sm toward E3-1 populations as observed in the antagonism assay. Overall, mutants 91-118:: opgH and 91-118:: opgH bcnB were significantly effective in reducing X. euvesicatoria populations in the greenhouse. Interestingly, 91-118:: opgH and 91-118:: opgH bcnB were more effective in reducing external E3-1 populations than internal populations Lindemann et al. (86) previ ously concluded that there is a direct correlation between phyllosphere populations and occurrence of disease. The existence of this correlation together with the reduction in X. euvesicatoria phyllosphere populations during antagonism experiments suggest that a biological control strategy for X. euvesicatoria by 91-118:: opgH bcnB may be effective at reducing diseas e by reducing phyllosphere populations below the threshold level necessary to cause lesion development. In 2004, two hurricanes during th e season introduced high ex ternal populations of wt X. perforans which reduced both E3-1 and 91-118:: opgH bcnB populations. Although naturally occurring populations of X. perforans were introduced into the pl ots, early sampling data along with AUDPC data suggest that weekly a nd biweekly treatments with 91-118:: opgH bcnB significantly reduced X. euvesicatoria populations. Similar trends were also observed in our 2005 field data evaluating phyllosphere levels. In 2005 experiments 91-118:: opgH bcnB effectively reduced X. euvesicatoria populations by up to 85 percent. In both years, weekly application of 91-118:: opgH bcnB at 5 106 CFU/mL significantly reduced X. euvesicatoria populations compared to plots receiving X. euvesicatoria alone. In two of three experiments, biweekly application was found to significantly reduce E3-1 populations. This reduction in X.

PAGE 93

93 euvesicatoria populations is similar to prev ious field data using 91-118:: bcnB (61). Hert et al. (61) found that co-inoculation of 91-118:: bcnB and an X. euvesicatoria in the field yielded less than 5 percent recovery of the X. euvesicatoria strain over all seasons tested. These results suggest that the 91-118:: opgH bcnB attenuated mutant has potenti al as a biological control agent of reducing X. euvesicatoria populations. The reduction in X. euvesicatoria populations by 91-118:: opgH bcnB in the antagonism assays does appear to be less inhibitory compared to 91-118:: bcnB This suggests that bacteriocin expression was al so affected by the 91-118:: opgH mutation. The repeated treatment of 91-118:: opgH bcnB in the field experiments was sufficient for maintaining 91-118:: opgH bcnB levels and consequently sufficien t bacteriocin levels to suppress X. euvesicatoria populations. Weekly application of 91-118:: opgH bcnB was more effective in reducing X. euvesicatoria populations than previously observed in similar experiments using hrpX. perforans mutants (88). The 91-118:: opgH bcnB mutant appears to colonize within the leaf tissue more effectively than non-pathogenic strain s and is similarly to what was observed in a previous study with R. solanacearum hrcV mutants (41). Although the opgHXcv mutant was effective in suppressing X. euvesicatoria populations, there is potential for identifying other gene targ ets that can help impr ove biological control efficacy. Several other pathogenicity factors and associated genes have previously been described in X. euvesicatoria with other genes associated with the hrp system ( hpaA hpaB hpaC ), avirulence genes ( avrBs2 xopA xopD ) and pathogenicity factors ( gumD ) (17, 18, 63, 106, 105, 167). In the previous chapter (Chapter 3) we tested these pat hogenicity-associated genes for their phenotype in X. perforans and the potential use for cr eating further pathogenicityattenuated biocontrol agen ts. When several of the genes were mutated in X. perforans strain

PAGE 94

94 91-118 there was an associated reduction in pathogenicity and growth curve in planta These mutants or other pathogenicity associated ge nes may improve our pat hogenicity attenuated biological control model system by allowing bett er internalization and subsequent competition between X. euvesicatoria and X. perforans populations without detrimen tal effects to the plant. Sequencing of the X. euvesicatoria genome has provided signif icant new possibilities for developing pathogenicity attenu ated candidates. In 2005, Thie me et al (150) published the X. euvesicatoria genome sequence and estimated over 480 putative pathogenicity factors and associated genes. These genes were grouped into 6 categories: (I) secretion systems, (II) flagellum, (III) secreted proteins (via type III secretion system), (IV) detoxification, (V) surface structure and adhesion and (VI) quor um sensing. Genomic sequencing of bacteria provides an opportunity to exploit these ge nes for our utilization. Further research is needed to optimize this sy stem to create a weakly aggressive biological control agent that is as antagonistic as wild-typ e. Recent information from related bacteria and genomic sequences can be used to provide opport unities to improve our understanding of how pathogenic bacteria colonize and subsequently infect the host. Continued exploration of new innovative ideas will help us to util ize this knowledge in effective ways.

PAGE 95

95 Table 4-1. Bacterial strains and plasmids used in this study Strain or plasmid Relevant characteristics Source or reference* Xanthomonas euvesicatoria E3-1 NalRSmR (61) 91-106 (155) ME-90 RifRKanR (155) X. perforans 91-118 RifR (155) 91-118:: bcnB BcnBRifR KmR (61) 91-118:: opgH OpgHRifR (101) 91-118:: opgH bcnB OpgHBcnBRifR KmR This study Escherichia coli DH5 Frec A BRL C2110 Nalr BRL PIR Host for pOK1; SpR ori R6K RK2 replicon UB Plasmids pOK1 Suicide vector; SacB (63) pBluescript-KS+ Phagemid, pUC derivative; AmpR Stratagene pLAFR3 Tcr rlx+ RK2 replicon BJS pRK2013 helper plasmid; Kmr tra+ (28) BRL, Bethesda Research Laborator ies, Gaithersburg, MD; Stratagene Stratagene Inc., La Jolla, CA; BJS, B. J. Staskawicz, University of Ca lifornia, Berkeley, CA; UB, U. Bonas, MartinLuther-Universitt, Halle, Germany.

PAGE 96

96 Table 4-2. In planta growth and aggressiveness of X. perforans strain 91-118 mutants as measured by area under the population progre ss curve (AUPPC) and percent disease severity, respectively, following inocul ation of Bonny Best tomato plants Strain X. perforans (91-118)118.3a 38.7a 91-118 opgH 104.2b 12.3b 91-118 opgH bcnB 111.6b 14.1b Growth curve [AUPPC]aDisease severity [% disease severity]b a AUPPC from growth curve of internal X. perforans strains inoculated at 5 x 106 CFU/mL over a 120 h period. b Percent disease severity 14 days after dip inoculation of each bacterium at 5 x 106 CFU/mL amended with 0.025% Silwet L-77. c Values followed by the same letter are not significantly different based on Waller-Duncan multiple range test ( P = 0.05).

PAGE 97

97 Table 4-3. Growth room in planta internal and phyllosphere an tagonism experiments measuring X. euvesicatoria strain E3-1 populations when co-inoculated with water, X. euvesicatoria strain E3-1 or wt and mutants of X. perforans strain 91-118 measured as area under the population progress curve (AUPPC) Strain water control115.3 abX. euvesicatoria (91-106)101.7b120.8a X. perforans (91-118)67.0d69.0c 91-118 opgH 86.7c79.1b 91-118 opgH bcnB 90.8c86.3b Internal antagonism [AUPPC]aPhyllosphere antagonism [AUPPC] a AUPPC from antagonism assay over a 96 h period based on recove red populations of X. euvesicatoria strain E3-1. b Values followed by the same letter are not significantly different based on Waller-Duncan multiple range test ( P = 0.05).

PAGE 98

98 Table 4-4. Incidence and recovery of X. euvesicatoria strain E3-1 in the field when treated with X. perforans mutant strain 91-118:: opgH bcnB Incidence and recovery were measured by area under the incidence progre ss curve (AUIPC) and percent recovery, respectively % Recoveryb% Recovery % Recovery 0cc 0 425b13157b7 563.8a 21 1273a37757ab15 614.5a 26 1435a541430a 30 0c 0 446b592b2 181.0b 5 263b12890ab25 158.2b 4 260b8477b11 2005 Quincy Citra Uninoculated control 2004 Citra AUIPCaAUIPCAUIPC Treatment opgH bcnB (1) + E3-1fGrower standarddE3-1 alone opgH bcnB alone opgH bcnB (2) + E3-1e a AUIPC from fields evaluating E3-1 popu lations for each season for each treatment. b The % recovery is the average percent recovery of E3-1 populations for the season for each treatment. c Values followed by the same letter are not significantly different based on Waller-Duncan multiple range test (P = 0.05). d Grower standard plots were treated with Copper + Manzate and Actigard biweekly throughout the season. e The 91 118:: opgH bcnB plots were treated with opgH bcnB every two weeks throughout the season. f The 91 118:: opgH bcnB (1) + T1 plots were treated with opgH bcnB weekly throughout the season.

PAGE 99

99 2 3 4 5 6 7 8 024487296120 Time after inoculation (h)Log10 CFU/cm2 of leaf tissue Figure 4-1. In planta growth of wild-type and mutant X. perforans strains. Plants were infiltrated with 5 x 105 CFU/mL of 91-118:: bcnB ( ), 91-118 opgH ( ), 91118:: opgH bcnB ( ) and wt 91-118 ( ) in tomato genotype Bonny Best. Error bars indicate the standard error.

PAGE 100

100 Figure 4-2. Disease severity on Bonny Best leaflets 2 weeks after dip inoculation 5 x 106 CFU/mL + 0.025% Silwet L-77) with X. perforans strains wild-typ e 91-118 (left), 91-118 opgH (center) and 91-118:: opgH bcnB (right).

PAGE 101

101 2 3 4 5 6 7 8 024487296 Time after inoculation (h)Log10 CFU/cm2 of leaf tissue Figure 4-3. Growth room intern al antagonism assay measuring X. euvesicatoria strain E3-1 in leaflets. Plants were infiltrated with 5 x 107 CFU/mL of 91-118 bcnB ( ), 91-118 opgH ( ), 91-118:: opgH bcnB ( ), wild-type 91-118 ( ), wt X. euvesicatoria strain 91-106 ( ) and water ( ) followed 18 h later by 5 x 106 CFU/mL of E3-1 in tomato genotype Bonny Best. E rror bars indicate the standard error.

PAGE 102

102 2 3 4 5 6 7 8 9 024487296 Time after E3-1 inoculation (h)Log10 CFU/cm2 of leaf tissue Figure 4-4. Growth room phyllosp here antagonism assay measuring X. euvesicatoria strain E3-1 in leaflets. Plants were dip i noculated with suspensions of 5 x 107 CFU/mL (amended with 0.025% Silwet L-77) of 91-118 opgH ( ), 91-118:: opgH bcnB ( ), wild-type 91-118 ( ), or X. euvesicatoria strain 91-106 ( ) followed 7 d later by spray inoculation of 5 x 107 CFU/mL of E3-1 in tomato genotype Bonny Best. Error bars indicate the standard error.

PAGE 103

103 Uninoculated control0 20 40 60 80 100 3045607590105120Sample Time (DAT)Recovery of strain s (%) Grower standard0 20 40 60 80 100 3045607590105120Sample Time (DAT)Recovery of strain s (%) E3-1 alone0 20 40 60 80 100 3045607590105120Sample Time (DAT)Recovery of strain s (%) opgH bcnB alone0 20 40 60 80 100 3045607590105120Sample Time (DAT)Recovery of strain s (%) opgH bcnB (2 weeks) + E3-10 20 40 60 80 100 3045607590105120Sample Time (DAT)Recovery of strain s (%) opgH bcnB (weekly) + E3-10 20 40 60 80 100 3045607590105120Sample Time (DAT)Recovery of strain s (%) Figure 4-5. Quincy 2004 field expe riment: Percent recovery of X. perforans strain 91-118:: opgH bcnB ( ), native wild-type strains ( ) and X. euvesicatoria strain E31 ( ) from lesions from plants from th e following treatments: (1) uninoculated control; (2) E3-1 followed by grower standa rd (copper + manzate & actigard); (3) E31 alone; (4) 91-118:: opgH bcnB alone; (5) E3-1 followed by 91-118:: opgH bcnB applied every two weeks; and (6) E3-1 followed by 91-118:: opgH bcnB applied weekly. Sample times are indicated as da ys after transplantin g (DAT). Error bars indicate the standard error.

PAGE 104

104 Uninoculated Control0 20 40 60 80 100 1525354555Sample Time (DAT)Recovery of Strain s (%) Grower standard0 20 40 60 80 100 1525354555Sample Time (DAT)Recovery of Strain s (%) E3-1 alone0 20 40 60 80 100 1525354555Sample Time (DAT)Recovery of Strain s (%) pgH bcnB alone0 20 40 60 80 100 1525354555Sample Time (DAT)Recovery of Strain s (%) pgH bcnB (2 weeks) + E3-10 20 40 60 80 100 1525354555Sample Time (DAT)Recovery of Strain s (%) pgH bcnB (weekly) + E3-10 20 40 60 80 100 1525354555Sample Time (DAT)Recovery of Strain s (%) Figure 4-6. Quincy 2005 field expe riment: Percent recovery of X. perforans strain 91-118:: opgH bcnB ( )and X. euvesicatoria strain E3-1 ( ) from asymptomatic leaves from plants that received the fo llowing treatments: (1) uninoculated control; (2) E3-1 followed by grower standard (c opper + manzate & actigar d); (3) E3-1 alone; (4) 91-118:: opgH bcnB alone; (5) E3-1 followed by 91-118:: opgH bcnB applied every two weeks; and (6 ) E3-1 followed by 91-118:: opgH bcnB applied weekly. Sample times are indicated as days after tr ansplanting (DAT). Error bars indicate the standard error.

PAGE 105

105 Uninoculated Control0 20 40 60 80 100 1525354555Sample Time (DAT)Recovery of Strain s (%) Growers standard0 20 40 60 80 100 1525354555Sample Time (DAT)Recovery of Strain s (%) E3-1 alone0 20 40 60 80 100 1525354555Sample Time (DAT)Recovery of strain s (%) opgH bcnB0 20 40 60 80 100 1525354555Sample Time (DAT)Recovery of strain s (%) pgH bcnB (2 weeks) + E3-10 20 40 60 80 100 1525354555Sample Time (DAT)Recovery of strain s (%) opgH bcnB (weekly) + E3-10 20 40 60 80 100 1525354555 Sample Time (DAT)Recovery of strain s (%) Figure 4-7. Citra 2005 field expe riment: Percent recovery of X. perforans strain 91-118:: opgH bcnB ( )and X. euvesicatoria strain E3-1 ( ) from asymptomatic leaves from plants that received the fo llowing treatments: (1) uninoculated control; (2) E3-1 followed by grower standard (c opper + manzate & actigar d); (3) E3-1 alone; (4) 91-118:: opgH bcnB alone; (5) E3-1 followed by 91-118:: opgH bcnB applied every two weeks; and (6 ) E3-1 followed by 91-118:: opgH bcnB applied weekly. Sample times are indicated as days after tr ansplanting (DAT). Error bars indicate the standard error.

PAGE 106

106 CHAPTER 5 SUMMARY AND DISCUSSION Xanthomonas perforans strain 91-118 produces at least three different bacteriocin-like compounds (BcnA, BcnB, BcnC), antagonistic toward X. euvesicatoria strains (155). Hert et al. previously (61) demonstrated that two of these bacteriocins previously discovered in X. perforans (BcnA and BcnC) (154) c ould effectively control X. euvesicatoria populations in greenhouse and field experiments. The goal of this study was to evaluate a new biological control strategy utilizing pathogeni city-attenuated, bacteriocin-producing X. perforans strains for control of bacteriocin-sensitive strains of X. euvesicatoria The objectives of this study were: (I) to further characterize the b acteriocins associated with X. perforans (II) to identify and individually delete pathogenicity genes to create partially pa thogenic mutants of X. perforans ; and (III) to determine the ability for these pathogenicity-attenu ated mutant strains to antagonize X. euvesicatoria in vitro in planta in the greenhouse and under field conditions. For the first objectiv e bacteriocins of X. perforans partially characterized by Tudor-Nelson et al. (155) were further charac terized to determine their acti vities and possible functions. Disruption analysis has shown that BcnA is part of a multiple component fa mily of bacteriocins. The toxin and immunity function to BcnA were localized to ORFA and ORF5, respectively. Using predicted localization, based on bioinfor matics software SOSUI and PSORT, a model was created to represent the possible positions and ro les of each ORF. The model suggests four steps: (I): pre-BcnA delivery into the peri plasm chaperoned by ORF4; (II) Processing of preBcnA and delivery of the active BcnA outside of the cell by ORF2 a nd ORF3; (III) Entry of active BcnA into Xanthomonas cells (unknown); (IVa) BcnA suppressed by ORF5; and (IVb) BcnA inhibition (either in the periplasm or in the cytoplasm). The type 2 secretion system (T2SS) was also shown to be involved in the activity of BcnA. A X. euvesicatoria strain

PAGE 107

107 transconjugant containing a T2SS mutant (91:106:: xpsD ) and a BcnA expressing plasmid (pXV12.1) did not produce any detectible BcnA activ ity in plate assays. Further research is needed to confirm the involvement of the T2SS in BcnA secretion. Previously BcnB and BcnC were localized to a 5.9 kb and 1.7 kb fragment, respectively (61, 154). In this study BcnB was shown to be produced by an ORF with endoproteinase Arg-C homology, while BcnC was found to have homol ogy toward the extracellular metalloprotease family of genes. Disruption mutagenesis and pr otease assays confirmed that BcnB is in the endoproteinase Arg-C family of serine endoproteases and BcnC to be a metalloprotease. There is preliminary information that BcnB a nd BcnC are secreted via the T2SS. A X. euvesicatoria strain transconjugant contai ning a T2SS mutant (91:106:: xpsD ) and a BcnB or BcnC expressing plasmid (pL5.8, and pL5.1, respectively) did not pr oduce any detectible BcnB or BcnC activity in plate assays. Further research is needed to confirm the involvement of the T2SS in BcnB and BcnC secretion. The second objective was to identify pathoge nicity-related genes fo r disruption to create mutants of X. perforans with reduced virulence. The goal of this objective was to identify genes that could be mutated that would provide a dram atic reduction in disease symptoms while still maintaining the significant expression levels of BcnA and BcnC. Two mutants, 91-118:: opgH and 91-118:: gumD, stood out as strong candidates for further study. Both mutants had a pathogenicity-attenuated phenotype in disease severity and grow th rate experiments, while exhibiting high levels of antagonism toward the bacteriocin-sensitive X. euvesicatoria strains. Mutant 91-118:: xopA was the only mutant to exhibit intermediate levels of disease severity and reduced grow th rate, while 91-118:: hpaB, 91-118:: hpaC and 91-118:: avrBs2 were only slightly reduced. Antagonism assa ys revealed that these mutants (91-118:: xopA

PAGE 108

108 91-118:: hpaB, 91-118:: hpaC and 91-118:: avrBs2 ) were affected in antagonism toward X. euvesicatoria below what was observed by wild-type (w t) 91-118. Overall disease levels caused by these mutants were still too high to be a viab le pathogenicity-attenuate d mutant for use as a biocontrol agent. In the final objective one pathogenically attenuated X. perforans mutant was selected as the model attenuated mutant for furthe r studies. An osmoregulation gene, opgHXcv was chosen (101) and, based on previous data (61), combined with an additional mutation ( bcnB ) to create 91-118:: opgh bcnB In greenhouse experiments the growth rate and disease severity of both attenuated mutants 91-118:: opgH and 91-118:: opgH bcnB exhibited reduced growth rates compared to 91-118. The 91-118:: opgH still maintained significant levels of antagonism toward E3-1 populations seen in the an tagonism assay. Overall, mutants 91-118:: opgH and 91-118:: opgH bcnB were very effective in reducing X. euvesicatoria populations in the greenhouse. Interestingly, 91-118:: opgH and 91-118:: opgH bcnB were more effective at reducing external E3-1 populations th an internal populations. Lindemann et al. (86) previously concluded that there is a dire ct correlation between phyllosphe re populations and occurrence of disease. The existence of this corr elation together with the reduction in X. euvesicatoria phyllosphere populations during antagonism expe riments suggest that a biological control strategy for X. euvesicatoria by 91-118:: opgH bcnB may be effective at reducing disease by reducing phyllosphere populations below the threshold level necessary to cause lesion development. Field experiments using 91-118:: opgH bcnB as a potential biologi cal control agent were quite successful in reducing X. euvesicatoria populations. Week ly application of 91-118:: opgH bcnB consistently reduced X. euvesicatoria population by up to 85 percent and

PAGE 109

109 in two out of three experiments application ev ery two weeks was also found to significantly reduce E3-1 populations. This reduction in X. euvesicatoria populations is similar to previous field data using 91-118:: bcnB (61). Hert et al. (61) found co-inoculation of 91-118:: bcnB and an X. euvesicatoria in the field, yielded less than 5 percent rec overy of the X. euvesicatoria strain over all seasons tested. The 91-118:: opgH bcnB mutant appears to colonize within the leaf tissue more effectively than non-pa thogenic strains, similarly to R. solanacearum hrcV mutants (41), resulting in more effective biological control. These results suggest that the 91-118:: opgH bcnB attenuated mutant has potential as a biological control agent for reducing X. euvesicatoria populations. Further research is needed to optimize this system. Recent information from related bacteria and genomic sequences and functional ge nomics can be used to provide opportunities to improve our understanding of how pa thogenic bacteria colonize and s ubsequently infect the host. Continued exploration of new innova tive ideas will help us to uti lize this knowledge in effective ways. Bacterial spot disease is one of the most severe bacterial di seases affecting production of peppers and tomatoes in regions where they are grown (123). No single strategy has been successful for controlling this di sease. The findings of this st udy warrant further investigation into the development of a biol ogical control strategy for bact erial spot disease utilizing attenuated, bacteriocin expressing X perforans strains. Further research needs to be done to definitiv ely determine the functi on and role of each bacteriocin. BcnA has been shown to need at le ast four proteins in or der to produce a functional bacteriocin and we have a model for how those proteins may function. However more research is needed to determine the accuracy of the model.

PAGE 110

110 APPENDIX A SEQUENCE AND ALIGNMENT OF BCNB AND PATHOGENICITY-RELATED GENES CHOSEN FOR DELETION CACTAAAGGGAACAAAAGCTGGTACCGGTGGTTGCCAGCAGATGACATCCAGCATGGAAA 60 ACCTAGCACTTTTACCAATGGATGGGCTTGTCGGCCACATTATCGTCACAATGCCGGTCA 120 CATGACGGCAGCACGGGACGGGCGTTCAGGTCGTCCGGGCGCTTGCGCTGCGTGCCGGCA 180 TTCCTTGCAACGGACACCTGCATGACCGACACGCCCCGCCGCGACGATGCGGCCACGCTG 240 CAACGCTTCGGCTACGCACAGGAGCTCAAGCGGCAGCTCACGCTGAAGGATCTGTTGATC 300 TACGGGCTGGTGTGCATGGTGCCCACGGCGCCGTTCTCGATCTTCGGCGGCGTCTTCGAC 360 ATCAGTGCCGGGATGGTGCCGCTGACCTACCTGGTCGGCTTCGTGGCAATGCTGTTCACC 420 GCACTGAGCTACCAGCAGATGTCGCAGGCATTTCCGGTGGCCGGTTCGGTATATGCCTAT 480 GTCGGGCGCGGGCTGAGCAGCGGCATGGGATTTCTGGCTGGCTGGGCGATTCTGCTCGAC 540 TACCTGCTGGTGCCGACGCTGTTGTACGTGGTCGGCGCCAATGCGATGCAGACCGTATTG 600 CCGGCGATACCGCAGCCGGCGTGGATTGGGTTCTTCGTGGTGCTCAACACGGTGGTGAAC 660 CTGCGTGGCATCGAAACCACCGCACGCGCCAATCGCTTTTTCCTGCTTGCGCAGTTGCTG 720 GTCCTGGCGATGTTCGTGGTGCTGGCGACGCTGGCGATCCAGCGCGGCGTCAATGGCGCG 780 CACTGGAGCTGGCGCCCCCTGTATAACCCGCAGGCGTTTTCTCCGCAGCTGATCTTCAGC 840 GCGTTGTCGGTGGCGGTGGTGTCGTTCCTCGGCTTCGATGCGATCTCCACCATGTCCGAG 900 GAAGCGCGCGGCGGCAACCGCGTGGTGGGGCGCGCAACCTTGCTGGCCTTGCTGATCGTG 960 GCAGGGCTGTTCATCCTGCAAACCTGGCTGGCGGCATTGCTGCAGCCTACCCTGCAGCGC 1020 TACCCAAGCCCGCAGGCATCCAACGATGCGTTCTTCGAGATCGGGCGGTTGATCGCCGGA 1080 CCATGGCTGCAAATTGTCATTGCGCTCACGGTGGCGATCAGCGCGGCAATCGCCAATTCA 1140 TTGGTTGCACAGGCGGCCACGTCGCGACTGTTGTTCGCCATGGCGCGCGACCGGCAATTG 1200 CCGGGCTTTCTGAGTTACATCCACCCGCGTACCGGCGTACCGCAGCGCGCGATCCTGCTG 1260 GTGGCGGGCCTGAGCCTGGTACTGGGCGAAGTCTTCGTCGGGCAGATCGCGCTGCTGTCG 1320 TCCTTGTGCAACGTCGGCGCACTGACCGCCTTCGTGTTGCTGCATATCGCAGTGCTGTGG 1380 CACTTCCGCACCCACGGCCGGCGTGTGCTGCATGGCGTGGTGCCCCTGATCGGGATCGTC 1440 ATCCTGGCCTACGTCCTGCTCAATGCCGACCTGCATGCACAACTGGGCGGCGCTGCGTGG 1500 ATGGCAGTGGGGCTGACGGTACTGGCGTTTCTGAAACTCAGCGGACGCTCCACCGAATTG 1560 CGCGCCAGCGACGCGCTGGAATAGCCTCAACAGCAGCACCGTCTGTTGTGCCGGCGTGCT 1620 CCTGGCCCACCCGTGCTGCGGTGTGCGACGGATCAATCTCGTGCTGGGCAACCTCAAACA 1680 CGTCATCAAGGTGGTGCATCACGCCATCAGCGCAAATACACAGGGCTTGACTTGCAGAGA 1740 TGGACGATCGCTCACCGTCGATCCGCTTCACGATGCGCTGCCACGACTTGCCACGGGCAT 1800 GTTGCAGTCCAAGCAGTGCCTGGAGCCGGTTTGACGTGCAGCAGCGATCTTCGTGGCCAG 1860 GAGTCAGGATCAATCAAAAATTTAGAAACTCTCTCTTGGCCTGGAAAACTCATTTGCAAT 1920 Figure A-1. BcnB sequence. Nucleotide sequence of a 5968 bp Kpn I and Eco RI fragment containing bcnB The deduced amino acid sequence is given for BcnB.

PAGE 111

111 TTATTGCCGGCTATCTAATCCGGCCGGCGAGTGGCCGGCGGCTGGCGGTACGGCAGCCAC 1980 GAGGTGAACAATTGCCTGCGGGGAAGACGCCGGTCTTCGGATGTTCCCTGCGACACTTCC 2040 ATGGGAGAAAAGATGCGCTGCAGATCGCCCGGCTCGATCTTGGACAGGAAATAAGAGCGC 2100 CACAACGTCAATTCGTGATTGTCCGCCAGGTTGACAGTGCTGGCCGGGCCGTCCGTCACG 2160 GTGATGAACTGAGCGAAGCGATCACCGCTGGTGCTGCCGGCGCGCTCGTCAGCAGCCCCG 2220 GGGTGATCCCAGTGCGGGTGATGCATGAAAAAGGGCACCGACTGGATCGGATCCAGGGCG 2280 CGCGTAAACTGGCGCAGCGTGAATCGCAGTTCGCTGTCGTTCGATCTTAAGCTTTGCATA 2340 TAGTGTGTGTCGGCGCCCGCTGGCCCGAGGCGCGGTGGATTGAATGAATAAATGCTGATC 2400 TTGCCCGGGGCTCCGGCCGCGCGCAGAAAGTGCGTGATGTCGAAGCCGGCGAGCGTGGCC 2460 GCTGCTGCGCCCAGACTGTGTCCCGTCAACGACATCGACAACGCCTTGCCGCTCTCCCGC 2520 GCGCGATTCGTGTAGCGTTTCAACACGTCCTTCATGACGGTTCCGTCGGCGCGCGGCAAC 2580 TGAGCTTCGGACTGCCACCATTCCTGCCAGCCTGCTCCAACCTTGCCCAAGGTCGGAAGG 2640 CGCTCGTCAAATATGTTGACGTGTACCTGTGCGATCTGGCTTTTCATATCGCATAACAGA 2700 TCGTCGGTCGCATTCAGGCATGTGCCGGAAAAACCCAGCACCGCCACAATCACATCGTCG 2760 GTCTCTCGCTCCCAGAGTTTCAACCGCCGCGTGGACATCGGCACATTCAGATGCAACCCA 2820 CCGATGTCAGCCACCGTCGCCGGATCGCCCTTGTAGGATTCGCCTGCGCATGCGCGCATG 2880 CTTTCGAGCACAGTGCGGCAGTGTATTTCCCATTGCGGATCCTGTCCGTGCGGCTGCGCC 2940 GATGCACGATCCACTTGGGCAATCGCGCTGACGAGATCGCATGCCTGGGTATCTACCTGC 3000 GGCAGACGCATGCCAGATGACCGCCAGGCACCTGCCGTGATCGATCGTGTCGCGTTACGT 3060 CCGCGTTTGACCGGCGCAGCACTGAGACCTGGAAACGCACTGGCGCTGTCACGCGCGTGC 3120 CAGGGGAAAGCGGTCGCGGAGGCAGGCACGTCGCTATCGCCGTCGCGCGATTGCGCGTGC 3180 GGCAGCCTTGGAAAAGAAGTTGAGGTGGGTCGCATGCGACATGTTTCAGGTGAGTGCGCA 3240 GAGATTCTGTTTATATCCCCACCTGGCACGCAGCCAGAGATGTCGTCCACCGCCGTGAAA 3300 ACGAACGAAGGATTCACTGGCTCCACCATCGCTGTGCATCCCTCAAGGCCGGGTTGAGGT 3360 CGCCCGAGGCCCTAGGAGCGCTGATCGCCACCACGACTATGACGTGGATTGAACACACCA 3420 AGCTGCCACTGCGCAGCTGGATGCTGGACTTGCACCTGCTGACTAGAACCAACATGGCCG 3480 CCCTGGGAGCCTATGCACCGCCCGGGGCCGACGGCATGACGGCCCAGTGGATGACAGACG 3540 CGTGCGGGCACTATATCGATCACGACCTTTCATTCCCGAACGAGATTTTCAGCACATATT 3600 CAGCTTTAAAGTTTCGCTAGGCTCGTTCGGCGCCAACGATGGCGCCCGCTTGTTGCAGCC 3660 CCCGCTGCAGCCTGACGATCCCTGACGAGAAGGACTACCCATGAATCGCAAGAACGCGAT 3720 M N R K N A M GTATCTGGCGTTGTTTTCCGCTGTTTCCGGCACCGCAGCGGCCGCCCCGCCGACCGAGAT 3780 Y L A L F S A V S G T A A A A P P T E M GGATGCAGCGCCGGTCACCACCGCGCCGCAGGCGGCCAAGCTTGGTGCGGCCACGTTGCA 3840 D A A P V T T A P Q A A K L G A A T L Q GTCGGCAAGTTTACGCGGCGGCGTTCTGCCCACGCGCGTGGTGCAGCTCACCGCACCTAC 3900 S A S L R G G V L P T R V V Q L T A P T Figure A-1 continued.

PAGE 112

112 CAGCGCCGAAATGGGTCGTGTGCGCGAACGTCGTATTGCACAGGTCAAGCATGGGCAGCC 3960 S A E M G R V R E R R I A Q V K H G Q P GTTGCAGATCGGTTTTTCGCGTGCGGTGACACAGCCGACGGTGAACCTGTCCAAGCTGGA 4020 L Q I G F S R A V T Q P T V N L S K L D TTGGCAGATGGCTCCCGATGGCTCGCGCGTGGCCAGTCTCAAAGTCAGCTCCGCACAGGC 4080 W Q M A P D G S R V A S L K V S S A Q A GGCGTCGCTGCGTGCATCGCTGGTACTGCGTGGCGCCGGTGCAACGCCTGGCGACCCATC 4140 A S L R A S L V L R G A G A T P G D P S CAAGGTGACGCTGCGCTTTGCCGGCAACGATGGTCGCGTGTTCGAGCAATCCGGCGCCAG 4200 K V T L R F A G N D G R V F E Q S G A S CTTTGCTGCCAGCGGCAACGATATCGGTTGGTCGCCGACGGTGAGTGGCGAGGATCTGCT 4260 F A A S G N D I G W S P T V S G E D L L GGTCGAGCTGTCGCTGCCGGCCGGGTTGTATCCGGAAAACTTCAGCCTCAGCATTCCGCA 4320 V E L S L P A G L Y P E N F S L S I P Q GTTGTCGCATCTGGATATCAGCCCCACTGCCAGCCCGCGCGACATGATGACCATTGCCAT 4380 L S H L D I S P T A S P R D M M T I A I CGGCGAAAGCGATTCCTGTCAGAACGACATCGTCTGCCGTGCCAACCCCACGACCGGCTT 4440 G E S D S C Q N D I V C R A N P T T G F CACCAGCGCCGCAAAGGCGGTGGCACGCATGGTGTTCACCACCAGCCAGGGCTCGTTCCT 4500 T S A A K A V A R M V F T T S Q G S F L GTGCACCGGCACGCTGCTCAACAATACCAATTCGCCCAAGCGCAACCTGTTCTGGACCGC 4560 C T G T L L N N T N S P K R N L F W T A GGCGCATTGCATCAGCACGCAGACGGTGGCCAACACGCTGCAGACCTACTGGTTCTACGA 4620 A H C I S T Q T V A N T L Q T Y W F Y D TGCGGCCAGCTGCAATGGCAGTACGGTCAGCTCGCAGGCAACCACATTGACGGGCGGCGC 4680 A A S C N G S T V S S Q A T T L T G G A GTTCCTGCGGCATGCCAACACCACGCGCGACACCGCATTGCTGGAGCTGAAGACCGCACC 4740 F L R H A N T T R D T A L L E L K T A P GCCAAGCGGCGCGTTCTATGCAGCGTGGAATAGCGCGGCGATCGGTTCCACCGGCACCTC 4800 P S G A F Y A A W N S A A I G S T G T S GATTGTCGGCATCCATCACCCATCGGGCGACGTCAAGAAGTATTCGCTGGGCACGGTGAA 4860 I V G I H H P S G D V K K Y S L G T V N TGCACTGAGCAGCTCCATCGATGGCAAGAGCCCGCTCTATCGCGTGGTGTGGAGCGACGG 4920 A L S S S I D G K S P L Y R V V W S D G TGTGACCGAAGGCGGCTCGTCCGGCTCGGGCCTGTTCACCGTTGCCAGCGGTGGCGCCTA 4980 V T E G G S S G S G L F T V A S G G A Y Figure A-1 continued.

PAGE 113

113 CCAGCTGCGTGGCGGCCTGTATGGCGGCTATTCGTTCTGTAGCGCGCAGACCGATCCGGA 5040 Q L R G G L Y G G Y S F C S A Q T D P D TTACTACTCGCGTTTCTCGGACGTGTATTCCACCATTTCCACCTACTTCGGCCAGTAAGC 5100 Y Y S R F S D V Y S T I S T Y F G Q GGTAAGCCGCGTGGCAGGCGACCTTGTGCCTGCCACGCTTGATCGCGCATCGCCGGGCAA 5160 TCCGTTTGCCGGGCGATGCGTCTTCGGCTCAGCGACCGTGCATGCGTTGATCCGTGCACT 5220 GAAGTGTCCGCAGGCTGTCCAATGCCACCACCATCTTCTTGCCACGCACTCAAGGCAACG 5280 GCGTGACGTGCTAGCGCGAGACATCGCTCCACGGCTGTTTTTTGCAGCCTCGCTCAACAC 5340 ATTGATCGGTGTGCGGTTGCCGGCGCCATAGGTGTACCAGATCCACTTCGGCCAATCGGG 5400 GAATGCCTGATTGACGTCGACTGCACCAACGGCGCACTAGCGGCGGCGAGCGGCGCAATG 5460 ATGCAGACTGATGGGGCCTGTTCCTTCACAACGGTCGCCCATGGTTCGCGCACGTGCCCC 5520 ACTGCTGGTCATCATCCTTGTTGCCGTGGCGCTGCAAGCCTCCGCCCTGGCGGCCGGCAA 5580 GCGGGCCGATCGCGAGGACATCAAGGCCTATGCGCTGCAGCGGTGCCTGGATAACAACTA 5640 CACGCGTAGCAGCAAGTACGCACCCGATCAGCTACGCGACCGCTCGTATCTGCTCACCAC 5700 ATATGCGATGGACAACGCCAAGGTCGGCGCGACCGACCGCCTGCATCGCTTTGTCGACGC 5760 CAGCACGGCGGGGTTTGACAAGCGCGAAGTCCCGATGAAGGACGAAGCGCGCCGCGGCCC 5820 GTTCAATCGCATCTTCGCCCAATGCATGGCGTTCTATCGTTCGCCGGCCTTGGATGAATT 5880 CCTGCAGCCCGGGGGATCCACTAGTTCTAGAGCGGCCGCCACCGCGGTGGAGCTCCAATT 5940 CGCCCTATAGTGAGTCGTATTACAATTC 5968 Figure A-1 continued.

PAGE 114

114 GCTCAAGCTGGTGGTGGTGGTGGCCGCCATTGCCGTGACGGCGCCGTGGGGCGCCTCGGC 60 CATCATGCAGTTCGGCCAGGCATTGATGCAGGCGGCGTTCCCATG ATCCGTCGCATCTCG 120 M I R R I S CCCGGCACCTTGCGGCCCTCGGTTTCCATCGAGCATGGCGGCTCGCATCACCATGGCGAT 180 P G T L R P S V S I E H G G S H H H G D CATCATTCCGATGCGGCAGGCACCAGCGCGGGGTCGCCTACCGACATTGCGGCACGAGCA 240 H H S D A A G T S A G S P T D I A A R A CCGCGTCTGCGCCCTGCTCCACCGCGCAGGCGACGCCGCGGGATCCGGTCGCTGGACGGC 300 P R L R P A P P R R R R R G I R S L D G CAGGAGGATGAGTTCGACGCCAACGAGCAGGAAGAGATCGAAGCCAAACGCGAGTGCGCA 360 Q E D E F D A N E Q E E I E A K R E C A CTGCGCGGACGGGTGAGCGTGGCGATCACGCCGGCCCAGAGCCGGGAGCATGGCCAGGGC 420 L R G R V S V A I T P A Q S R E H G Q G GACCAGCACGGCGGTGACCGCAGCACGCATACCGACGATCCGCAGGCCGGTCCGTGGCAC 480 D Q H G G D R S T H T D D P Q A G P W H AGCGCCGCTCCAGCGCAATTGAGCGACAGCATCCGCACGTGCATCGACGCGATTGTCGAT 540 S A A P A Q L S D S I R T C I D A I V D CGCTACGTGACCAGGCGCGACGCCGATCCAGTGGCCAAGCGACACGCGCTGGCTGCTGCG 600 R Y V T R R D A D P V A K R H A L A A A CTTGTCGAGCTGCGCGCCGTCGGCGTTTCGCATCCTGCCGTCGCCCCCTTGACCGCAACG 660 L V E L R A V G V S H P A V A P L T A T ATCTGGAGGTTGATGCGCGAGCATCTACGCGCGTACGACAAGGCCACTGCGGCGGAAAAC 720 I W R L M R E H L R A Y D K A T A A E N CTGCTGGCCTTGCGGACCCGATTGCTGGAGTTGATGCCATCGGAGTTGGAGCCGGTGCCC 780 L L A L R T R L L E L M P S E L E P V P GCGCTACGCAATTTTCATCTCCTGCTTCCACTGATCCTGCTGAATGCGGAAAAACCGCGC 840 A L R N F H L L L P L I L L N A E K P R AGACGCATCGACCGTACACATGCCATCACACGTCTGAACACACTGCTGATTGAGCAACCA 900 R R I D R T H A I T R L N T L L I E Q P CAACAGGCGGCTCAGGAGGTTCGCCCA TGA CCATGCAGCTTCGCGTACTGACAGGAACCC 960 Q Q A A Q E V R P ACGCAGGCGCACGATTAGAT 980 Figure A-2. Nucleotide sequence of a 980 bp fragment containing the hpaA ORF. The deduced amino acid sequence is given for HpaA. The positions of the annealing points of four primers (hpaAF, hpaAR, hpaADF and hpa ADR) used for PCR are indicated. hpaAF hpaAR hpaA2F hpaADF hpaADR hpaA2R

PAGE 115

115 GCGACGCTGCGCGACACCCTGGCCGACCCCGCCCTCACCGTCCGTTACAAGGGCAATGGC 60 GTTTTCTCGGTTTCCGGCAGCAGCGACTACGCAGAACGCGCAAGCCGCCGGATTGCCGAC 120 GTCCGCAGCGACCTGGGCCCGGAGGTCCGCGCGCTGCACGTAGAGATCAGCCAGCAAGAC 180 CCGTCCATCAAACCACCGGACAATTACGACGCGGCATTGCTCGCCGATGGCCTGCACTAT 240 GTCGAAACGCCGGATGGCACCAAGCATATGACATCCCTGCCGCAGCAAGCGGCGCAATGA 300 GAGATCACACCATGTTCGATGCAATGACCGATGCGGTCACTCAGGATATGAGCAAGATTC 360 TGCAGGCCAAGGCGATGGATCTGTCCGGCGAGCGGCTGCGCAACGTGGAGACAGCGCTGG 420 ATGCCACTGCGCAACAGATTCGCGTGCACTGGTCTGCAGCGAGCGACCAGGTCGCGCGCA 480 ACGACTTCAACGTCTTGTATGACGGCATCACGGCCGCGCGCAACATCGTCGCGCACATCG 540 CAAGCATGCCGTAAATCGATTTCTTGGTCGCTCGGACCAAGTAAAACGCGCCGATAGGCG 600 CACCATTACATGCTCACTACGAAGGAGTACTGGTATGCAAATTTTTCCTGAAGTAAGCTC 660 GTGGAGGTCCCGTGTTGGCCAATGCATGGATGGCTTCACCGGTGGTTTGTCCAATGGAAT 720 TTCCGGCGCTGCTGCACTCTCTGGTGCAAACGGCCAGATGGATTCGCTACTCGGCGACAT 780 GTCAGCCTCGGACGACGCTCAGAAGTCCATGAACAACAAGATCACGATGCTCAAAAACGA 840 TCTGGACTTCAACGTGGCACTCAACAAGTTCATCGGCAAGGCGGGCGACAACGCTAAGCA 900 GCTCGTTGGCCAGTAATGGGCTCGGGCATGCGCAGGACCTGTAGGGGGTCCTGTGCATGT 960 CTCACCACAGATG AGCAGCGCGCGATTCGAAACCATCGTTCGACAGATGTGCGAGGCACT 1020 M S S A R F E T I V R Q M C E A L GGACTTGCCGGATGTGGAGTCGGTGCTGGATCGGCGTGTGCTTTGGGTCGAGGGCTTTGA 1080 D L P D V E S V L D R R V L W V E G F E GGTCTATCTGCACCTGCCGACCCCACAGCCCGAAGACGACGTAAAGGAAGAAGCACTGTA 1140 V Y L H L P T P Q P E D D V K E E A L Y TCTGCGCATCGCCTACGGGTTACCACCTGCCGGGCGGACATTGACCGTTTTCAGGCTTCT 1200 L R I A Y G L P P A G R T L T V F R L L GCTCGAAGCCAATCTCTCGGTGTATGCGCAGGACCAGGCCCAGCTTGGACTGAACGATGA 1260 L E A N L S V Y A Q D Q A Q L G L N D D CGGCGTTATCGTGCTGATTGTGCGGGTACC GCTGGACGACGACGTGGATGGAGCGTGGAT 1320 G V I V L I V R V P L D D D V D G A W I CTGCGATCTGCTTGCCCACTACGCCGAGCATGGCCGCTACTGGAACAACAATATCTTCGT 1380 C D L L A H Y A E H G R Y W N N N I F V GGCGCACGACGAGATGTTCGAAGGAATCGCGACGGGCAACTATCTGTGGTTACGCGCCTG 1440 A H D E M F E G I A T G N Y L W L R A Figure A-3. Nucleotide sequence of a 2345 bp fragment containing the hpaB ORF. The deduced amino acid sequence is given for HpaB. The positions of the annealing points of two primers (hpaBF, hpaBR, hpaB2F and hpaB2 R) used for PCR are indicated. The Kpn I restriction site used fo r disruption is underlined. hpaBF hpaB2F

PAGE 116

116 ATACAGGCTTGTTGGCAGCTTGCCGAAGCCTCAATCACGGCCGCAGCGGCGCCCATTCGC 1500 CATGGCACAGCAACCTAGCTTGCCGGCCAACTCCACCGGCCTGGCACTCCTCTGCCCGAT 1560 GGCGCCGGTGGCGCTGAGTGAAACCCACTTGCCATGTCGAAGCCTGACACGCCTGCATCT 1620 GCGCCGGCTTGCTCCCCGATATAGGTTTGGCGGGAGATCGCAGGTCTGTCGTGCCGGCTC 1680 ACCGCGATGGCCGACACTGCACAGAACAAAGGCAGGCGCAGACGCATGGTCTACACCGTG 1740 CACATCCGCGACGGCAGCTGATGCACCGGTGCTGTTGGCGCTGACGAGGGACACGACCAC 1800 TTTGCCGGCGAGTTCTTTCCGGTGTTTGACGTGCGCGTCAAGGTCGACGCGTTCCGGAGC 1860 GACGGCGCCATCGAGTCGCAGATGCAATGCCTGCCTCGTGCAGTGAGCGACCTGCGCAGA 1920 TCGCGGCATGAGGCCGTGGTGTTCCATGATTGCTGGTAATGGCTAAGCTGCCAGACACCA 1980 CCAAGCCACAGACGCGACGAACGCGACGTCTGTGGCGGCGGTGTACAACACTCCTGAGTA 2040 TCAACGCGGCAGCTGAGCGCTCATGCTCGCCCGCTCTGCCACTGGCGAAGTGCTTCGCTG 2100 TCATGCAGTGCACGCTTGAGCGTTTGCAGCTGGCCATCCAGTGCCTGCGTGGCCGCGTCG 2160 CGGGGGACGCCGCTAAAGGTTTGACTCCACCCCTCGCGTAGTGCAGTGCGTACCGCATCG 2220 ATGGTCGCGATCGAAACGGACTGCGTGCCGATCTCCGTACGCAATGTCTGCAGCGCAGTA 2280 CGCAGTGTCTGGCCAAGCGGCCCAGGGCCAGGCGCGGCGCGCATCCCCTCCTCCATGGCA 2340 GCCTT 2345 Figure A-3, continued. hpaBR hpaB2R

PAGE 117

117 CGCATACCGCAACCGCAGGCACCGGGCAGCTGCCCGCAGTGATGCTCGACCATGCTGTCG 60 AACAACTGATCAGGCAATCGATTCGCGCGACCGCGGCCGGCAATTTCCTGGCACTTCCGC 120 CCGAGCAGGCGAACCAGCTTGTCGAACAGGTGGAACGCATCGTCGGCGACCATGCGCAGC 180 ATCCGCTGGCGGTGGTCGCGTCGATGGACGTGCGCAGGTATGTGCGCCGCATGATCGAAG 240 CACGGCTGACCTGGCTGCAGGTCTATTCGTTCCAGGAGCTGGGTTCGGAGGTGCAGCTGC 300 AGCCGATCGGTAGAGTGGTGGTGTGATGCGCAAGCCGCCCCTCCGCCATGTGCGCATCCT 360 M R K P P L R H V R I L GCCGGTCAGCGGCGCACTGCAGCGGCCGGCGGCTCCAGCCACGCCGGCCCGGTCGGCGCT 420 P V S G A L Q R P A A P A T P A R S A L CCGCTCCAGCTTCCTGCAACTGCGCCAGCGCTTGCGCAGCGCACAGCTCGCGCTTCCCTG 480 R S S F L Q L R Q R L R S A Q L A L P C CATGGTGTTGCCACCGCAGTGCGACGAGGATCGGCCGGAGCCTGATGCCGAGGAGGGCTT 540 M V L P P Q C D E D R P E P D A E E G F CACCGAGGCGCATGACAGCGTGCCCGTGCAAACCGACCCACCCTTACGTATCGAGGGGAC 600 T E A H D S V P V Q T D P P L R I E G T GAAACACCAAGAGCCGTCCCAAGGCAACGGCGATGGCGCAGTGGGGCGACAGATTGCGAC 660 K H Q E P S Q G N G D G A V G R Q I A T CGAGTGGATACGCACGCAACGCGCCCACATGGCGATCGACCACATCGCGCTGCGGGTGGC 720 E W I R T Q R A H M A I D H I A L R V A CGAGTTCTGCAATGCCCAACCGGTGCGCAGCGCAGGGAGCTGGGAAGCGTGGCTGGCTAT 780 E F C N A Q P V R S A G S W E A W L A I CGACCAAGAGGTCGTCGCACAGACGACGTTGTTTTTGCGGCTTTCGCCGGACCAGCTATC 840 D Q E V V A Q T T L F L R L S P D Q L S GCTTCGCTTCAATACCAGTTCGCCAGATGCGCGCGAGGTACTTTGGTGCGGAAAGCAGCG 900 L R F N T S S P D A R E V L W C G K Q R CCTGGAGGCTGCACTGACGTCCACGCTGAGTAGCACGCTCCAGATCAGCATCGAGGTTGT 960 L E A A L T S T L S S T L Q I S I E V V Figure A-4. Nucleotide sequence of a 1419 bp fragment containing the hpaC ORF. The deduced amino acid sequence is given for HpaC. The positions of the annealing points of four primers (hpaCF, hpaCR, hpa C2F, hpaC2R, hpaCDF and hpaCDR) used for PCR are indicated. hpaCF hpaC2F hpaCDF hpaCDR hpaC2R

PAGE 118

118 C TAA CGCGGCACTGGCGACAGAAAGCCATTTGGAGGAACTCCATCTTGCTAACCGAGCAG 1020 AGCCAGACTCCGCCGGCATGTGCACTCTCTCAGGCGTTGACGCGCGTTGCGGCCGAGCGC 1080 GCCCAGCTTGGCAGAGTGTTCGGCGACCCACGCGCAGCACGCCAATGCGGCTTTGCCACG 1140 CATTGCCGTCGCATCTCGCCCAATGATGCGGCCCGCCTGCGCCTGCAGCTGGATGCCGGC 1200 CAGATGGAGTTGCGGATTGCCGCGCGCGACGGCTTGGCGCTGCTATTGAACGAGGACGAT 1260 GACTCACTACGCGTTTCGATCGCGGGAATGCTACTGGCTGATCGCCTGGGCGCATTTGCG 1320 CCGCTCGGCCTGGGCGCTGCCGAGGTGATCGCCTTCGAGCGCGATGCAGAGCCAGACGAC 1380 TGCCACGGCATCGGCATGACACTGGGCGATCTCGATGCC 1419 Figure A-4. continued. hpaCR

PAGE 119

119 AATCCGCGCGTGCGAAAGGCGCCGCGCATGCCAGCGCTGCCGCAAACAAAATGCGCTGCG 60 GACCAGCCCCTGCCCGACGACCACGTCCGGACCATTGCGCGCGCATGCGCCGCTTCGGCG 120 AAAAGCCCGGACGAGGCCTCCCACTGCCAATTGGATACTTTATCGACATACTTGAATGCG 180 ATCGTTGAATTGATCATGTGTCGTTACCTCGATCTCGATTGAATACAGGTCTCCAGGTGA 240 GTAGAGCGCCGGTACACAACGACCCACGACGCTCGACGGCGGCAACATAACGCCACCTGT 300 ACGCGATTTA CTCGGG ACCTACGAAACACTTAATAGGCAAGCGAAAAAAGTTTTTCAACG 360 ACACCCCGTGGATTTTCTTATCGACCCTAAGAATCTTTTTTATTTACCTCTTCGCTTGCA 420 CAAGCGTAATTTCGCATACTCTATGGCGCCGATGATTTTCAGCTTCTACTGTTCAGCGGG 480 GCGGAAAGCGACACAGCAGTCGTTGCCTCCTGGCCCCGGCAGTTGAGCGAAAGACAAATC 540 CTAGTTAAACCAGAGAGAAATCGCTATGATCAATTCATTGAATACGTCGCACCTCGGCGT 600 M I N S L N T S H L G V CGACTCTTCCTTTATGCAAGTCAACCCGGACCAATTTCAAAAATTCGATTCAAATCAAAG 660 D S S F M Q V N P D Q F Q K F D S N Q S CAATCAAGGCATCTCGGAAAAGCAGCTGGACCAACTGCTGACCCAGTTCATCTTTTCAAT 720 N Q G I S E K Q L D Q L L T Q F I F S M GCTTCTGCAGGACGACAATGCTGATGATTCCCCGAACTCTGACAAGCCCACCGATTTTCC 780 L L Q D D N A D D S P N S D K P T D F P GTCGCCACGCACCCAGATGCTAATGAATGTCATCGGAGACATTTTACAGGCGAAGAATGG 840 S P R T Q M L M N V I G D I L Q A K N G CGGACGCCTCGGTGGTTTATCCGATGGAGGGCTCAACACTAGCCTCAGCCTCTCGGGCGA 900 G R L G G L S D G G L N T S L S L S G D CACTGCATCGATGCAGTAAACAACTGTTGGTCGCCCCTACCGACCACGGCCTGCAGGTTG 960 T A S M Q CCGTTTCATGCTTCAGCCTGCAGGCACCCTCATTGAATTCGCTTGGGACCCGTGTT CTCG 1020 GG CGAATTCCGCAGGGGTTCCATGCCAGTGCACTGTGGGGCCGCTCCTCGTTATAGAAGC 1080 GCCGCCACGCTTCGATTTTGCTCCGTGCATCGGCTAAAGACAAGAACCAATGCGCATTCA 1140 GGCACTCCTGTCGCAGCCGGCCGTTGAAGCTTTCCACCATCGCGTTGTCCGTCGGCGTGC 1200 CACGTCGGGAAAAGTCCAGCTCTACGCCGTTTTCATAGGCCCAGCGGTCCAAGACCTTGC 1260 CGGCAAACTCACTGCCGTTATCCACTTTGATCGCTTCCGGCTTGCCTCGTTGAACGACCA 1320 Figure A-5. Nucleotide sequence of a 1716 bp fragment containing the xopA ORF. The deduced amino acid sequence is given for XopA. The positions of the annealing points of four primers (XopAF, XopAR, XopA2F, XopA2R XopA3F and XopA3R) used for PCR are indicated. Two Ava I restriction sites are in bold XopAF XopA3F XopA2F XopA2R XopA3R

PAGE 120

120 GCCGCACCACTGCGTCGGCGACATCGTCAGCGCCCAAGGACTGATCCACGAAGATCTCCA 1380 GGCACTCATGCGTGAAGTGGTCCAGCACCGGCAGCAGCCGGAAGCGACGGCCATCGAACA 1440 GCGCGTCACTGACGAAGTCCATGCCCCACAGCGTGTTGGGCGCCGTGGCCACCTTGATCG 1500 GTTGCCGGCGACGACTGCTGCGACTACGTCGCGGGCGACAATGCCGCAACGACAGGCCTT 1560 CTTCCTTGTAGATGCGATGCACCCGCTTGTGGTTGTCCCGCCAGCCTTCGCGCCGCAGCA 1620 CCACCAATACGCGTTCGCAGCCATAGTGGATGCGCGTCTGCGTGATCTCGCGCATCCGCA 1680 AGCGGATGGCGCTGCAGTCGCGCGCCTTGGCCTTGTACGAAAACGCCGAGCGCGACATCG 1740 CAACGATCCGCAATGCGCGTCGTTCGCTCACGCCGAAGCGCTCTCTCAAGCGGACAACCC 1800 AGCTGCGCTTCTGGGGCGCCCTTAGAGTTTTTTTGTGACCACCTCCTGGAGCATCGCCTT 1860 GTCCAGGCTCAGGTCGGCAACCAGCTGCTTGAGCTTGCGGTTTTCCTCTTCCAGCA 1716 Figure A-5. continued. XopAR

PAGE 121

121 TGCTGCCTTTTTGATG GACAGTGGCTTGTCGCATGTAAATGGCAGGCAGATGCTTCAGGA 60 ACTGAATGAAGATCAACGCGACCAAGTCATACATCAAATAATAAGACGAATTGAGTATTG 120 TGCGGATCCTGAATATCGAGAAGTTGCGCTGAGCCGACTGGAGTCGGATTGCAGTGGAAA 180 AATTACGCTAAGTCAACGGACTTTGGATCGCATTGACAAAGCCAAAGCCAAAGCCGAAGC 240 CAAAGCCAAAGCCAAAGCCAAAGCCAAAGCCAAAGCCAAAGTCGGAGTCGAAGCCGGAGC 300 CCAATGCAAGATCAACGAAATTATGGAATATATACCAAGATATGAAGCATTAGAGAAAGT 360 M E Y I P R Y E A L E K V GCCAGTGCGTGTGGGATTCCATGCTTACCTGCGTGGTGATGGCTCATTCGGCCCAGGGCT 420 P V R V G F H A Y L R G D G S F G P G L ATCTGGCATCCTTCGATACATGACCCCAGATCAGAAGAAAAGATTGTATCTAGCAAGTGA 480 S G I L R Y M T P D Q K K R L Y L A S E GAGACGCAAACTGGCCTTGGCCGCTCCAAAAAGCAAGCCTCTAAAAGGCGTATTCCGGAC 540 R R K L A L A A P K S K P L K G V F R T CCTCCATCAAAAACCAAATTTGCTTCTTGAGATTTCGAGCAAATTCAGCAATAGAGCGTA 600 L H Q K P N L L L E I S S K F S N R A Y CAGCATCAATGATTCAAGCAGCGCATATTTATCACAAGCAGACCTGGAAGAAATGGTCGA 660 S I N D S S S A Y L S Q A D L E E M V D CGAGGAAACCGGCGAACTGACTCGTCTGGGTGAAGCAGTGATTTCAGGAGCATCCCAAGG 720 E E T G E L T R L G E A V I S G A S Q G CATCCAGACGGCAATTCGAGCCAACTTCAGAATGCGTTATCAACAACCGGATCTGCCTCC 780 I Q T A I R A N F R M R Y Q Q P D L P P ATACAGTCCCCCTCAGGCCTTCCATCGGCCAGAAGAAACGTGGAATCCCCATACTCCGGC 840 Y S P P Q A F H R P E E T W N P H T P A GGGTTCTTCCTATTCGTCCCTGTTCCCGCCCACCCCTTCTGGCGGTTGGCCGCAGAACGC 900 G S S Y S S L F P P T P S G G W P Q N A ATCAGGTGAGTGGCATCCCGATACTCCGGCGGATTCTTCCTATTCGTCCCTGTTCCCGCC 960 S G E W H P D T P A D S S Y S S L F P P CACCCCTTCTGGCGGTTGGCCGCAGAACGCATCAGGTGAGTGGCATCCCGATACTCCGGC 1020 T P S G G W P Q N A S G E W H P D T P A Figure A-6. Nucleotide sequence of a 2173 bp fr agment containing the ORF. The deduced amino acid sequence is given for XopD. The positions of the annealing points of four primers (XopDF, XopDR, XopD2F, X opD2R, XopD3F, XopDDF and XopDDR) used for PCR are indicated. XopDF XopD2F XopDDF XopDDR XopD3F

PAGE 122

122 GGATTCTTCCTATTCGTCCCTGTTCCCGCCCACCCCTTCTGGCGGTTGGCCGCAGAACGC 1080 D S S Y S S L F P P T P S G G W P Q N A ATCAGGTGAGTGGCATCCCGATACTCCGGCGGGTTATTCCCATCGTGCATGGCCAGCCCA 1140 S G E W H P D T P A G Y S H R A W P A Q GCCCGAAGCGTCGAGTTCCACCTTCGATGATCTTGAGTCCTTGGACTATAGGCAGAACTA 1200 P E A S S S T F D D L E S L D Y R Q N Y TGGTTATCGCGAATTCGACCTTAACACCCCCCAGGAAATCGAGCAGCCAGGGTGGTGGCA 1260 G Y R E F D L N T P Q E I E Q P G W W Q GCAAGCCACGCCCGCCCAAAGCACGGACTCGACCTTCGATGGCCTCTCCTCCATGAGCCA 1320 Q A T P A Q S T D S T F D G L S S M S H TTACGGTAGCGAATTCGACCTCAACATCCCCCAGCAAGAAGAGTACCCTAATAACCATGG 1380 Y G S E F D L N I P Q Q E E Y P N N H G CACGCAGACCCCCATGGGATATTCGGCCATGACTCCTGAAAGGATCGATGTGGACAATCT 1440 T Q T P M G Y S A M T P E R I D V D N L GCCGTCGCCCCAGGACGTCGCAGACCCCGAACTTCCTCCAGTGAGGGCCACTTCGTGGCT 1500 P S P Q D V A D P E L P P V R A T S W L GCTGGATGGACATTTGCGCGCCTACACCGATGACCTAGCTCGCCGATTGCGAGGGGAGCC 1560 L D G H L R A Y T D D L A R R L R G E P CAACGCCCATTTACTCCACTTTGCCGACTCGCAGGTAGTGACCATGCTGAGCTCCGCAGA 1620 N A H L L H F A D S Q V V T M L S S A D TCCAGACCAACAGGCCCGCGCACAGCGCCTTCTTGCCGGAGACGACATCCCACCTATCGT 1680 P D Q Q A R A Q R L L A G D D I P P I V GTTCCTGCCGATCAATCAGCCCAACGCTCATTGGTCATTGCACGTCGTCGACCGGCGTAA 1740 F L P I N Q P N A H W S L H V V D R R N CAAGGACGCTGTTGCGGCCTACCACTATGATTCCATGGCACAGAAGGACCCACAGCAACG 1800 K D A V A A Y H Y D S M A Q K D P Q Q R CTACCTTGCTGATATGGCGGCCTATCACCTTGGCCTTGATTATCAACAAACTCATGAAAT 1860 Y L A D M A A Y H L G L D Y Q Q T H E M GCCCATCGCGATACAGTCGGACGGTTATTCCTGCGGCGATCATGTGCTGACCGGGATAGA 1920 P I A I Q S D G Y S C G D H V L T G I E GGTGTTGGCCCACAGGGTACTCGACGGCACCTTCGACTACGCAGGCGGCAGGGACCTGAC 1980 V L A H R V L D G T F D Y A G G R D L T TGATATCGAACCAGACCGCGGCCTCATCAGGGATCGTCTTGCCCAAGCGGAGCAAGCTCC 2040 D I E P D R G L I R D R L A Q A E Q A P Figure A-5. continued. XopD2R

PAGE 123

123 AGCAGAAAGCAGCATCAGGCAAGTTCCGCACGATCCAACGAACAGAAGAAAAAGAAAAGC 2100 A E S S I R Q V P H D P T N R R K R K A AAGTGGTGGAAAAAGTCCTAGCAGTTCGACCATCAGCGGTAAAGTAGGGTTGGCAGGATT 2160 S G G K S P S S S T I S G K V G L A G L AGGGAAGGTCTGA 2173 G K V Figure A-6. continued. XopDR

PAGE 124

124 ATCGCCCGCATCGCCTTCGAACCTGCGCCGCGGCGTGGCGTTGCATTGCCGTATCCCACG 60 CTGCAGGTCGGGCAGGTGACGGCGCATGCCACCGTGCTGGGCGCGGCCATGCTGCCGTTC 120 AAGGAAACGCTTTTCTAGCGCACCGCATGCGCGGCGATATGCGCGCCAAGTGCTGGCAAC 180 GCGTCCAAACAAGGCCTGCGCCGCACGCCTGCCAGCGCGCGCAACGCAGGCATCGTTTCG 240 CATCCGGGCGGTACTTTTCGCCTAATTTGCCAATTGTCATATG CCACGCGCTTTACTGGC 300 M P R A L L A CGCCCGCCGCGTTTTCGAGGTCATCATGCGTATCGGTCCTCTGCAACCTTCTATCGCGCA 360 A R R V F E V I M R I G P L Q P S I A H CACTGCCGCGCCGGCACTACCGACGCACACCAGTGCGATCAGCCCGACGCAGGTGCCGCA 420 T A A P A L P T H T S A I S P T Q V P H CATGCCGGGCAACACTCCGCCACTGCGAGAAAGGCCGCGTCGCCGCGCCACCAATATGCC 480 M P G N T P P L R E R P R R R A T N M P CCCGCTGGTGCCGTTGAACGACAGCGCGATGACCGGCAAGCCGGCCCTGGTGGCGCTCGA 540 P L V P L N D S A M T G K P A L V A L D CAGCGAATTCTCAGAGCAGCGTCTGGCCGAAGTGCAGGCGCGCCAGATCACCGTGCAGAC 600 S E F S E Q R L A E V Q A R Q I T V Q T ACTGCAAGCCAAGCTTGCCACGCATCTTGCGCAGGCCGGCACGGCGCTCAAACCCGACAG 660 L Q A K L A T H L A Q A G T A L K P D S CATTGCCGCACGCTTTGCCGCCGGCACACTGGAGCCGGTGTATCTGGATACCGCCGCCTT 720 I A A R F A A G T L E P V Y L D T A A F CAATGCCATGTCGCGCGGGCTGCCCGCACGCGCACGTGCGGCCGCAGGCCCGGTGCTGAT 780 N A M S R G L P A R A R A A A G P V L I CGATGCACAACAAGGTCGCATCATCTTCAATCTGCAGCGCGCGTTCGCCCCTGGCGACAC 840 D A Q Q G R I I F N L Q R A F A P G D T CTTCAGCGACGCGGCGCTTGCCGCGCTTGGCAAGCAGTTGAATCTTTCCGGCCACGGGCT 900 F S D A A L A A L G K Q L N L S G H G L GGCAACGCCGAACTGGCTGCAGCCTGCCGCAGGCACGCCGGGGCGGCGCAAGCTGCAGCA 960 A T P N W L Q P A A G T P G R R K L Q Q AGCCGCGCGCTACCACGGCCACGAGGTGCCGGCCCGCGACGGTGGCGCCGGTTTCTTCAA 1020 A A R Y H G H E V P A R D G G A G F F K GGCCAACGACCATCGCCTGCTGGAAGGCAAGCAAGTGCTGTTGCGCAATCATCAGAAGTC 1080 A N D H R L L E G K Q V L L R N H Q K S Figure A-7. Nucleotide sequence of a 2735 bp fragment containing the avrBs2 ORF. The deduced amino acid sequence is given for Av rBs2. The positions of the annealing points of four primers (avrBs2F, avrBs2R, avrBs2DF and avrBs2DR) used for PCR are indicated. avrBs2F

PAGE 125

125 GCTCGTGCACAACCACTACTTCGAAGCACCCAGCACGCGTGCGTCCGGAAAGGACGTCAT 1140 L V H N H Y F E A P S T R A S G K D V M GGTGCATCGCGGGCTGTTCGATAATCACGCCGGCATTCCGGAAAACTCGCTGGCGTCCAT 1200 V H R G L F D N H A G I P E N S L A S I CGATCATGCCTACGAGGTCGACCAGGGCTACCGCAATCTGGAGCTGGACGTCGAAGTCAG 1260 D H A Y E V D Q G Y R N L E L D V E V S TTCCGATGGCGTGCCGGTGTTGATGCACGATTTCAGCATCGGCCGCATGGCAGGCGACCC 1320 S D G V P V L M H D F S I G R M A G D P GCAGAACCAGTTGGTGTCGCAGGTGCCGTTTGCCGAGCTGCGTGAAATGCCGTTGGTGAT 1380 Q N Q L V S Q V P F A E L R E M P L V I TCGCAACCCGTCTGACGGAAACTACGTCAAGACCGACCAGACCATCGCCGGTGTGGAGCA 1440 R N P S D G N Y V K T D Q T I A G V E Q GATGCTGGAGCACGTGCTCAAAAAGCCCGAGCCGATGTCGGTGGCGCTGGACTGCAAGGA 1500 M L E H V L K K P E P M S V A L D C K E AAACACCGGCGAAGCAGTGGCGATGCTGCTGATGCGCCGGCCGGACCTGCGCAAGGCTGC 1560 N T G E A V A M L L M R R P D L R K A A GGCGATCAAGGTCTATGCCAAGTACTACACGGGCGGCTTTGACCAATTCCTGTCCAATTT 1620 A I K V Y A K Y Y T G G F D Q F L S N L GTACAAGCACTACCAGATCAACCCGTTGCACTCGCAGGATGCGCCACGTCGCGCCGCGCT 1680 Y K H Y Q I N P L H S Q D A P R R A A L GGATCGCTTGCTGGCCAAGATCAACGTGGTGCCGGTCTTGAGCCAGGGCATGTTGAACGA 1740 D R L L A K I N V V P V L S Q G M L N D CGAGCGCTTGCGCGGCTTCTTCCGAAGCAATGAGCAGGGCGCCGCGGGGCTCGCAGACAC 1800 E R L R G F F R S N E Q G A A G L A D T CGCAATGCAGTGGCTGGACAGCTGGACCAAGATGCGCCCGGTGATCGTGGAGGCGGTGGC 1860 A M Q W L D S W T K M R P V I V E A V A CACCGACGACAGCGATGCCGGCAAGGCCATGGAAATGGCTCGGACGCGGATGCGCCAGCC 1920 T D D S D A G K A M E M A R T R M R Q P GGACTCGGCCTACGCGAAGGCCGCGTATTCGGTGAGCTACCGGTATGAGGACTTTTCCGT 1980 D S A Y A K A A Y S V S Y R Y E D F S V GCCGCGCGCCAATCACGACAAGGACTACTACGTTTACCGCAACTTCGGTGAGCTCCAAAA 2040 P R A N H D K D Y Y V Y R N F G E L Q K GCTCACCAACGAGGCCTTCGGCGTCAAGCGCACCACGGCCGGCGCGTTTCGCGACGATGG 2100 L T N E A F G V K R T T A G A F R D D G CGAAAGCCTGTTGACCGATCAGCCCGAGGCCGAATTGCTCGCCATCCTGGAAAACCGCAC 2160 E S L L T D Q P E A E L L A I L E N R T Figure A-7. continued

PAGE 126

126 GCTGGCCCGGGGCCATACCGGCAACGAACTCGACGTACCGCCGGAAACGCCCATCGATAT 2220 L A R G H T G N E L D V P P E T P I D I CAACCGGGACGCCGAGATCGTGAAGCAGCGAACCCAGCAATTTCAGGCCAGCTCCATCCC 2280 N R D A E I V K Q R T Q Q F Q A S S I P CGCCGACCCGAACCACATCGCCGCGGTTCGCGAAGGCAAGCAACACGATCACACCGCAGA 2340 A D P N H I A A V R E G K Q H D H T A D CATGGTCAATGACCCTGCGGCAACGCGTGCGCTGGACAAGCGCGCCAAAGCGCTTGGTTT 2400 M V N D P A A T R A L D K R A K A L G L GCTGACCGACAAATACCGTGGCGCGCCTGTGACCCACTACCTCAATGAGCAGGCCAGGCA 2460 L T D K Y R G A P V T H Y L N E Q A R Q GACCGAGACGGAT TGA ACCTTCTAAGCTCACCGCTTCGGTTGGTAGGTCGAGCGCGTGGG 2520 T E T D GCAGCATGCAGATGCTCCAAGCCACGCTCGGGCAGCGTCAGGGATTGGGAGAGATCACGC 2580 GCCTGCACCGGCAAGCCGGCGCCCGGAGCGGTGGGGCGGCCGATCAGCCGCGTTGCAGTG 2640 CCCTGCGTGCAGGTGCCGCGGCGTGCGCCGCCGCGTTACCTGCATACGCATCCGACGTTG 2700 CCTCCCATCGATTCATCGGTGGAGGCGACTGCGTG 2735 Figure A-7. continued avrBs2R

PAGE 127

127 X.p. MIRRISPGTLRPSVSIEHGGSHHHGDHHSDAAGTSAGSPTDIAARAPRLRPAPPRRRRRG X.e. MIRRISPGTLRPSVSIEHGGSHHHGDHHADASGTSAGSPTDIAARAPRLRPAPPRRRRRG X.o. MIRRISPGPLQPSVSTEHGDSHHNADHHADAAGTSAGSPANIAARATRLRPAPPRRRRRG X.o.2 MIRRISPGPLQPSVSTEHGGSHHHADHHADAASTSAGSPADIAARATRLRPAPPRRRRRG X.p. IRSLDGQEDEFDANEQEEIEAKRECALRGRVSVAITPAQSREHGQGDQHGGDRSTHTDDP X.e. IRSLDGQEDEFDANEQEEIEAKRECALRGRVSVAITPAQSREHGQGDQHGGDRSTHTDDP X.o. NRSLEGQEDEFDANEQEEIEAKRECALRGRVSVAIAPAQSREHGQSDQHGGDRNTHTDDT X.o.2 NRSLDGHEDEFDANEQEEIEAKRECALRGRVSVAIAPAQSREHGQSDQHGGDRNTHTDDT X.p. QAGPWHSAAPAQLSDSIRTCIDAIVDRYVTRRDADPVAKRHALAAALVELRAVGVSHPAV X.e. QAGPWHSAAPAQLSDSIRTCIDAIVDRYVTRRDADPVAKRHALAAALVELRAVGVSHPAV X.o. QAGPWRGAAPAQVTDSIRTCIDAILDRYVARRDADPVAKRNALAAALVELRAIGVSHPAV X.o.2 QAGPWHGGAPRQVTDSIRTCIDAILNRYVARRDADPVAKRDALAAALVELRAIGVSHPAV X.p. APLTATIWRLMREHLRAYDKATAAENLLALRTRLLELMPSELEPVPALRNFHLLLPLILL X.e. APLTATIWRLMREHLRAYDKATAAENLLALRTRLLELMPSELEPVPALRNFHLLLPLILL X.o. APLTETVWRLMREHLSSYDRATAAENLQALRTRLLELMPSELEPVPALRNFHLLLPLILL X.o.2 APLTETVWRLMREHLSSYDKATAAENLQTLRTRLLELMPSELEPVPALRNFHLLLPLILL X.p. NAEKPRRRIDRTHAITRLNTLLIEQPQQAAQEVRP X.e. NAEKPRRRVDRTHAITRLNTLLIEQPQQAAQEVRP X.o. NAEKPRRQVDRRHAITRLNTLLIEQPQQAAQEVRP X.o.2 NAEKPRRQVDRRHAITRLNTLLIEQPEQAAQEVRP Figure A-8. Similarity of amino acid sequence of HpaA in X. perforans (X.p.) to homologs in X. euvesicatoria (X.e.), X. oryzae pv. oryzicola (X.o.) and X. oryzae pv. oryzae (X.o.2) using the multiple sequence alignment tool CLUSTALW. Mismatch AA sequences are indicated by an asterisk. * * * * * * * *** ** * * * ** * ** *

PAGE 128

128 X.p. MSSARFETIVRQMCEALDLPDVESVLDRRVLWVEGFEVYLHLPTPQPEDD X.e. MSSARFETIVRQMCEALDLPDVESVLDRRVLWVEGFEVYLHLPTPQPEDD X.o. MSSARFETIVRQMCESLDLPDVESVLSRRVLWVEGFEVYLHLPTPQPGDD X.o.2 MSTARFETIVRQMCESLDLPDVDSVLSRRVLWVEGFEVYLHLPTPQPGDD X.p. VKEEALYLRIAYGLPPAGRTLTVFRLLLEANLSVYAQDQAQLGLNDDGVI X.e. VKEEALYLRIAYGLPPAGRTLTVFRLLLEANLSVYAQDQAQLGLNDDGVI X.o. AQQEALYLRIAYGLPPAGRTLTVFRLLLEANLSVYAQDQAQLGLNDNGVI X.o.2 AQEEALYLRIAYGLPPAGRTLNVFRLLLEANLSVYAQDQAQLGLNDDGVI X.p. VLIVRVPLDDDVDGAWICDLLAHYAEHGRYWNNNIFVAHDEMFEGIATGN X.e. VLIVRVPLDDDVDGAWICDLLAHYAEHGRYWNNNIFVAHDEMFEGIATGN X.o. VLIVRVPLDDDVDGAWICDLLAHYAEHGRYWNNNIFVAHDEMFEGIATGN X.o.2 VLIVRVPLDNDVDGAWICDLLAHYAEHGRYWNNNIFVAHDEMFEGIATGN X.p. YLWLRA X.e. YLWLRA X.o. YLWLRA X.o.2 YLWLRA Figure A-9. Similarity of ami no acid sequence of HpaB in X. perforans (X.p.) to homologs in X. euvesicatoria (X.e.), X. oryzae pv. oryzicola (X.o.) and X. oryzae pv. oryzicola (X.o.2) using the multiple sequence ali gnment tool CLUSTALW. Mismatch AA sequences are indicated by an asterisk. ** * * *

PAGE 129

129 X.p. MRKPPLRHVRILPVSGALQRPAAPATPARSALRSSFLQLRQRLRSAQLALPCMVLPPQCD X.e. MRKPPLRHVRILPVSGALQRPAAPATPARSALRSSFLQLRQRLRSAQLALPCMVLPPQCD X.o. MRKLPLRHVRILPVSGALQRPATPATPARSAQRSSFLQLRQRLRSVELALPCMVLPPQCD X.o.2 --------MRILPVSGALQRPATPATPARSAQRSSFLQLRQRLRSVELALPCMVLPPQCD X.p. -EDRPEPDAEEGFTEAHDSVPVQTDPPLRIEGTKHQEPSQGNGDGAVGRQIATEWIRTQR X.e. -EDRPGPDAEEGFTEAHDSVPVQTDPPLRIEGTKHQEPSQGNDDGAVGRQIATEWIRTQR X.o. DEDRPEPDAEERFTEARDSAPVQTDPPLRLDGPKHREPPQGNDNGAVGRQIATEWIRTQR X.o.2 DEDRPEPDAEERFTEARDSAPVQTDPPLRLDGPKHREPPQGNDNGAVGRQIATEWIRTQR X.p. AHMAIDHIALRVAEFCNAQPVRSAGSWEAWLAIDQEVVAQTTLFLRLSPDQLSLRFNTSS X.e. AHMAIDHIALRVAEFCNAQPVRSAGSWEAWLAIDQEVVAQTTLFLRLSPDQLSLRFNTSS X.o. AQMAIDHIALRVADFCNAKPVRSAGSWEAWLDIDQEVVAQTTLFLRLSPHQLSLRFNTSS X.o.2 AQMAIDHIALRVADFCNAKPVRSAGSWEAWLDIDQEVVAQTTLFLRLSPHQLSLRFNTSS X.p. PDAREVLWCGKQRLEAALTSTLSSTLQISIEVV X.e. PDAREVLWCGKQRLEAALTSTLSSTLQISIEVV X.o. PDAREVLWCGRQRLEAALTSTLSSTLQISIEVV X.o.2 PDAREVLWCGRQRLEAALTSTLSSTLQISIEVV Figure A-10. Similarity of am ino acid sequence of HpaC in X. perforans (X.p.) to homologs in X. euvesicatoria (X.e.), X. oryzae pv. oryzicola (X.o.) and X. oryzae pv. oryzicola (X.o.2) using the multiple sequence ali gnment tool CLUSTALW. Mismatch AA sequences are indicated by an asterisk. ********* * ** * ** * ** * ******

PAGE 130

130 X.p. MPRALLAARRVFEVIMRIGPLQPSIAHTAAPALPTHTSAISPTQVPHMPGNTPPLRERPR X.e. ---------------MRIGPLQPSIAHTAAPALPTHTSAISPTQVPHMPGNTPPLRERPR X.c. ---------------MRIGPLQPSVAHTAAPALPTHTSVVSPTQVPHMPGDIPPLRERPR X.o. -----------MRSFMRIGPPQTSIAHTDALAIPTHTSASSPTQVPHVQGNTPPLRERAP X.p. RRATNMPPLVPLNDSAMTGKPALVALDSEFSEQRLAEVQARQITVQTLQAKLATHLAQAG X.e. RRAGNMPPLVPLNDSAMTGKPALVALDSEFSEQRLAEVQARQITVQTLQGKLATHLAQAG X.c. RRADSAPPLVPLNDSAMTGKPALVALDSEFSEQRLAEVQARQITVQTLQTKLATHLAQAG X.o. RRADSAPPLVPLNDSAMTGKPALVALDSEFSEQRLAEVQARQITVQTLQTKLATHLAQAG X.p. TALKPDSIAARFAAGTLEPVYLDTAAFNAMSRGLPARARAAAGPVLIDAQQGRIIFNLQR X.e. TALKPDSIAARFAAGTLEPVYLDTAAFNAMSRGLPARARAAAGPVLIDAQQGRIIFNLQR X.c. TPLKPDSIAARFAAGTLEPVYLDTAAFNAMSRGLPARARAAAGPVLIDAQQGRIIFNLQR X.o. TALTPDSIAARFAAGALEPVYLDTAAFNAMSRGLPARARAASGPVLIDAQQCRIVFNLQR X.p. AFAPGDTFSDAALAALGKQLNLSGHGLATPNWLQPAAGTPGRRKLQQAARYHGHEVPARD X.e. AFAPGDTFSDAALAALGKQLNLSGHGLATPNWLQPAAGTPGRRKLQQAARYHGHEVPARD X.c. AFAPGDSSSDAALAALGKQLNLSGHGLATPNWLQPAARTPARRKLQQAARYHGHEVPARD X.o. AFASGDTFSDAALTALGKALDLPGHGLATPDWLQPAARTPTRRKLQHAPRYHGHEVPARD X.p. GGAGFFKANDHRLLEGKQVLLRNHQKSLVHNHYFEAPSTRASGKDVMVHRGLFDNHAGIP X.e. GGAGFFKANDHRLLEGKQVLLRNHQKSLVHNHYFEAPSTRAFGKDVMVHRGLFDNHAGIP X.c. GGAGFFKANDHRLLEGKQALLRNHRQSLVHDHYFEAPSTRAFGKDVMVHRGLFDNHAGIP X.o. GGAAFFKANDHRLLEGKQALLRNHRKALVHDHYFEAPSTRALGKDVMVHRGLFDNHAGIP X.p. ENSLASIDHAYEVDQGYRNLELDVEVSSDGVPVLMHDFSIGRMAGDPQNQLVSQVPFAEL X.e. ENSLASIDHAYE--QGYRNLELDVEVSSDGVPVLMHDFSIGRMAGDPQNRLVSQVPFAEL X.c. ENSLASIDHAYE--QGYRNLELDVEVSADGVPVLMHDFSVGRMAGDPQNRLVSQVPFAEL X.o. ENSLSSIDNAYA--KGYRNLELDVEVSADGVPVLMHDFSVGRMAGDPQNRLVSQVPFAEL X.p. REMPLVIRNPSDGNYVKTDQTIAGVEQMLEHVLKKPEPMSVALDCKENTGEAVAMLLMRR X.e. REMPLVIRNPSDGNYVKTDQTIAGVEQMLEHVLKKPEPMSVALDCKENTGEAVAMLLMRR X.c. REMPLVIRNPSDGNYVKTDQTIAGVEQMLEHVLKKPEPMSVALDCKENTGEAVAMLLMRR X.o. REMPLVIRNPSDGNYVKTDQTIPGVEQMLEHVIKKPEPMSVALDCKENTGEAVAMLLMRR X.p. PDLRKAAAIKVYAKYYTGGFDQFLSNLYKHYQINPLHSQDAPRRAALDRLLAKINVVPVL X.e. PDLRKAAAIKVYAKYYTGGFDQFLSNLYKHYQINPLHSQDAPRRAALDRLLAKINVVPVL X.c. PDLRKAAAIKVYAKYYTGGFDQFLSNLYKHYQINPLHSQDAPRRAALGRLLAKINVVPVL X.o. PDLRNAAAIKVYAKYYTGGFDQFLSNLYKHYQINPLHSQDAPRRAALDRLLAKINVVPVL X.p. SQGMLNDERLRGFFRSNEQGAAGLADTAMQWLDSWTKMRPVIVEAVATDDSDAGKAMEMA X.e. SQGMLNDERLRGFFRSNEQGAAGLADTAMQWLDSWTKMRPVIVEAVATDDSDAGKAMEMA X.c. SQGMLNDERLRGFFRSNEQGAAGLADTAMQWLDSWTKMRPVIVEAVATDDSDAGKAMEAA X.o. SQAMLNDEHLRGFFRSNDDGAEGLADTAMQWLESWTRMRPVIVEAVATDDSDAGKAMAMA X.p. RTRMRQPDSAYAKAAYSVSYRYEDFSVPRANHDKDYYVYRNFGELQKLTNEAFGVKRTTA X.e. RTRMRQPDSAYAKAAYSVSYRYEDFSVPRANHDKDYYVYRNFGELQKLTNEAFGVKRTTA X.c. RARMRQPDSAYAKAAYSVSYRYEDFSVPRANHDKDYYVYRNFGELQKLTDEAFGVKRTTA X.o. RERMRQPDSAYAKAAYSVSYRFEDFSVPRANHDRDYYVYRNFGELQKLTNEEFGVKRTTA Figure A-11. Similarity of ami no acid sequence of AvrBs2 in X perforans (X.p.) to homologs in X. euvesicatoria (X.e.), X. axonopodis pv. citri (X.c.) and X oryzae pv. oryzae (X.o.) using the multiple sequence alignment tool CLUSTALW. Mismatch AA sequences are indicated by an asterisk. ***************************** *** ****** * * ***** ***** * ******* ** ********* *****

PAGE 131

131 X.p. GAFRDDGESLLTDQPEAELLAILENRTLARGHTGNELDVPPETPIDINRDAEIVKQRTQQ X.e. GAFRDDGESLLTDQPEAELLAILENRALARGHTGNELDVPPETPIDINRDAEIVKQRTQQ X.c. GAFRDDGESLLTDQPEAELLAILENRTLARGHTGNELDVPPETPIDINRDAEIVKQRTQQ X.o. GAFRDDGESLLTDQAEAELLAILENRTLARGHTGNELDVPPETPIDINRDAEIVKQRTRE X.p. FQASSIPADPNHIAAVREGKQHDHTADMVNDPAATRALDKRAKALGLLTDKYRGAPVTHY X.e. FQASSIPADPNHISAVREGKQHDHTADMVNDPAATRALDKRAKALGLLTDKYRGAPVTHY X.c. FQASSIPADPNHIAAVREGRQRDHTADMVNDPAATRAMDKRAKALGVLTDKYRGAPVTHY X.o. FQAGSIPADPNHIAAVREGKQRDHRADMVHDPAATRAVDKRAQASGLLTEKYRGAPVTHY X.p. LNEQARQTETD X.e. LNEQARQTETD X.c. LNEQAKQTETD X.o. LNERANQTEPE Figure A-11. continued. *** *** ******* ****

PAGE 132

132 X.p. MINSLNTSHLGVDSSFMQVNPDQFQKFDSNQSNQGISEKQLDQLLTQFIFSMLLQDDNAD X.e. MINSLNTSHLGVDSSFMQVNPDQFQKFDSNQSNQGISEKQLDQLLTQFIFSMLLQDDNAD X.c. -MNSLNT-QIGANSSFLQVDPSQNTQFGPNQGNQGISEKQLDQLLTQLIMALLQQSNNAD X.g. -MNSLNT-QLGANSSFFQVDPGQNTQSSPNQGNQGISEKQLDQLLTQLIMALLQQSNNAE X.p. ----------------DSPNSDKPTDFPSPRTQMLMNVIGDILQAKNGGRLGGLSDGGLN X.e. ----------------DSPNSDKPTDFPSPRTQMLMNVIGDILQAKNGGRLGGLSDGGLN X.c. QGQG----GDSGGQGGNSRQAGQPNGSPSAYTQMLMNIVGDILQAQNGGGFGGGFGGGFG X.g. QGQGQGQGGDSGGQGGNPRQAGQSNGSPSQYTQALMNIVGDILQAQNGGGFGGGFGGGFG X.p. ------TSLSLSGDTASMQ X.e. ------TSLSLSGDTASMQ X.c. GGLGTSLGSSLASDTGSMQ X.g. ILVTSLASDTGSMQ Figure A-12. Similarity of am ino acid sequence of XopA in X perforans (X.p.) to homologs in X euvesicatoria (X.e.), X citri (X.c.) and X axonopodis pv. glycines (X.g.) using the multiple sequence alignment tool CLUSTALW. Mismatch AA sequences are indicated by an asterisk. ** *** * ****** * *** * *************************** ** ** ** ** ** ********* **

PAGE 133

133 X.p. MEYIPRYEALEKVPVRVGFHAYLRGDGSFGPGLSGILRYMTPDQKKRLYL X.e. MEYIPRYEALEKVPVRVRFHAYLRGDGSFGPGLPGILRYMTPDQKKRLYL X.c. ----------------------MESQDPAGSSYSSYSSLIPPTPSG---X.e. ASERRKLALAAPKSK------PLKGVFRTLHQKPNLLLEISSKFSNRAYS X.e. ASERRKLALAAPKSKPTPKSKPLKGVFRTLHQKPNLLLEISSKFSNRAYS X.c. ---------GWPQN-----------------------------ASGEWYP X.p. INDSSSAYLSQADLEEMVDEETGELTRLGEAVISGASQGIQTAIRANFRM X.e. INDSSSGYLSQADLEEMVDEETGELTRLGEAVISGASQGIQTAIRANFRM X.c. DTPADSSYRAWPARPEASSSTFDDLESLDS--------------RRNY-X.p. RYQQPDLPPYSPPQAFHRPEETWNPHTPAGSSYSSLFPPTPSGGWPQNAS X.e. RYQQPDLPPYSPPQAFHRPEETWNPHTPAGSSYSSLFPPTPSGGWPQNAS X.c. SYREFDLN---TPQEIEQP------------------------------X.p. GEWHPDTPADSSYSSLFPPTPSGGWPQNASGEWHPDTPADSSYSSLFPPT X.e. -------------------------------------------------X.c. -------------------------------------------------X.p. PSGGWPQNASGEWHPDTPAGYSHRAWPAQPEASSSTFDDLESLDYRQNYG X.e. ----------GEWHPDTPAGYSHRAWPAQPEASSSTFDDLESLDYRQNYG X.c. ----------GWWQHATSAGSSYRAWPAQPEASSSTFDDLESLDSRQNYG X.p. YREFDLNTPQEIEQPGWWQQATPAQSTDSTFDGLSSMSHYGSEFDLNIPQ X.e. YREFDLNTPQEIEQPGWWQQATPAQSTDSTFDGLSSMSHYGSEFDLNIPQ X.c. YREFDLNTPQEIEQPG-WQYATPAQSTDSTFDGLSSMSHYGREFDLNTPQ X.p. QEEYPNNHGTQTPMGYSAMTPERIDVDNLPSPQDVADPELPPVRATSWLL X.e. QEEYPNNHGTQTPMGYSAMTPERIDVDNLPSPQDVADPELPPVRATSWLL X.c. EEDEPWDYGTQTPVGHSAMSPERIDVDNLPSPQDVADPELPQVTDTSWLL X.p. DGHLRAYTDDLARRLRGEPNAHLLHFADSQVVTMLSSADPDQQARAQRLL X.e. DGHLRAYTDDLARRLRGEPNAHLLHFADSQVVTMLSSADPDQQARAQRLL X.c. DGHLRAYTDDLARRLRGQPNAHLLHFADSQVVTMLSSTDPGQQARARRLL X.p. AGDDIPPIVFLPINQPNAHWSLHVVDRRNKDAVAAYHYDSMAQKDPQQRY X.e. AGDDIPPIVFLPINQPNAHWSLLVVDRRNKDAVAAYHYDSMAQKDPQQRY X.c. VGDDVPPIVFLPINQPNFHWSLLVVDRRNKDAAAAYYYDSMAQTQPQQRY X.p. LADMAAYHLGLDYQQTHEMPIAIQSDGYSCGDHVLTGIEVLAHRVLDGTF X.e. LADMAAYHLGLDYQQTHEMPIAIQSDGYSCGDHVLTGIEVLAHRVLDGTF X.c. LADMAAYHLGLDYKEIHEMPTAIQPDGYSCGDHVLTGIETLAHRVIDGTF X.p. DYAGGRDLTDIEPDRGLI------------RDRLAQAEQAPAESSIRQVP X.e. DYAGGRDLTDIEPDRGLI------------RDRLAQAEQAPAESSIRQVP X.c. DSADGRDLSEIAPGRGLITDRLAQAQAQAERDRLAQAEQAPAESSVRQVS Figure A-13. Similarity of am ino acid sequence of XopD in X perforans (X.p.) to homologs in X. euvesicatoria (X.e.) and X campestris pv. campestris (X.c.) using the multiple sequence alignment tool CLUSTALW. Mism atch AA sequences are indicated by an asterisk. ***************************** *********** ******** *********** ******************************** *** ***** *************** ** **************** *** *** ***** **** ******************************* ************************************************** ********** *** * * * * ** *** * * ** * * * * * ** *** * * * ** * ************ *

PAGE 134

134 X.p. HDPTNRRKRKASGGKSPSSSTISGKVGLAGLGKV X.e. ARSNEQKKKKSKWWKKF----------------X.c. ERSIEQKKKKSKWWKKF----------------Figure A-13. continued. ******* **** ******************

PAGE 135

135 LIST OF REFERENCES 1. Alfano, J. R. and A. Collmer. 2004. Type III secretion system effector proteins: Double agents in bacterial disease and pl ant defense. Annu. Rev. Phytopathol. 42: 385-414. 2. Arts, J., R. van Boxtel, A. Filloux, J. Tommassen, and M. Koster. 2007. Export of the pseudopilin XcpT of the Pseudomonas aeruginosa type II secretion system via the signal recognition particle-sec pa thway. J. Bacteriol. 189: 2069-2076. 3. Astua-Monge, G., G. V. Minsavage, R. E. Stall, C. E. Vallejos, M. J. Davis, and J. B. Jones. 2000. Xv4-avrXv4 : a new gene-for-gene interaction identified between Xanthomonas campestris pv. vesicatoria race T3 and wild tomato relative Lycopersicon pennellii Mol. Plant Microbe Interact. 13: 1346-1355. 4. Astua-Monge, G., G. V. Minsavage, R. E. Stall, M. J. Davis, U. Bonas, and J. B. Jones. 2000. Resistance of tomato and pepper to T3 strains of Xanthomonas campestris pv. vesicatoria is specified by a plant-inducible avir ulence gene. Mol. Plant Microbe Interact. 13: 911-921. 5. Balogh, B., J. B. Jones, M. T. Momol, S. M. Olson, A. Obradovic, P. King, and L. E. Jackson. 2003. Improved efficacy of newly formulat ed bacteriophages for management of bacterial spot of tomato. Plant Dis. 87: 949-954. 6. Bashan, Y., S. Diab, and Y. Okon. 1982. Survival of Xanthomonas campestris pv. vesictoria in pepper seeds and roots in symptomless and dry leav es in non-host plants and in the soil. Plant Soil 68: 161-170. 7. Becker, A., F. Katzen, A. Puhler, and L. Ielpi. 1998. Xanthan gum biosynthesis and application: a biochemical/genetic pers pective. Appl. Microbiol. Biotechnol. 50: 145-152. 8. Bieler, S., F. Silva, C. Soto, and D. Belin. 2006. Bactericidal activity of both secreted and nonsecreted Microcin E492 requires th e mannose permease. J. Bacteriol. 188: 7049-7061. 9. Bogdanove, A. J., J. F. Kim, Z. Wei, P. Kolchinsky, A. O. Charkowski, A. K. Conlin, A. Collmer, and S. V. Beer. 1998. Homology and functional sim ilarity of an hrp-linked pathogenicity locus, dspEF, of Erwinia amylovora and the avirulence locus avrE of Pseudomonas syringae pathovar tomato. PNAS 95: 1325-1330. 10. Bouzar, H., J. B. Jones, R. E. Stall, N. C. Hodge, and N. Daouzli. 1994. Physiological, chemical, serological, and pathologenic analyses of worldwide collection of Xanthomonas campestris pv. vesicatoria strains. Phytopathology 31: 753-755. 11. Bouzar, H., J. B. Jones, R. E. Stall, M. Schneider, J. L. Rademaker, F. J. de Bruijn, and L. E. Jackson. 1999. Multiphasic analysis of Xanthomonads causing bacterial spot

PAGE 136

136 disease on tomato and pepper in the caribbean and central america: Evidence for common lineages within and between countries. Phytopathol. 89: 328-335. 12. Bonas, U., R. Shulte, S. Fenselau, G. V. Mi nsavage, B. J. Staskawicz, and R. E. Stall. 1991. Isolation of a gene-cluster from Xanthomonas campestris pv. vesicatoria that determines pathogenicity and the hypersen sitive response on peppe r and tomato. Mol. Plant-Microbe Interact. 4: 81-88. 13. Bond, J. S. and P. E. Butler. 1987. Intracellular Proteases. Annu. Rev. Biochem. 56: 333364. 14. Braun, V., S. I. Patzer, and K. Hantke. 2002. Ton-dependent colicins and microcins: Modular design and evolution. Biochimie 84: 365-380. 15. Buell, C. R., V. Joardar, M. Lindeberg, J. Sele ngut, I. T. Paulsen, M. L. Gwinn, R. J. Dodson, R. T. Deboy, A. S. Durkin, J. F. Kolonay, R. Madupu, S. Daugherty, L. Brinkac, M. J. Beanan, D. H. Haft, W. C. Ne lson, T. Davidsen, N. Zafar, L. Zhou, J. Liu, Q. Yuan, H. Khouri, N. Fedorova, B. Tran, D. Russell, K. Berry, T. Utterback, S. E. Van Aken, T. V. Feldblyum, M. D'Ascenzo W. Deng, A. R. Ramos, J. R. Alfano, S. Cartinhour, A. K. Chatterjee, T. P. Delane y, S. G. Lazarowitz, G. B. Martin, D. J. Schneider, X. Tang, C. L. Bender, O. White, C. M. Fraser, and A. Collmer. 2003. The complete genome sequence of the Arabidopsis and tomato pathogen Pseudomonas syringae pv. tomato DC3000. PNAS 100: 10181-10186. 16. Buttner, D. and U. Bonas. 2002. Getting across-bacterial type III effector proteins on their way to the plant cell. EMBO J. 21: 5313-5322. 17. Buttner, D., D. Gurlebeck, L. D. Noel, and U. Bonas. 2004. HpaB from Xanthomonas campestris pv. vesicatoria acts as an exit control protein in type III-dependent protein secretion. Mol. Microbiol. 54: 755-768. 18. Buttner, D., C. Lorenz, E. Weber, and U. Bonas. 2006. Targeting of two effector protein classes to the type III secr etion system by a HpaCand HpaB-dependent protein complex from Xanthomonas campestris pv. vesicatoria Mol. Microbiol. 59: 513-527. 19. Wilson, C.L. 1997. Biological control and plant dis eases-a new paradigm. Journal of Industrial Microbio logy and Biotechnology 19: 158-159. 20. Cascales, E. and P. J. Christie. 2003. The versatile bacterial t ype IV secretion systems. Nat. Rev. Microbiol. 1: 137-149. 21. Chou, F., H. Chou, Y. Lin, B. Yan g, N. Lin, S. Weng, and Y. Tseng. 1997. The Xanthomonas campestris gumD Gene required for synthesis of xanthan gum is involved in normal pigmentation and virulence in causing black rot. Biochem. Biophys. Res. Commun. 233: 265-269.

PAGE 137

137 22. Chuang, D., A. G. Kyeremeh, Y. Gunji, Y. Takahara, Y. Ehara, and T. Kikumoto. 1999. Identification and cloning of an Erwinia carotovora subsp. carotovora bacteriocin regulator gene by insertional mutagenesis. J. Bacteriol. 181: 1953-1957. 23. Chen, L., D. Chen, J. Miaw, and N. Hu. 1996. XpsD, an outer membrane protein required for protein secretion by Xanthomonas campestris pv. campestris forms a multimer. J. Biol. Chem. 271: 2703-2708. 24. Collmer, A., M. Lindeberg, T. Petnicki-Ocwieja, D. J. Schneider, and J. R. Alfano. 2002. Genomic mining type III secr etion system effectors in Pseudomonas syringae yields new picks for all TTSS prosp ectors. Trends Microbiol. 10: 462-469. 25. Conover, R. A. and N. A. Gerhold. 1981. Mixtures of copper and maneb or mancozeb for control of bacterial s pot of tomato and their compatabil ity for control of fungus diseases. Proc. Fla. State Hort. Soc. 94: 154-156. 26. Cursino, L., J. Smarda, E. Chartone-Souza, and A. M. Nascimento. 2002. Recent updated aspects of colicins of Ente robacteriaceae. Braz. J. Microbiol. 33: 185-195. 27. Da Silva, A. C. R., J. A. Ferro, F. C. Reinach, C. S. Farah, L. R. Furlan, R. B. Quaggio, C. B. Monteiro-Vitorello, M. A. V. Sl uys, N. F. Almeida, L. M. C. Alves, do Amaral, A. M., M. C. Bertolini, L. E. A. Camargo, G. Camarotte, F. Cannavan, J. Cardozo, F. Chambergo, L. P. Ciapina, R. M. B. Cicarelli, L. L. Coutinho, J. R. Cursino-Santos, H. El-Dorry, J. B. Faria, A. J. S. Ferreira, R. C. C. Ferreira, M. I. T. Ferro, E. F. Formighieri, M. C. Franco, C. C. Greggio, A. Gruber, A. M. Katsuyama, L. T. Kishi, R. P. Leite, E. G. M. Lemos, M. V. F. Lemos, E. C. Locali, M. A. Machado, A. M. B. N. Madeira, N. M. Martinez-Rossi, E. C. Martins, J. Meidanis, C. F. M. Menck, C. Y. Miyaki, D. H. Moon, L. M. Moreira, M. T. M. Novo, V. K. Okura, M. C. Oliveira, V. R. Oliveira, H. A. Perei ra, A. Rossi, J. A. D. Sena, C. Silva, de Souza, R. F., L. A. F. Spinola, M. A. Takit a, R. E. Tamura, E. C. Teixeira, R. I. D. Tezza, Trindade dos Santos, M., D. Truffi, S. M. Tsai, F. F. White, J. C. Setubal, and J. P. Kitajima. 2002. Comparison of the genomes of two Xanthomonas pathogens with differing host specificities. Nature 417: 459-463. 28. Daniels, M. J., C. E. Barber, D. C. Turner, W. G. Cleary, and M. Sawage. 1984. Isolation of mutants of Xanthomonas campestris pv. campestris showing altered pathogenicity. J. Gen. Microbiol. 130: 2447-2455. 29. Davis, R. M., W. T. Lanini, G. Hamilton, T. H. Spreen, and C. Osteen. 1998. The importance of pesticides and other pest mana gement practices in U.S. tomato production. USDA. 30. De los Santos, P. E., A. H. Parret, and R. De Mot. 2005. Stress-related Pseudomonas genes involved in production of bacter iocin LlpA. FEMS Microbiology Letters 244: 243250.

PAGE 138

138 31. DebRoy, S., R. Thilmony, Y. Kwack, K. Nomura, and S. Y. He. 2004. A family of conserved bacterial effectors i nhibits salicylic acid-mediated basal immunity and promotes disease necrosis in plants. PNAS 101: 9927-9932. 32. Dianese, A. C. 1997. Importance of nutritional similar ity in pre-emptive biocontrol of bacterial spot. Masters Thesis. Auburn University, Auburn, Ala. 33. Doidge, E. M. 1921. A tomato canker. Ann. Appl. Biol. 7: 407-430. 34. Dow, J. M., B. R. Clarke, D. E. Milligan, J. L. Tang, and M. J. Daniels. 1990. Extracellular proteases from Xanthomonas campestris pv. campestris the black rot pathogen. Appl. Environ. Microbiol. 56: 2994-2998. 35. Dye, D. W., J. F. Bradbury, R. S. Dickey, M. Goto, and A. C. Hayward. 1980. international standards for naming pathovars of phytopathogenic bacteria and a list of pathovar names and pathotype strains. Rev. Plant Pathol. 59: 156-168. 36. Echandi, E. 1976. Bacteriocin production by Corynebacterium michiganensis Phytopathology 66: 430-432. 37. Echandi, E. and J. W. Moyer. 1979. Production, properties, and morphology of bacteriocins for Erwinia chrysanthemi Phytopathology 69: 1204-1207. 38. El-Ghaouth, A. 1997. Biologically-based al ternatives to synthetic fungicides for the control of postharvest diseases. Journal of Industrial Microbiology and Biotechnology 19: 160-162. 39. El-Morsy, G. A., G. C. Somodi, J. W. Scott, R. E. Stall, and J. B. Jones. 1994. Aggressiveness of Xanthomonas campestris pv. vesicatoria tomato race strains over tomato race 1 strains: Evidence for antagonism. Phytopathology 84: 551-556. 40. Ennahar, S., T. Sashihara, K. Sonomoto, and A. Ishizaki. 2000. Class IIa bacteriocins: biosynthesis, structure and ac tivity. FEMS Microbiology Reviews 24: 85-106. 41. Etchebar, C., D. Trigalet-Demery, G. Gijsegem, J. Vasse, and A. Trigalet. 1998. Xylem colonization by a hrcV mutant of Ralstonia solanacearum is a key factor for the efficient biological control of tomato bacterial wilt. Mol. Plant Microbe Interact. 11: 869877. 42. Fath, M. J., R. Skvirsky, L. Gilson, H. K. Mahanty, and R. Kolter. 1992. The secretion of colicin V., p. 331-348. In R. James, F. Laz dunski and F. Pattus (ed.), Bacteriocins, Microcins, and Lantibiotics. Springer-Verlag, Heidelberg. 43. Fath, M. J., L. H. Zhang, J. Rush, and R. Kolter. 1994. Purification and characterization of Colicin V from Escherichia coli culture supernatants. Biochemistry 33: 6911-6917.

PAGE 139

139 44. Fenselau, S., I. Balbo, and U. Bonas. 1992. Determinants of pathogenicity in Xanthomonas campestris pv. vesicatoria are related to proteins involved in secretion in bacterial pathogens of animals. Mol. Plant Microbe Interact. 5: 390-396. 45. Fett, W. F., M. F. Dunn, G. T. Maher, and B. E. Maleeff. 1987. Bacteriocins and temperate phage of Xanthomonas campestris pv. glycines. Curr. Microbiol. 16: 137-144. 46. Flaherty, J. E., J. B. Jones, B. K. Harbaugh, G. C. Somodi, and L. E. Jackson. 2000. Control of bacterial spot of tomato in the greenhouse and field with h-mutant bacteriophages. HortScience 35: 882-884. 47. Fravel, D. R. 2005. Commercialization and implemen tation of biocontrol. Annu. Rev. Phytopathol. 43: 337-359. 48. Frey, P., J. Smith, L. Albar, P. Prior, G. Saddler, D. Trigalet-Demery, and A. Trigalet. 1996. Bacteriocin typing of Burkholderia ( Pseudomonas ) solanacearum race 1 of the French West Indies and correlation with genom ic variation of the pa thogen. Appl. Environ. Microbiol. 62: 473-479. 49. Frey, P., P. Prior, C. Marie, A. Kotouj ansky, D. Trigalet-Dem ery, and A. Trigalet. 1994. Hrpmutants of Pseudomonas solanacearum as potential biocontrol agents of tomato bacterial wilt. Appl. Environ. Microbiol. 60: 3175-3181. 50. Furutani, A., S. Tsuge, K. Ohnishi, Y. Hikic hi, T. Oku, K. Tsuno, Y. Inoue, H. Ochiai, H. Kaku, and Y. Kubo. 2004. Evidence for HrpXo-dependent expression of type II secretory proteins in Xanthomonas oryzae pv. oryzae J. Bacteriol. 186: 1374-1380. 51. Gardner, M. W. and J. B. Kendrick. 1923. Bacterial spot of tomato and pepper. Phytopathology 13: 306-315. 52. Gardner, M. W. and J. B. Kendrick. 1921. Bacterial spot of tomato. J. Agric. Res. 21: 123-156. 53. Gassmann, W., D. Dahlbeck, O. Chesnokova, G. V. Minsavage, J. B. Jones, and B. J. Staskawicz. 2000. Molecular evolution of virulen ce in natural field strains of Xanthomonas campestris pv. vesicatoria. J. Bacteriol. 182: 7053-7059. 54. Gijsegem, F. V., J. Vasse, R. D. Rycke, P. Castello, and C. Boucher. 2002. Genetic dissection of the Ralstonia solanacearum hrp gene cluster reveals that the HrpV and HrpX proteins are required for Hrp p ilus assembly. Mol. Microbiol. 44: 935-946. 55. Gilson, L., H. K. Mahanty, and R. Kolter. 1990. Genetic analysis of an MDR-like export system: the secretion of Colicin V. EMBO J. 9: 3875-3884. 56. Gilson, L., H. K. Mahanty, and R. Kolter. 1987. Four plasmid genes are required for colicin V synthesis, export, and immunity. J. Bacteriol. 169: 2466-2470.

PAGE 140

140 57. Goodman, R. N. 1982. Phytopathogenic prokaryotes. In M. S. Mount and G. H. Lacy (ed.), The Infection Process. Academic Press, New York. 58. Gurlebeck, D., F. Thieme, and U. Bonas. 2006. Type III effector proteins from the plant pathogen Xanthomonas and their role in the interaction wi th the host plant. J. Plant Physiol. 163: 233-255. 59. He, S. Y. and Q. Jin. 2003. The Hrp pilus: learning from flagella. Curr. Opin. Microbiol. 6: 15-19. 60. Hert, A. P. 2001. Relative importance of bacteriocin-lik e genes in antagonism of T3 to T1 strains of Xanthomonas campestris pv. vesicatoria Masters Thesis. University of Florida, 61. Hert, A. P., P. D. Roberts, M. T. Momol, G. V. Minsavage, S. M. Tudor-Nelson, and J. B. Jones. 2005. Relative importance of bacterio cin-like genes in antagonism of Xanthomonas perforans tomato race 3 to Xanthomonas euvesicatoria tomato race 1 strains. Appl. Environ. Microbiol. 71: 3581-3588. 62. Horns, T. and U. Bonas. 1996. The rpoN gene of Xanthomonas campestris pv. vesicatoria is not required for pathogenicit y. Mol. Plant Microbe Interact. 9: 856-859. 63. Huguet, E., K. Hahn, K. Wengelnik, and U. Bonas. 1998. hpaA mutants of Xanthomonas campestris pv. vesicatoria are affected in pathogenicity but retain the ability to induce host-specific hypersensi tive reaction. Mol. Microbiol. 29: 1379-1390. 64. Hwang, J., X. Zhong, and P. Tai. 1997. Interactions of dedi cated export membrane proteins of the colicin V secretion system : CvaA, a member of the membrane fusion protein family, interacts with CvaB and TolC. J. Bacteriol. 179: 6264-6270. 65. Jack, R. W., J. R. Tagg, and B. Ray. 1995. Bacteriocins of Gram-positive bacteria. Microbiol. Rev. 59: 171-200. 66. James, R., C. N. Penfold, G. R. Moore, and C. Kleanthous. 2002. Killing of E. coli cells by E group nuclease colicins. Biochimie 84: 381-389. 67. Ji, P. and M. Wilson. 2003. Enhancement of population size of a biological control agent and efficacy in control of bacterial speck of tomato through sali cylate and ammonium sulfate amendments. Appl Environ. Microbiol. 69: 1290-1294. 68. Jones, J. B., H. Bouzar, G. C. Somodi, R. E. Stall, K. Pernezny, G. El-Morsy, and J. W. and Scott. 1998. Evidence for the preemptive nature of tomato race 3 of Xanthomonas campestris pv. vesicatoria in Florida. Phytopathol. 88: 33-38. 69. Jones, J. B., G. H. Lacy, H. Bouzar, R. E. Stall, and N. W. Schaad. 2004. Reclassification of the xanthomonads associated with bacterial spot disease of tomato and pepper. Syst. Appl. Microbiol. 27: 755-762.

PAGE 141

141 70. Jones, J. B., K. L. Pohronezny, R. E. Stall, and J. P. Jones. 1986. Survival of Xanthomonas campestris pv. vesicatoria in Florida on tomato crop residue, weeds, seeds, and volunteer tomato plants. Phytopathology 76: 430-434. 71. Jones, J. B., R. E. Stall, and H. Bouzar. 1998. Diversity among xanthomonads pathogenic on pepper and toma to. Annu. Rev. Phytopathol. 36: 41-58. 72. Jones, J. B., R. E. Stall, J. W. Scott, G. C. Somondi, H. Bouzar, and N. C. Hodge. 1995. A third tomato race of Xanthomonas campestris pv. vesicatoria Plant Dis. 79: 395398. 73. Jones, J. B., J. P. Woltz, J. P. Jones, and K. L. Portier. 1991. Population dynamics of Xanthomonas campestris pv. vesicatoria on tomato leaflets treated with copper bacteriocides. Phytopathol. 81: 714-719. 74. Kaniga, K., I. Delor, and G. R. Cornelis. 1991. A wide-host-range suicide vector for improving reverse genetics in gram-neg ative bacteria: inactivation of the blaA gene of Yersinia enterocolitica Gene 109: 137-141. 75. Katzen, F., D. U. Ferreiro, C. G. Oddo, M. V. Ielmini, A. Becker, A. Puhler, and L. Ielpi. 1998. Xanthomonas campestris pv. campestris gum Mutants: Effects on xanthan biosynthesis and plant virulence. J. Bacteriol. 180: 1607-1617. 76. Kearney, B. and B. J. Staskawicz. 1990. Widespread distribution and fitness contribution of Xanthomonas campestris avirulence gene avrBs2 Nature 346: 385-386. 77. Kerr, A. and K. Htay. 1974. Biological control of cr own gall through bacteriocin production. Physiol. Plant pathol. 4: 37-44. 78. Kjemtrup, S., Z. Nimchuk, and J. L. Dangl. 2000. Effector proteins of phytopathogenic bacteria: bifunctional signals in virulence and host recogn ition. Curr. Opin. Microbiol. 3: 73-78. 79. Koller, W. 1998. Chemical approaches to managing plant pathogens. In J. R. Ruberson (ed.), Handbook of integrated pest management. Dekker, New York. 80. Kyeremeh, A. G., T. Kikumoto, D. -. Chua ng, Y. Gunji, Y. Takahara, and Y. Ehara. 2000. Biological control of soft rot of chin ese cabbage using singl e and mixed treatments of bacteriocin-producing avirulent mutants of Erwinia carotovora subsp. carotovora J. Gen. Plant Pathol. 66: 264-268. 81. Langridge, J. 1958. An osmotic mutant of Arabidopsis thaliana Aust. J. Biol. Sci. 11: 457-470. 82. Lavermicocca, P., S. L. Lonigro, F. Va lerio, A. Evidente, and A. Visconti. 2002. Reduction of olive knot dis ease by a bacteriocin from Pseudomonas syringae pv. ciccaronei Appl. Environ. Microbiol. 68: 1403-1407.

PAGE 142

142 83. Lazzaroni, J., J. Dubuisson, and A. Vianney. 2002. The Tol proteins of Escherichia coli and their involvement in the transloc ation of group A co licins. Biochimie 84: 391-397. 84. Leben, C. 1974. Survival of plant pathogenic bacter ia. In Anonymous Ohio Agriculture Reseach and Development Center, Wooster, Ohio. 85. Lee, B., Y. Park, D. Park, H. Kang, J. Kim E. Song, I. Park, U. Yoon, J. Hahn, B. Koo, G. Lee, H. Kim, H. Park, K. Yoon, J. Kim, C. Jung, N. Koh, J. Seo, and S. Go. 2005. The genome sequence of Xanthomonas oryzae pathovar oryzae KACC10331, the bacterial blight pathogen of rice. Nucl. Acids Res. 33: 577-586. 86. Lindemann, J., D. C. Arny, and C. D. Upper. 1984. Use of an apparent infection threshold population of Pseudomonas syringae to predict incidence and severity of brown spot of bean. Phytopathol. 74: 1334-1339. 87. Lindow, S. E., D. K. Willis, and N. J. Panopoulos. 1987. Biological control of bacterial brown spot disease of bean with Tn5-induced avirulence mutants of the pathogen. (abstract). Phytopathol. 77: 1768. 88. Liu, T. 1998. Biological control of tomato bact erial spot with Hrpmutants of Xanthomonas campestris pv. vesicatoria. Masters Thesis. University of Florida, Gainesville, FL. 89. Louws, F. J., M. Wilson, H. L. Campbell, D. A. Cuppels, J. B. Jones, P. B. Shoemaker, F. Sahin, and S. A. Miller. 2001. Field control of bacterial spot and bacterial speck of tomato using a plant activator. Plant Dis. 85: 481-488. 90. Marco, G. M. and R. E. Stall. 1983. Control of bacterial spot of pepper initiated by strains of Xanthomonas campestris pv. vesicatoria which differ in sensitivity to copper. Plant Disease 67: 779-781. 91. Masaki, H. and T. Ogawa. 2002. The modes of action of colic ins E5 and D, and related cytotoxic tRNases. Biochimie 84: 433-438. 92. McAuliffe, O., R. P. Ross, and C. Hill. 2001. Lantibiotics: Struct ure, biosynthesis and mode of action. FEMS Microbiology Reviews 25: 285-308. 93. McGuire, R. G. and J. B. Jones. 1986. Tween media for semiselective isolation of Xanthomonas campestris pv. vesicatoria from soil and plant material. Plant Dis. 70: 887891. 94. McGuire, R. G., J. B. Jones, and J. W. Scott. 1991. Epiphytic populations of Xanthomonas campestris pv. vesicatoria on tomato cultigens resistant and susceptible to bacterial spot. Plant Dis. 75: 606-609. 95. McInnes, T. B. and R. D. Gitaitis. 1988. Airborne dispersal of bacteria in tomato and pepper transplant fields. Plant Dis. 72: 575-579.

PAGE 143

143 96. Mew, T. W., C. I-pin, and J. S. Huang. 1984. Scanning electron microscopy of virulent and avirulent strains of Xanthomonas campestris pv. oryzae on rice leaves. Phytopathology 74: 635-641. 97. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 98. Miller, M. B. and B. L. Bassler. 2001. Quorum sensing in bact eria. Annu. Rev. Microbiol. 55: 165-199. 99. Miller, K. W., R. Schamber, Y. Chen, and B. Ray. 1998. Production of active chimeric pediocin AcH in Escherichia coli in the absence of processing and secretion genes from the pediococcus pap operon. A ppl. Environ. Microbiol. 64: 14-20. 100. Minsavage, G. V., J. B. Jones, and R. E. Stall. 1996. Cloning and sequencing of an avirulence gene ( avrRx3 ) isolated from Xanthomonas campestris pv. vesicatoria tomato race 3. Phytopathology 86: S25. 101. Minsavage, G. V., M. B. Mudgett, R. E. Stall, and J. B. Jones. 2004. Importance of opgHXcv of Xanthomonas campestris pv. vesicatoria in host-parasite interactions. Mol. Plant Microbe Interact. 17: 152-161. 102. Moss, W., J. M. Byrne, and M. Wilson. 1997. Interactions between Xanthomonas axonopodis pv. vesicatoria hrp mutants and the pathogenic parent. Phytopathol. 87: S68. 103. Mudgett, M. B. 2005. New insights to the function of phytopathogenic bacterial type III effectors in plants. Ann. Rev. Plant Biol. 56: 509-531. 104. Noel, L., F. Thieme, J. Gabler, D. Buttner, and U. Bonas. 2003. XopC and XopJ, two novel type III effector Proteins from Xanthomonas campestris pv. vesicatoria J. Bacteriol. 185: 7092-7102. 105. Noel, L., F. Thieme, D. Nennstiel, and U. Bonas. 2001. cDNA-AFLP analysis unravels a genome-wide hrpG-regulon in the plant pathogen Xanthomonas campestris pv. vesicatoria Mol. Microbiol. 41: 1271-1281. 106. Noel, L., F. Thieme, D. Nennstiel, and U. Bonas. 2002. Two novel type III-secreted proteins of Xanthomonas campestris pv. vesicatoria are encoded within the hrp pathogenicity siland. J. Bacteriol. 184: 1340-1348. 107. Novotny, C. P. and P. Fives-Taylor. 1974. Retraction of F pili. J. Bacteriol. 117: 13061311. 108. O'Garro, L. W. and E. Charlemange. 1994. Comparison of bacteria l growth activity of glucanase and chitinase in pepper leaf and flower tissue infected with Xanthomonas campestris pv. vesicatoria Physiol. Mol. Plant Pathol. 45: 181-188.

PAGE 144

144 109. O'Garro, L. W. and S. Tudor. 1994. Contribution of four races of Xanthomonas campestris pv. vesicatoria to bacterial spot in Barbados. Plant Dis. 78: 88-90. 110. Okabe, N. 1954. Studies of Pseudomonas solanacearum Antagonism among the strains of P. solanacearum ., p. 37-40. In Anonymous Reports of the Faculty of Agriculture. Shizuoka University, Japan. 111. Papathanasopoulos, M. A., F. Krier, A. Revol-Junelles, G. Lefebvre, Le Caer, Jean Pierre, A. von Holy, and J. W. Hastings. 1997. Multiple bacteriocin production by Leuconostoc mesenteroides TA33a and other Leuconostoc/Weissella strains. Curr. Microbiol. 35: 331-335. 112. Parke, J. L. and D. Gurian-Sherman. 2001. Diversity of the Burkholderia cepacia complex and implications for assessment of biological control strains. Annu. Rev. Phytopathol. 39: 225-258. 113. Paulitz, T. C. and R. R. Belanger. 2001. Biological control in greenhouse systems. Annu. Rev. Phytopathol. 39: 103-133. 114. Pernezny, K., L. E. Datnoff, T. Mueller, and J. Collins. 1996. Losses in fresh-market tomato production in Florida due to target sp ot and bacterial spot and the benefits of protectant fungicides. Plant Dis. 80: 559-563. 115. Pohornezny, K., R. E. Stall, S. Subramanya, and K. Shuler. 1993. Integrated control of bacterial spot on pepper. Grower and Rancher 86: No. 8: 88. 116. Pohronezny, K., M. A. Moss, W. Dankers, and J. Schenk. 1990. Dispersal and management of Xanthomonas campestris pv. vesicatoria during thinning of direct-seeded tomato. Plant Dis. 74: 800-805. 117. Pohronezny, K., R. E. Stall, B. I. Cant eros, M. Kegley, L. E. Datnoff, and R. Subramanya. 1992. Sudden shift in the prevalent race of Xanthomonas campestris pv. vesicatoria in pepper fields in Southern Florida. Plant Dis. 76: 118-120. 118. Pohornezny, K. and R. B. Volin. 1983. The effect of bacterial spot on yield and quality of fresh market tomatoes. Hortscience 18: 69-70. 119. Porter, W. H., L. W. Cunningham, and W. M. Mitchell. 1971. Studies on the active site of clostripain. The specific inactivation by th e chloromethyl ketone derived from agr-NTOSYL-l-LYSINE. J. Biol. Chem. 246: 7675-7682. 120. Pugsley, A. P. 1984. The ins and outs of colicins. Pa rt I: Production, and translocation across membranes. Micrbiol. Sci. 1: 168-175. 121. Pugsley, A. P. 1984. The ins and ours of colicins. Pa rt II. Lethal action, immunity and ecological implications. Micrbiol. Sci. 1: 203-205.

PAGE 145

145 122. Quezado-Duval, A. M., R. P. Leite, D. Truffi, and L. E. Camargo. 2004. Outbreaks of bacterial spot caused by Xanthomonas gardneri on processing tomato in Central-West Brazil. Plant Dis. 88: 157-161. 123. Ramos, L. J. and R. B. Volin. 1987. Role of stomatal opening and frequency on infection of Lycopersicon spp. by Xanthomonas campestris pv. vesicatoria Phytopathol. 77: 13111317. 124. Reece, K. S. and G. J. Phillips. 1995. New plasmids carrying antibiotic-resistance cassettes. Gene 165: 141-142. 125. Reeves, P. 1972. The bacteriocins. Molecular bi ology, biochemistry, and biophysics. Springer-Verlag, New York, New York. 126. Ribeiro, O. K. 1978. Isolation techniques: Antibiotic media for general isolation of Phytophthora spp., p. 55-57. In Anonymous A source book fo the genus Phytophthora J. Cramer, Germany. 127. Riley, M. 1993. Molecular mechanisms of co licin evolution. Mol Biol Evol 10: 1380-1395. 128. Riley, M. A. and D. M. Gordon. 1999. The ecological role of b acteriocins in bacterial competition. Trends in Microbiology 7: 129-133. 129. Riley, M. A. and J. E. Weltz. 2002. Bacteriocin diversity: ec ological and evolutionary perspectives. Biochimie 84: 355-356. 130. Roden, J. A., B. Belt, J. B. Ross, T. Tachibana, J. Vargas, and M. B. Mudgett. 2004. A genetic screen to isolate t ype III effectors translocat ed into pepper cells during Xanthomonas infection. PNAS 101: 16624-16629. 131. Romantschuk, M., E. Roine, and S. Taira. 2001. Hrp Pilus Reaching through the pant cell wall. Eur. J. Plant Pathol. 107: 153-160. 132. Sahin, F. and S. Miller. 1996. Characterization of Ohio strains of Xanthomonas campestris pv. vesicatoria, causal agent of bact erial spot of pepper. Plant Dis. 80: 774-449. 133. Salanoubat, M., S. Genin, F. Artiguenave, J. Gouzy, S. Mangenot, M. Arlat, A. Billault, P. Brottier, J. C. Camus, L. Cattolico, M. Chandler, N. Choisne, C. ClaudelRenard, S. Cunnac, N. Demange, C. Gaspin, M. Lavie, A. Moisan, C. Robert, W. Saurin, T. Schiex, P. Siguier, P. Theb ault, M. Whalen, P. Wincker, M. Levy, J. Weissenbach, and C. A. Boucher. 2002. Genome sequence of the plant pathogen Ralstonia solanacearum Nature 415: 497-502. 134. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular Clon ing: A laboratory manual. Cold Spring Harbor, Cold Spring Harbor, NY. 135. Scott, J. W. and J. B. Jones. 1989. Inheritance of resi stance for bacterial spot of tomato initiated by Xanthomonas campestris pv. vesicatoria J. Amer. Hort. Soc. 114: 111-114.

PAGE 146

146 136. Shaner, G. and R. Finney. 1997. The effect of nitrogen ferti lization on the expression of slow mildewing resistance in Knox wheat. Phytopathol. 67: 1051-1056. 137. Sharon, E., Y. Bashan, Y. Okon, and Y. Henis. 1982. Presymptomatic multiplication of Xanthomonas campestris pv. vesicatoria on the surfaces of pepper leaves. Can. J. Bot. 60: 1041-1045. 138. Simmonds, R. S., W. J. Simpson, and J. R. Tagg. 1997. Cloning and sequence analysis of zooA a Streptococcus zooepidemicus gene encoding a bacteriocin-like inhibitory substance having a domain structure sim ilar to that of lysostaphin. Gene 189: 255-261. 139. Smidt, M. L. and A. K. Vidaver. 1986. Isolation and characteri zation of syringacin W-1, a bacteriocin produced by Pseudomonas syringae pv. syringae Can. J. Microbol. 32: 231236. 140. Stall, R. E. 1993. Xanthomonas campestris pv. vesicatoria : Cause of bacterial spot of tomato and pepper., p. 57-60. In J. G. Swings and E. L. Civerolo (ed.), Xanthomonas. Chapman and Hill, London. 141. Stall, R. E. 1995. Xanthomonas campestris pv. vesicatoria ., p. 167-181. In U. S. Singh, R. P. Singh and K. Kohmoto (ed.), Pathogenesis and host specificity in plant diseases: Histopathological, biochemical, genetic, and mol ecular bases. Elsevier Science, New York. 142. Stall, R. E. and P. L. Thayer. 1962. Streptomycin resistance of the bacterial spot pathogen and control with streptom ycin. Plant Dis. Reptr. 46: 389-392. 143. Stall, R. E., D. C. Loschke, and J. B. Jones. 1986. Linkage of copper resistance and avirulence loci on a self-transmissible plasmid in Xanthomonas campestris pv. vesicatoria Phytopathology 76: 240-243. 144. Sutic, D. 1957. Bakterioze crvenog patlidzana (Tomat o bacteriosis). (English Summary) Posebna Izd. Inst. Zashi Bilija Beograd (Spec. Edit. Inst. Plant Prot. Beograd). Rev. Appl. Mycol. 36: 734-735. 145. Swings, J., L. Vauterin, and K. Kersters. 1993. The bacterium Xanthomonas, p. 121-156. In J. G. Swings and E. L. Civerolo (ed.), Xanthomonas Chapman and Hill, London. 146. Swords, K., D. Dahlbeck, B. Kearney, M. Roy, and B. Staskawicz. 1996. Spontaneous and induced mutations in a singl e open reading frame alter both virulence and avirulence in Xanthomonas campestris pv. vesicatoria avrBs2 J. Bacteriol. 178: 4661-4669. 147. Tagg, J. R. and A. R. McGiven. 1971. Assay system for bacteriocins. Appl. Microbiol. 21: 943. 148. Tai, T. H., D. Dahlbeck, E. T. Clark, P. Gajiwala, R. Pasion, M. C. Whalen, R. E. Stall, and B. J. Staskawicz. 1999. Expression of the Bs2 pepper gene confers resistance to bacterial spot disease in tomato. PNAS 96: 14153-14158.

PAGE 147

147 149. Tampakaki, A. P., V. E. Fadouloglou, A. D. Gazi, N. J. Panopoulos, and M. Kokkinidis. 2004. Conserved features of type III secretion. Ce ll. Microbiol. 6: 805-816. 150. Thieme, F., R. Koebnik, T. Bekel, C. B erger, J. Boch, D. Buttner, C. Caldana, L. Gaigalat, A. Goesmann, S. Kay, O. Kirchner, C. Lanz, B. Linke, A. C. McHardy, F. Meyer, G. Mittenhuber, D. H. Nies, U. Niesbach-Klosgen, T. Patschkowski, C. Ruckert, O. Rupp, S. Schneiker, S. C. Schuster, F. J. Vorholter, E. Weber, A. Puhler, U. Bonas, D. Bartels, and O. Kaiser. 2005. Insights into genome plasticity and pathogenicity of the plant pathogenic bacterium Xanthomonas campestris pv. vesicatoria revealed by the complete genome sequence. J. Bacteriol. 187: 7254-7266. 151. Timmer, L. W., J. J. Marois, and D. Achor. 1987. Growth and survival of Xanthomonads under conditions nonconducive to disease to disease development. Phytopathology 77: 1345. 152. Trigalet, A. and D. Trigalet-Demery. 1990. Use of avirulent mutants of Pseudomonas solanacearum for the biological control of bacterial wilt of tomato plants. Physio. Mol. Plant Physiol. 36: 27-38. 153. Tudor, S. M. 1995. An analysis of antagonism in tomato race 3 strains of Xanthomonas campestris pv. vesicatoria Masters Thesis. University of Florida, Gainesville, FL. 154. Tudor, S. M. 1999. Molecular characterization of bacter iocin-like activity in tomato racethree strains of Xanthomonas campestris pv. vesicatoria PhD Dissertation. University of Florida, Gainesville, FL. 155. Tudor-Nelson, S. M., G. V. Minsavage, R. E. Stall, and J. B. Jones. 2003. Bacteriocinlike substances from tomato race 3 strains of Xanthomonas campestris pv. vesicatoria Phytopathol. 93: 1421. 156. Vakili, N. G. 1967. Importance of wounds in bacterial spot ( Xanthomonas vesicatoria ) of tomatoes in the field. Phytopathology 57: 1099-1103. 157. Vauterin, L., B. Hoste, K. Kersters, and J. Swings. 1995. Reclassification of Xanthomonas Int. J. Syst. Bacteriol. 45: 472-489. 158. Vidaver, A. K. 1983. Bacteriocins: the lure a nd the reality. Plant Dis. 67: 471-475. 159. Vidaver, A. K. 1976. Prospects for control of phytopat hogenic bacteria by bacteriophages and bacteriocins. Annu. Rev. Phytopathol. 14: 451-465. 160. Volcani, Z. 1969. The effect of mode of irrigation and wind direction on disease severity caused by Xanthomonas vesicatoria in Isreal. Plant Dis. Reptr. 6: 161. Voulhoux, R., A. Lazdunski, and A. Filloux. 2001. Colicin A hybrids: a genetic tool for selection of type II secretion-proficient Pseudomonas strains. EMBO reports 2: 49-54.

PAGE 148

148 162. Walker, J. C. and P. N. Patel. 1964. Splash dispersal and wind as factors in epidemiology of halo blight of bean. Phytopathology 54: 140-147. 163. Wang, L., Y. He, Y. Gao, J. E. Wu, Y. Don g, C. He, S. X. Wang, L. Weng, J. Xu, L. Tay, R. X. Fang, and L. Zhang. 2004. A bacterial cell-cell communication signal with cross-kingdom structural an alogues. Mol. Microbiol. 51: 903-912. 164. Waters, V. L. and J. H. Crosa. 1991. Colicin V virulence plasmids. Microbiol. Rev. 55: 437-450. 165. Wengelnik, K., G. Van den Ackerveken, and U. Bonas. 1996. HrpG, a key hrp regulatory protein of Xanthomonas campestris pv. vesicatoria is homologous to twocomponent response regulators. Mol. Plant Microbe Interact. 9: 704-712. 166. Whalen, M. C., J. F. Wang, F. M. Carland, M. E. Heiskell, D. Dahlbeck, G. V. Minsavage, J. B. Jones, J. W. Scott, R. E. Stall, and B. J. Staskawicz. 1993. Avirulence gene avrRxv from Xanthomonas campestris pv. vesicatoria specifies resistance on tomato line Hawaii 7998. Mol. Plant Microbe Interact. 6: 616-627. 167. Wichmann, G., D. Ritchie, C. S. Kousik, and J. Bergelson. 2005. Reduced genetic variation occurs among genes of the highly clonal plant pathogen Xanthomonas axonopodis pv. vesicatoria including the effector gene av rBs2. Appl. Environ. Microbiol. 71: 24182432. 168. Wilson, M., H. L. Campbell, J. B. Jones, and D. A. Cuppels. 2002. Biological control of bacterial speck of tomato unde r field conditions at several locations in North America. Phytopathol. 92: 1284-1292. 169. Xu, G. W. and C. F. Gonzalez. 1991. Plasmid, genomic, and bact eriocin diversity in U.S. strains of Xanthomonas campesteris pv. oryzae Phytopathol. 81: 628-631. 170. Zhang, L. H., M. J. Fath, H. K. Mahanty, P. C. Tai, and R. Kolter. 1995. Genetic Analysis of the Colicin V Secretion Pathway. Genetics 141: 25-32. 171. Zhu, W., M. M. MaGbanua, and F. F. White. 2000. Identification of two novel hrpassociated genes in the hrp gene cluster of Xanthomonas oryzae pv. oryzae. J. Bacteriol. 182: 1844-1853.

PAGE 149

149 BIOGRAPHICAL SKETCH Aaron Hert was born in Quinc y, Illionis, on May 2, 1976, to D onald and Linda Hert. He attended QuincyUniversity in Illinois. Aaron gr aduated with a B.S. in biological sciences from Quincy University in Illinois. After graduating in 1998, he moved to Florida and worked as a lab technician for Dr. Pamela Roberts at the Sout hwest Florida Research and Education Center (SWFREC) in Immokalee, Florida for one year. In 1999, Aaron received an assistantship from the SWFREC to pursue an M.S. degree in plan t pathology at the Univ ersity of Florida, Gainesville. In December 2001, Aaron obtained a ma sters degree in Plant Pathology with Drs. Pamela D. Roberts and Jeffrey B. Jones. In spring 2002 Aaron received an assistantship from the University of Florida (half from Drs. M. Timur Momol from the North Florida Research and Education Center (NFREC) and Dr. Jeffrey B. Jones with the plant pa thology department at Gainesville) to pursue a Ph.D. degree in plan t pathology at the Univ ersity of Florida.


Permanent Link: http://ufdc.ufl.edu/UFE0019980/00001

Material Information

Title: Evaluation of Bacteriocins in Xanthomonas perforans for Use in Biological Control of Xanthomonas euvesicatoria
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0019980:00001

Permanent Link: http://ufdc.ufl.edu/UFE0019980/00001

Material Information

Title: Evaluation of Bacteriocins in Xanthomonas perforans for Use in Biological Control of Xanthomonas euvesicatoria
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0019980:00001


This item has the following downloads:


Full Text





EVALUATION OF BACTERIOCINS IN Xanthomona~s perforans FOR USE IN BIOLOGICAL
CONTROL OF Xanthomona~s euvesicatoria






















By

AARON PAUL HERT


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

2007









ACKNOWLEDGMENTS

My sincere thank out to Dr. Jeffrey B. Jones, who served as my supervising committee

cochair for my M. S. and Ph.D. degrees. I appreciate all of the time that he spent sharing his

ideas. He was always happy to discuss new ideas and theories. I also thank Dr. Timor Momol,

who served as co-chair for my PhD and had the faith in me to complete my degree. I owe Dr.

Pamela D. Roberts, who served as a co-chair during my masters degree and a committee member

for my PhD. Her support both technically and financially allowed me to complete my degree. I

truly appreciate all her continued support of my work. My thanks also goes to Dr. Steve Olson,

Dr. Jim Preston, and Dr. Martin Handfield for serving on my committee and all the helpful

discussions about ideas and experiments.

I would like to especially acknowledge Dr. Mizuri Marutani whom has been instrumental

in much of the molecular aspects of my research and has played an integral part in bacteriocin

analysis and attenuation mutants creation. Her input and knowledge was utilized on a daily

basis, I don't know if I could have done it with out her.

I would like to thank Jerry Minsavage for his technical support and donation of several

mutants to aid in my masters and Ph.D. degrees. The faculty and staff of the Department of

Plant Pathology were very supportive throughout my program, providing assistance whenever it

was needed. I also would like to thank my fellow graduate students, who provided continual

support and friendship.

Last but not least, I thank my family. My mother, father, brother, sister, and my niece

and nephew have been in my heart and mind with me every step of the way. I'd like to thank all

of my family for being there for me since day one, with all the love and belief in me that I could

ask for.












TABLE OF CONTENTS


page



ACKNOWLEDGMENT S .............. ...............2.....


LI ST OF T ABLE S .........._.... ...............6.._.._ ......


LI ST OF FIGURE S .............. ...............7.....


AB S TRAC T ......_ ................. ............_........9


CHAPTER


1 INTRODUCTION ................. ...............11.......... ......


History, Etiology, and Strain Diversity ................. ...............11........... ...
Epidemiology .............. ...............13....
Disease Control .............. ...............14....
Bacteriocins ................. ....... ... ._. ..... ...............15...
The Type III Secretion System (T3 SS) ................. ........._..._. ...._.__......._ .. 16
Biological Control .............. ...............17....

2 CHARACTERIZATION OF GENETIC DETERMINANTS AND EVALUATION OF
THEIR ROLE IN ANTAGONISM ................. ...............19.......... .....


M materials and M ethods .............. .. ........... .... ............2
Bacterial Strains, Plasmids and Culture Conditions ................. ................ ........ .22
DNA Manipulations ................... ...............22.
Construction of Bacteriocin Mutants............... ...............23
Construction of T2SS Mutant................ ...... ...............2
Bioinformatics Characterization of Bcn+ Cosmid Clones .............. .....................2
Subcloning of BcnB and BcnC............... ...............24..
Protease Activity Assay................. .... ...... .........2
BcnA Timing of Activation and Size Analysis ................ ............ ........ .........25
In vitro Antagonistic As say .............. ...............25....
Evaluation of Immunity ................. ...............27....___ ......
In vitro a;ss ay~ssssssssss ................ ...............27....... ......
hr planta a~ssays. .............. ...............27....
Re sults............... .. ........ ..... ..... ... ... .... ..... .......2

Sequence Analysis of Genes Involved in BcnA Activity ........._._ ..... ..._._...........28
BcnA Activity Requires ORFA, ORF2, ORF3 and ORF4..........._...._ ............ ........29
Localization of BcnA Activation............... ...............2
Identification of the Immunity Gene ................. ...............29........... ...
Sequence Analysis of BcnB and BcnC ...................... ...............30
Purification and Characterization of BcnB and BcnC ................. ............... ....__.3 1

Type II Secretion Mutant Lost Secretion of Amylase and Bacteriocins..............._..__......32












Discussion ................. ...............32.................


3 ANALYSIS OF PATHOGENICITY MUTANTS OF XANTH2OMONASPERFORANS
AND THEIR EFFECT ON BACTERIOCIN EXPRESS SION. ......____ ...... ....__..........5 1


M materials and M ethods ........... ........... ... ..... ...............5
Bacterial Strains, Plasmids and Culture Conditions ......____ ..... ... ._ ...............54
Plant M material .............. ...............54....
Primer Design ............ ..... ._ ...............55....
Generation of Mutants ............ ..... ._ ...............55...
Growth Room Growth Curve Assays............... ...............58.
Greenhouse Disease Severity Assay .............. ...............59....
Growth Room Antagonism Assay ............ ......__ .......___ ........ 5
R e sults.............. ......__ _. ....... ........... ..... ..... ..........6

Sequence Analysis of Attenuated Mutant Candidate Genes ..........._... ......._........61
Population Dynamics and Pathogenicity Assays .............. ...............62....
Antagonism As say s .............. ...............63....
Discussion ................. ...............63.................


4 EVALUATION OF XANTH2OMONASPERFORANS MUTANTS IN CONTROLLING
X. EUVESICA TORIA IN GREENHOUSE AND IN THE FIELD ................ ................ ...80


M materials and M ethods ........... ........... ... ..... ...............8
Bacterial Strains, Plasmids and Culture Conditions ................. ................. ........ 83
Generation of the 91-1 18::AopgH~bcnB Attenuation Mutant.........._.._.. ........._.._.. ..84
Plant M materials ........._...... ........_. ...............84...
Growth Room Growth Curve Assays............... ...............84.
Greenhouse Disease Severity Assay .............. ...............85....
Growth Room Antagonism Assay ........._.. ...._._..... ...............86...
Internal antagonism assays. .............. ...............86....
Phyllosphere antagonism assays. .............. ...............86....
Field Experiments ................. ...............87.................
Field plot design. .............. ........ .. ..... ..............8
Bacterial strains, inoculum production, inoculation and plant material. ................87
Incidence ofstrains in lesions. ............. ...............88.....
Incidence ofphyllosphere populations ................. ......_.. ........._.__.......89
R e sults................ ........... __ ..... ..... .. .... ........... ... ... ..........8
91-118 and Mutants Reduce E3-1 in Growth Room and in the Greenhouse ..................89
Field Study............... ...............90.
Discussion ................. ...............91._ ___.......


5 SUMMARY AND DISCUSSION .............. ...............106....


APPENDIX


SEQUENCE AND ALIGNMENT OF BCNB AND PATHO GENICITY-RELATED
GENES CHOSEN FOR DELETION ................. ...............110...............












LIST OF REFERENCES ................. ...............135................


BIOGRAPHICAL SKETCH ................. ...............149......... ......










LIST OF TABLES


Table page

2-1. Bacterial strains and plasmids used in this study .............. ...............35....

2-2. PCR primer used in bcnA, bcnB and bcnC analyses for genetic manipulation ....................36

2-3. Characterization of bacteriocin ORFs associated with the expression of BcnA, BcnB
and BcnC ........... _... ...._. ...............37...

3-1. Bacterial strains and plasmids used in this study .............. ...............68....

3-2. PCR primers used in hpaA, hpaB, hpaC, xopA, xopD, avrBs2 and gumD analyses for
genetic manipulations .............. ...............69....

3-3. List of attenuation candidate genes .............. ...............70....

3-4. Homology of X perforans genes............... ...............71.

3-5. In planta growth and aggressiveness of X perforans strain 91-118 mutants................_._....72

3 -6. Growth room in planta internal and phyllosphere antagoni sm experiments ........................73

4-1. Bacterial strains and plasmids used in this study .............. ...............95....

4-2. In planta growth and aggressiveness of X perforans strain 91-118 mutants as .................96

4-3. Growth room in planta internal and phyllosphere antagoni sm experiments ........................97

4-4. Incidence and recovery ofX. euvesicatoria strain E3-1 in the field............... ..................9











LIST OF FIGURES


Figure page

2-1. BcnA diagram showing position of individual ORFs .............. ...............38....

2-2. BcnB diagram showing position of individual ORFs............... ...............39..

2-3. BcnC diagram showing position of individual ORFs............... ...............40..

2-4. Antagonism assays of bacteriocin-like activity against I euvesicatoria .............. ..............41

2-5. Antagonism activation assays of bacteriocin-like activity .............. ...............42....

2-6. Secretion assays of bacteriocin-like activity .............. ...............43....

2-7. Immunity assays of bacteriocin-like activity ................. ...............44...............

2-8. In vitro populations of I. euvesicatoria strain 9 1 -106 transconjugants ........._..... ..............45

2-9. In planta populations of I euvesicatoria strain 91-106 transconjugants .............................46

2-10. Antagonism assays of bacteriocin-like activity ................. ....._._ ........... .......4

2-11. Protease activity of bacteriocin candidates............... ...............4

2-12. Evaluation of type two secretion system mutant xpsD ................. ................ ........ .49

2-13. BcnA Model for ORFs predicted involved in BcnA activity .............. .....................5

3-1. Illustration of deletion constructions .............. ...............74....

3-2. In planta growth of wild-type (wt) and mutant X perforans strains. ..........._..._ ...............75

3-3. Disease severity on Bonny Best leaflets 2 weeks after dip inoculation .............. ................76

3 -4. Phenotype in leaves of Bs2 transgenic tomato VS3 6 and pepper (ECW-20R). .........._.......77

3-5. Growth room internal antagonism assay .............. ...............78....

3-6. Growth room phyllosphere antagonism assay .....__.....___ ..........._ ..........7

4-1. In planta growth of wild-type and mutant X perforans strains ........._. ..... ...._._.........99

4-2. Disease severity on Bonny Best leaflets 2 weeks after dip inoculation .............. ................100

4-3. Growth room internal antagonism assay .............. ...............101....

4-4. Growth room phyllosphere antagonism assay ....__ ......_____ .......___ ............0











4-5. Quincy 2004 Hield experiment ........._._ ...._.._.. ...............103.

4-6. Quincy 2005 Hield experiment ........._..... ...............104._._.......

4-7. Citra 2005 Hield experiment............... ..............10

A-1. BcnB sequence. Nucleotide sequence of a 5968 bp KpnI and EcoRI fragment
containing bcnB. ................ ...............110......... ......

A-2. Nucleotide sequence of a 980 bp fragment containing the hpaA ORF ............... ... ............1 14

A-3. Nucleotide sequence of a 2345 bp fragment containing the hpaB ORF ................... ..........1 15

A-4. Nucleotide sequence of a 1419 bp fragment containing the hpaC ORF ................... .........1 17

A-5. Nucleotide sequence of a 1716 bp fragment containing the xopA ORF ................... ..........1 19

A-6. Nucleotide sequence of a 2173 bp fragment containing the xopD ORF ............................ 121

A-7. Nucleotide sequence of a 2735 bp fragment containing the avrBs2 ORF ................... ........124

A-8. Similarity of amino acid sequence of HpaA............... ...............127.

A-9. Similarity of amino acid sequence of HpaB ................ ...............128.............

A-10. Similarity of amino acid sequence of HpaC ................ ....___ ...............129

A-11i. Similarity of amino acid sequence of AvrB s2 ................. ...............130...........

A-12. Similarity of amino acid sequence of XopA ................. ...............132........... .

A-13. Similarity of amino acid sequence of XopD ................. ...............133...........









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

EVALUATION OF BACTERIOCINS IN Xanthomona~s perforans FOR USE IN BIOLOGICAL
CONTROL OF Xanthomona~s euvesicatoria

By

Aaron Paul Hert

May 2007

Chair: Jeffrey B. Jones
Cochair: M. Timur Momol
Major: Plant Pathology

Xanthomonas perforans~X~XX~XX~X strain 91-118 produces at least three different bacteriocin-like

compounds (BcnA, BcnB, BcnC) antagonistic toward I. euvesicatoria strains. Previous research

showed that deletion of one bacteriocin (BcnB) produced the highest level of antagonism toward

sensitive X. euvesicatoria strains. One aspect for this study was to further characterize each

bacteriocin by deletion mutagenesis to establish which open reading frames (ORFs) were

responsible for bacteriocin activity for each bacteriocin as well as determining their possible

functions. BcnA has been shown to contain at least four essential genes for activity and a model

has been created to suggest the role of each gene. BcnB and BcnC were both found to be

proteinases (endoproteinase Arg-C and extracellular metalloproteinase, respectively).

A second aspect of this study was to develop a viable biocontrol strategy by creating

pathogenicity-attenuated mutants such that these attenuated mutants on the plant surface would

suppress bacteriocin-sensitive strains. Several candidate genes were chosen based on mutant

phenotypes in either X. perforans (OpgHxcy) or the closely related I. euvesicatoria strain 85-10

(hpaA, hpaB, hpaC, xopA, xopD, avrBs2 and gumD). Each candidate gene was amplified and

PCR-assisted deletion mutagenesis was performed for final marker exchange into wild-type (wt)










X. perforans to create attenuation mutants. Each mutant was tested for growth rate, disease

severity and antagonism toward I. euvesicatoria-sensitive strains. Mutations in opgH and gumD

gave the most significant reduction in disease and growth rate while maintaining the ability to

reduce X. euvesicatoria populations.

One attenuated mutant, 91-118::AopgH, was chosen for further investigation.

Greenhouse and Hield experiments were conducted using 91-118::AopgH~bcnB to determine its

ability to reduce X. euvesicatoria populations. Greenhouse and Hield experiments indicate

91-1 18::AopgH~bcnB significantly reduced I. euvesicatoria populations. In the Hield, weekly

application of 91-1 18::AopgH~bcnB consistently reduced I. euvesicatoria populations as

compared to the standard control (application of copper + manzate and actigard every two

weeks). 91-118::AopgH~bcnB applied every two weeks also significantly reduced I.

euvesicatoria populations in one season, but were not significantly different from the grower

standard control.









CHAPTER 1
INTTRODUCTION

Bacterial spot of tomatoes and peppers is caused by the phytopathogenic bacterium

XanthomonasXX~~XX~~~XX~~XX euvesicatoria, X. vesicatoria and X. perforans (69). It is a worldwide disease and

is a maj or problem in Florida, particularly during periods of high temperatures and high

humidity. Pohronezny et al. (1 15) estimated as high as 50% loss of marketable fruit due to

bacterial spot on tomatoes.

The pathogen is a Gram-negative, rod-shaped bacterium which can readily be isolated

from diseased tissue. It is motile, possessing a single polar flagellum, strictly aerobic, and

measures 0.7 to 1.0 Clm by 2.0-2.4 Clm. On nutrient agar the bacterium produces a characteristic

yellow water-insoluble pigment called xanthomonadin and an extracellular polysaccharide (EPS)

termed xanthan gum (149).

Bacterial spot of tomato affects the aerial portions of the plant, with symptoms consisting

of numerous small (1 to 5 mm) circular lesions on leaves, stems and fruit. Bacterial spot can be

distinguished from fungal leaf spots by a greasy, water-soaked appearance on the abaxial side of

leaves. Chlorosis and epinasty of leaves occurs, eventually leading to complete necrosis of

tomato leaflets (140).

History, Etiology, and Strain Diversity

Bacterial spot of tomato and pepper is one of the earliest recorded bacterial diseases. I.

euvesicatoria was first described as bacterial canker in South Africa by Doidge in 1921 (33).

That same year Gardner and Kendrick (52) in the United States discovered a similar organism

and referred to it as bacterial spot. Doidge performed a comprehensive study on the etiology of

bacterial spot, which she termed tomato canker, and identified the causal agent as Bacterium

vesicatorium, whereas in the United States, Gardner and Kendrick (52) originally named the










organism Bacterium exitiosa. Over time, as a result of comprehensive studies (51), the two

bacteria were discovered to be very similar and, in the mid 1920s, were designated Bacterium

vesicatoria. Since that time the bacterium was renamed several times, changing genera from

Bacterium to Pseudomonas to Phytomonas and finally to Xanthomonas (71).

Once transferred to Xanthomonas, the bacterium underwent several species and pathovar

changes. In 1980 it was transferred to X campestris py. vesicatoria (X~c. vesicatoria) (35). In

the 1990s, maj or changes occurred in the taxonomy of the genus Xanthomonas which resulted in

the renaming of many species. Extensive comparison of strains using DNA-DNA hybridizations

resulted in the identification of two groups, A and B (140, 157). Group A (tomato race 1) strains

were transferred to I. axonopod'is and designated I. axonopod'is py. vesicatoria, while Group B

strains (tomato race 2) were placed in X. vesicatoria (157). In the 1990s a new group of strains

was identified in Florida (72) that was phylogenetically most closely related to group A, but was

phenotypically and genotypically distinct enough from group A that it was designated as group

C. Because group C is most closely related to group A based on DNA-DNA hybridization this

new group was designated within I. axonopodis py. vesicatoria. In 2004, the most recent

changes occurred in nomenclature (69). Group A (tomato race 1) and Group C (tomato race 3)

strains were removed from X. axonopod'is, since they shared less than 70% DNA relatedness to

other I. axonopod'is strains, and were placed in I. euvesicatoria and X. perforans, respectively.

Several avirulence genes have been characterized in xanthomonads associated with tomato.

In 1993, Whalen et al. (166) found I. euvesicatoria, tomato race 1 (T1), strains to carry the

avirulence gene avrRxy, which induces an incompatible reaction that activates localized cell

death also known as a hypersensitive response (HR) on the genotype H17998 carrying the

corresponding resistance gene Rxy; X. perforans, tomato race 3 (T3), strains were determined to









carry avrXv3, which induces an HR in H7981 that contains the resistance gene Xv3(4, 100). In

2000, a new avirulence gene avrXv4 was described in X. perforans strains based on reactions on

the tomato genotype, LA716 (Lycopersicum pinnellii) which carries the Xv4 resistance gene (3,

4). Therefore, X. perforans strains carrying this new avirulence gene (avrXv4) but lacking a

functional avrXv3, have been designated as tomato race 4.

Epidemiology

Leben (84) introduced the concept of a resident phase where the pathogen is able to

replicate in the phyllosphere (leaf surface) without causing visible symptoms. Xanthomona~s

euvesicatoria has been shown to colonize pepper and tomato leaves epiphytically (93, 137).

Long-term survival occurs in crop residue and volunteer plants (70, 140). Seed contamination

was proposed as an important mechanism for transmission by Bashan et al. (6); however, Jones

et al. (70) concluded that survival on seed occurs at extremely low levels and may be less

important in the epidemiology of the disease than other inoculum sources when the pathogen is

endemic. In soils artificially infested with Xc. vesicatoria, survival is poor with the bacterium

being detected for only 16 days (6).

The bacterium gains entry into the plant when: (I) conditions are favorable for disease

development (57), or (II) a threshold of epiphytic populations is reached (94) or (III) if a plant is

compromised by wounding (156). Xanthomona~s euvesicatoria enters the plant in many ways.

Routine farming operations damage the plant causing wounds that act as entry points (6, 116).

Epidermal abrasions, leaf hair breakage and water congestion of the intercellular spaces increase

entry of the bacterium up to 100-fold over healthy plants (156). The bacterium can also enter

through natural openings such as stomates and hydathodes (96, 123, 145). High humidity is

conducive to bacterial ingress and survival; high relative humidity has been shown to increase









infection of X euvesicatoria by 10- to 100-fold on tomato leaves compared to low humidity

(151).

Dissemination of the bacterium is a maj or factor in the epidemiology of the disease.

Overhead irrigation enhances dissemination compared to furrow irrigation (160). Dissemination

also occurs in the form of aerosols and wind-blown rain (95, 162). Infected seed serves as a

mechanism for dissemination (116). Farming practices, such as thinning, tying, and mechanized

spraying also serve as factors in dissemination (116).

Disease Control

Bacterial spot of tomato is difficult to control when high temperatures and high moisture

exist. Bactericides, such as fixed coppers and streptomycin, have provided the major means of

control (90, 142). Streptomycin-resi stant mutants were rapidly selected on streptomycin-treated

plants (142).

As a result of rapid selection for streptomycin-resi stant mutants, copper compounds have

been used almost exclusively. However, Marco and Stall in the 1980s (90) showed that many I.

euvesicatoria strains were tolerant to copper and determined that copper resistance is mediated

by genes located on a self-transmissible plasmid (143). Adding mancozeb, a fungicide, to

copper sprays was shown to improve control efficiency (25) and was shown by Marco and Stall

(90) to control copper-tolerant strains. However, they also showed that this treatment is

insufficient when conditions favorable for disease development exist.

Because of the presence of copper-tolerant strains, other control strategies need to be

considered. Identification of resistance genes and introgression into commercial genotypes has

been a focus of breeding programs (135). There are currently no commercially available tomato

varieties resistant to all races of bacterial spot.









Bacteriocins

Bacteriocins are proteins or peptides with antibacterial properties, in most instances

targeting related bacteria belonging to the same species or genus (135). Bacteriocins of Gram-

positive bacteria, such as lactobacilli, typically are small peptides (40, 92). Bacteriocins from

Gram-negative bacteria are often larger proteins among which the colicins from Escherichia coli

represent the best-known examples (14, 66, 83, 91). There is considerable structural diversity

among them which is reflected in widely different modes of action, including membrane

disruption, non-specifie degradation of nucleic acids and inhibition of peptidoglycan synthesis,

and proteases.

Reports on the production of bacteriocin-like compounds by phytopathogenic bacteria are

scarce. The observed report of a phytopathogenic bacterium producing bacteriocin-like

compounds was reported by Okabe in the early 1950s (110). He reported that strains of

Pseudonzona~s (Ralstonia) solan2acearunt were inhibitory exclusively to other P. solan2acearunt

strains. Since then, these types of compounds have been reported for several other genera:

Agrobacteriunt (77), Clavibacter (36), Erwinia (22, 37, 80) and Pseudonzona~s (30, 48, 82, 139,

152). There have been reports of xanthomonads producing bacteriocins as well (45, 155, 169).

In 1991, X. perforans were first identified in Florida. In fields where both I. euvesicatoria

and X. perforans were present, the X. perforans strains became predominant (72). hz vitro assays

have shown that X. perforans strains inhibit growth of I euvesicatoria strains (39). Jones et al.

(71) characterized this relationship in greenhouse experiments tested on three genotypes

including a T3 resistant genotype. Under Hield conditions X. perforans strains had a competitive

advantage over I. euvesicatoria strains (71, 1 17). Tudor (153) identified at least three

antagonistic compounds in X. perforans strains that closely resembled bacteriocins. These

compounds were determined to have narrow inhibition spectra (restricted to Xanthonzona~s









strains) and fit the definition of a bacteriocin described by Reeves (125) based on the following

criteria: (I) the presence of a biologically active protein moiety, (II) inducibility with mitomycin

C and (III) non-self inhibition. All three bacteriocin-like groups (BcnA, BcnB and BcnC) were

unique in activity and specificity against I. euvesicatoria strains (155).

The Type III Secretion System (T3SS)

Pathogenicity of bacterial spot of tomato is determined by the type III secretion system

(T3 SS), which is highly conserved in most Gram-negative bacterial pathogens of plants and

animals (149). The T3SS is composed of a secretion apparatus and an array of diverse proteins,

known as effectors, that are injected into plant cells via the secretion apparatus (103).

The T3SS is encoded by a gene cluster termed the hrp (hypersensitive response and

pathogenicity) cluster (1). One group of genes located within the hrp cluster encodes for a

secretion apparatus known as the hrp pilus. The hrp pilus serves as a secretion apparatus for the

translocation of T3 SS effector proteins (59, 13 1). Hrp pilus mutants no longer cause disease in

susceptible plants and are unable to induce resistance in resistant host and non-host plants (12,

16). Inside the host cell, type III effectors have specific functions and interact with specific

targets in the host (17, 58, 63, 106); however, the functions of many effectors are unknown and

their deletion produces no detectible phenotype.

Some plants have developed resistance to these invading pathogenic bacteria via resistance

genes (R genes) which recognize specific effector proteins called avirulence (avr) genes. When

the pathogenic bacterium inj ects an avirulence gene into a resistant plant carrying the

corresponding R gene, an incompatible reaction, or HR, occurs which localizes the invading

bacteria and limits secondary infection of surrounding cells (103).









Biological Control

Biological control is another important approach for control of the disease. Pathogen

resistance to fungicides has prompted interest in development of biocontrol agents (19), to

provide additional tools for disease management. Unlike biocontrol of insects, biocontrol of

plant diseases is a relatively new field. In the last 25 years biocontrol has become an established

sub-discipline in plant pathology (113). Agrobacterium radiobacter strain K84, registered with

the United States Environmental Protection Agency (EPA) for control of crown gall in 1979

(EPA registration number 11,4201), and was the first commercially available biological control

agent against a bacterial plant disease (77). Since then, a total of 14 bacteria and 12 fungi have

been registered with EPA for control of plant diseases (47, 1 12).

Several promising biological control approaches that include antagonistic microorganisms,

natural fungicides and induced resistance are available for use in disease control today (3 8).

However, achieving success using biocontrol agents for many bacterial diseases has been

difficult. Some success has been achieved in this area through empirical selection of biocontrol

agents, as indicated by the commercialization of the products AgriphageTM (a mixture of

bacteriophages for control of bacterial spot of tomato (46)), GalltrolTM, for control of crown gall,

and BlightBanTM A506, for control of fire blight and frost injury (86). For bacterial spot of

tomato, fi eld experiments have b een conducted utilizing a non-pathogeni c b acteri ocin-producing

X. perforans strain to control disease incited by I. euvesicatoria strains (88). The non-

pathogenic strain was able to reduce X. euvesicatoria-incited disease incidence and severity

when applied prophylactically; however, the disease was still above acceptable levels (forty

percent) (88).

For Ralstonia solan2acearum, efforts to obtain a biological control strategy utilizing

bacteriocin-producing non-pathogenic hrp- mutants gave low to moderate levels of control of









wildtype (wt) R. solan2acearum (152). However, control using a partially pathogenic hrp mutant

(hrc V), which is capable of higher levels of colonization of the root and stem tissue, achieved

better control levels (49). Research into colonization has been conducted to understand the

possible relationship between invasion efficiency of the biocontrol agent and its ability for

disease control. Etchebar et al. (41) suggested that there was a positive correlation between

colonization of the xylem by the hrp mutant and the level of control of the wt R. solan2acearum.

The goal of this study was to evaluate a new biological control strategy utilizing

pathogenicity-attenuated, bacteriocin-producing X. perforans strains for control of bacteriocin-

sensitive strains of I euvesicatoria. The objectives of this study were: (I) to further characterize

the bacteriocins associated with X. perforans, (II) to identify and individually delete genes that

contribute to pathogenicity to create less virulent mutants of X perforans, and (III) to determine

the ability of these pathogenicity-attenuated mutant strains to antagonize X. euvesicatoria in

vitro, in planta and under field conditions.









CHAPTER 2
CHARACTERIZATION OF GENETIC DETERMINANTS AND EVALUATION OF THEIR
ROLE IN ANTAGONISM

Bacterial spot of tomato is incited by four Xanthomona~s species: I. euvesicatoria, X.

vesicatoria, X. perforans and X. gardneri. The first three bacterial species were previously

known as tomato races 1 (T1), 2 (T2) and 3 (T3), respectively, based on their reaction on three

tomato genotypes: Hawaii 7998 (H7998), Hawaii 7981 (H7981) and Bonny Best (71, 72, 139).

I. gardneri has only been found in former Yugoslavia, Costa Rica and Brazil (10, 123, 143).

In 1991, X. perforans was first identified in Florida (72). In fields where both I.

euvesicatoria and X. perforans were present, X. perforans became predominant (72). This

phenomenon was due to bacteriocin-like activity of X perforans strains (152). Bacteriocins are

substances, usually proteinaceous, that are inhibitory or harmful toward closely related bacteria

(124). Bacteriocins of Gram-negative bacteria represent a diverse group of proteins in terms of

size, microbial target, mode of action and immunity mechanism. They are high molecular

weight proteins that gain entry into susceptible cells by binding to surface receptors. Their mode

of action varies from degradation of cellular DNA, to disruption of cleavage of 16S RNA,

inhibition of synthesis of the peptidoglycan and pore formation in the cytoplasmic membrane

(26).

The most extensively studied bacteriocins are the colicins produced by Escherichia coli

(14, 66, 83, 91, 120, 121, 126, 163). A model system has been developed for proteinaceous

bacteriocin production consisting of three components: the toxin, the immunity gene and a

mechanism for delivery (126). Several known bacteriocins are transcribed in an inactive form

(pre-bacteriocin), which, upon secretion, is processed to its active form (eg. Colicin V; (169))

(1 11).









Reports on the production of bacteriocin-like compounds by phytopathogenic bacteria are

limited. In the 1950s Okabe et al (110) published the first article on phytopathogenic bacteria,

where strains of Pseudonona~s (Ralstonia) solan2acearunt were inhibitory only to other P.

solan2acearunt strains. Since the original description, production of such compounds has been

reported for several other genera: Agrobacteriunt (77), Clavibacter (36), Erwinia (22, 38, 80),

Pseudonzona~s (30, 82, 138, 151) and Ralstonia (41, 48). There have been a few reports of

xanthomonads producing bacteriocins (45, 154, 168).

In order to further characterize the bacteriocin-like activity of X perforans, a genomic

library was screened to localize bacteriocin activity (154). Three groups of clones were

identified that showed unique bacteriocin activity and all three bacteriocin-like groups (BcnA,

BcnB and BcnC) were unique in activity and specifieity based on I. euvesicatoria-sensitive

strains (154). None of the clones conferred immunity to the other bacteriocins.

BcnA was localized to an 8.0-kb fragment containing seven open reading frames (ORF)

identified in the sequenced region. The largest ORF (ORFA), approximately 3.6-kb, is required

for BcnA+ activity. The ORFA protein contains 1012 amino acids with a theoretical molecular

weight of approximately 1 11-kDa. BcnA' activity was detected in unconcentrated, cell-free

extracts of strains expressing ORFA. In some bacteria, an immunity function is necessary in

order to avoid self-inhibition of the producing strain (1 11, 169). The putative immunity function

of BcnA was mapped to a 4.5-kb Ba~nHI/EcoRI fragment downstream of ORFA (154).

Southern hybridization analysis using an ORFA-specific probe indicated that among

bacterial spot strains tested, only X. perforans strains hybridized. Hybridization of the probe to a

chromosomal location suggests that BcnA' is in the chromosomal DNA. Homology searches

using the deduced amino acid sequence of the ORF revealed significant homology to only two









known proteins, WapA and Rhs. Both of these proteins contain multiple copies of an almost

identical ligand-binding motif, thought to be involved in carbohydrate binding. Seven copies of

a similar motif were found in ORFA.

Xanthomona~s campestris py. glycines (X c. glycines) is one of the few xanthomonads that

produces multiple bacteriocin-like compounds with activity against selected xanthomonads (45).

According to Tudor (153) at least one of the X c. glycines bacteriocin-like compounds is very

similar in activity to BcnA. In X c. glycines, bacteriocin-like compounds were heat sensitive and

trypsin resistant (45), suggestive of the involvement of a high molecular weight protein.

Relatively little is known about BcnB and BcnC. BcnB and BcnC were previously

subcloned to 8.9-kb and 5.1-kb fragments, respectively (154). Both were sequenced (60, 154).

No immunity factor was associated with BcnB or BcnC activity. It is unknown how a

heterologous strain that expresses either bacteriocin is not inhibited. Although the exact ORF

involved in BcnC expression was not identified, one ORF within this fragment showed

significant homology to extracellular metalloproteases secreted by Aeromona~s hydrophila,

Armillarial~~~~~11111~~~~ mellea, Pleurotus ostreatus, Grifola frondosa, Aspergillus fumiga~tus and Penicillum

citrinum (153). Enzymes produced by bacteria may mimic the action of bacteriocins. The zooA

gene of Streptococcus zooepidemicus, which encodes a bacteriocin-like inhibitory substance,

contains a region with significant homology to several known endopeptidases (137).

It was reported that some proteinases and bacteriocins are secreted via the type two

secretion pathway (T2SS) (2, 149, 160). In order to determine the involvement of the T2SS in X.

perforans bacteriocin production, disruption mutants were created in the closely related bacteria,

I. euvesicatoria, then transformed with a cosmid expressing either BcnA, BcnB, or BcnC. The

T2SS is composed of 12 proteins (Xps) for translocating extracellular proteins across the outer









membrane in Xanthomona~s campestris py. campestris (X c. campestris) (23). I c. campestris

secretes several hydrolytic enzymes, including co-amylase, protease, pectate lyase and cellulase

by the type 2 secretion pathway (23). The XpsD T2SS protein, an outer membrane lipoprotein,

is required for protein secretion via the T2SS (23). An XpsD mutant from Xanthomona~s oryzae

py. oryzae also lost T2SS function (50). Thus, xpsD was chosen for disruption to create a T2SS

mutant. In this study we further characterized three bacteriocins of X. perforans to determine

their activity, delivery mechanisms and immunity.

The goals of this study were (I) to further characterize the role of each bacteriocin-like

compound secreted by X. perforans; (II) to determine the delivery mechanism of each

bacteriocin and; (III) to understand the possible functions of each bacteriocin.

Materials and Methods

Bacterial Strains, Plasmids and Culture Conditions

Strains of I euvesicatoria and X. perforans (Table 2-1) were grown on nutrient agar

(NA) medium (Difco Laboratories, Detroit, MI) at 280C. Strains ofEscherichia coli (Table 2-1)

were grown on Luria-Bertani (LB) medium at 370C (97). All strains were stored in 20%

glycerol in sterile tap water at -800C. Antibiotics were used to maintain selection for resistance

markers at the following concentrations: tetracycline (Tc) 12.5 Clg/mL; rifampicin (Rif) 100

Clg/mL; spectinomycin (Sp) 50 Clg/mL; kanamycin (Km) 25 Clg/mL; chloramphenicol (Cm) 34

Clg/mL; streptomycin (Sm) 200 Clg/mL and nalidixic acid (Nal) 50 Clg/mL.

DNA Manipulations

Standard techniques for molecular cloning were conducted as described by Sambrook et

al. (133). Restriction endonuclease digestions were performed according to manufacturer's

specifications. All enzymes were obtained from Promega (Madison, WI) or Biolab (Ipswich,










MA). All DNA extractions were done as described by Sambrook et al. (135). T4 DNA ligase

(M180A) was used according to manufacturer's specifications (Promega). Constructs were

transformed into competent Escherichia coli DH500 cells prepared as described in Sambrook et

al. (135) and stored at -800C until transformations.

Construction of Bacteriocin Mutants

According to an ORF search, there are 5 genes in the BcnA fragment, designated ORFA,

ORF2, ORF3, ORF4, and ORF5. To characterize the function of BcnA, genes ORFA, ORF2,

and ORF3 were disrupted either by deletion or transposon mutagenesis to create

91-118::AORFA, 91-118::AORF2, and 91-118::AORF3. AORF4 was previously disrupted (153).

The 91-118::AORFA mutant was constructed by deleting an EcoRV and BglII fragment. AORF2

and AORF3 mutants were created by using surrounding sequences up and downstream of the

target ORFA. For ORFA, PCR was performed with primers A5 and A3 (Table 2-2), then the

resulting PCR product was inserted upstream of ORF3 subcloned in the phagemid vector

pBluescript II KS (pBS) (Stratagene, La Jolla, CA) (Figure 2-la). Final AORF2 was moved into

suicide vector pOK1 using restriction enzymes Apal and Spel (Figure 2-1b). To make

91-118::AORF3, PCR was performed with primers ORF2F and ORF3R, then the resulting PCR

product was inserted upstream of ORF4 in pBS:ORF4 to create pBS:AORF3. The fragment

containing AORF3 was subcloned into suicide vector pOK1 with Apal and Sall creating

pOKl:AORF3. The final plasmid constructs were mated into 91-118 to make each mutant via

suicide vector-assisted mutagenesis as described previously (74). Candidates were screened

using PCR primers designed to amplify flanking regions of the cross-over region (Table 2-2).

Each mutant was tested for bacteriocin activity against I. euvesicatoria strain 91-106.









BcnB was disrupted by adding an insertion stop codon (TAA) using Quick-change XL Site

Directed Mutagenesis kit (Stratagene).

Construction of T2SS Mutant

Mutations were created to determine if BcnA, B and C are secreted by T2SS. In order to

clone xpsD gene, primers xpsDF and xpsDR were designed using 85-10 genome sequence (149).

A 2,229 bp xpsD gene was amplified and subcloned into pGEM vector (Promega). A

chloramphenicol-resi stance cassette from pRKP10 (123) was inserted in a kpnI site to disrupt

xpsD. This disrupted xpsD gene was subcloned into suicide vector pOK1 with Apal and Xbal.

The final plasmid constructs were mated into 91-106 to make 91-106::AxpsD via suicide vector-

assisted mutagenesis as described previously (74). Candidates were screened using PCR primers

designed to amplify flanking regions of the cross-over region (Table 2-2).

Bioinformatics Characterization of Ben+ Cosmid Clones

Each ORF was analyzed for sequence homology (BlastP,

http://www.ncbi .nlm.nih.gov/BLAST/), signal peptides and hydrophobicity (SOSUI,

http://sosui .proteome.bio.tuat.ac.j p/sosuiframe0.html), and localization (PSORT,

http://www. psort. org/).

Subcloning of BenB and BenC

The BcnB+ clone pXV6.0 (6.0-kb BcnB+ fragment) was subcloned using a Kpnl/EcoRI

fragment from pXV442 (BcnB ). The 6.0 kb fragment was sequenced (Appendix A-1). For

sequence analysis (Figure 2-2), the Kpnl/EcoRI region was cloned into pBS using Kpnl/EcoRI

enzymes and was sequenced using T7 and SP6 primers from pGEM and custom designed

oligonucleotides generated by the DNA Sequencing Core Laboratory of the Interdisciplinary

Center for Biotechnology Research, University of Florida, Gainesville, FL (UF-ICBR). DNA

sequencing was performed by the DNA Sequencing Core Laboratory of UF-ICBR using the









Applied Biosystems model 373 system (Foster City, CA). Further subcloning of this fragment

was performed using PCR primers (Table 2-2).

The BcnC pXV1.7 clone (1.7-kb BcnC+ Sall/EcoRI fragment) was directionally cloned for

expression in the plus and minus direction for expression of an extracellular metalloprotease

(plus direction) or a glycine-rich protein (minus direction) (Figure 2-3).

Protease Activity Assay

Proteolytic activity was measured by a diffusion assay in agar plates containing skim milk

(casein) as substrate as described previously (34). Five microliters of each bacterial suspension

were applied onto the surface of plates containing 20 mL of 0.5% (wt/vol) skim milk, 2%

(wt/vol) agar and 50mM Tris hydrochloride (pH 8.0) and allowed to incubate for 24 h at 280C.

Zones of clearing around the bacteria due to the degradation of the substrate were measured.

BenA Timing of Activation and Size Analysis

Bacteriocin activity was assessed to determine which fraction(s) contain active BcnA

protein. Supernatants and cells were collected from 18 h nutrient broth (NB) cultures. Cells

were collected via centrifugation, suspended with phosphate buffered saline (PBS) (135) and

sonicated using a digital Sonifier@ unit model S-150D (Branson Ultrasonics Corporation,

Danbury, CT). Fractions were sonicated for 30 s two times on ice. Supernatant and sonicated

cell fractions were assessed for bacteriocin activity by plate assay described below.

Once fractions were prepared, size analysis was performed by separating cell fractions by

Microcon protein filtration system (Millipore, Billerica, MA) with filter cut-offs of 50 kDA

(YM-50) and 100 kDa (YM-100).

In vitro Antagonistic Assay

Each mutant was evaluated for its relative bacteriocin activity produced toward a sensitive

X. euvesicatoria strain based on an in vitro zone of inhibition assay (153). Strains to be tested









were shaken at 280C overnight in NB. The cells were pelleted and resuspended in sterile tap

water. Resuspended cells were then standardized to A600= 0.3 which is approximately 5 x 10s

CFU/mL. A 25 Cll sample was spotted on a Petri plate (100 x 15mm) containing 20 mL NA (five

samples per plate) and grown for 18 h at 280C. After 18 h a suspension (5 x 107 CFU/mL) of X

euvesicatoria strain 91-106 (sensitive indicator strain) was sprayed over the plate using a Sigma

aerosol spray unit (Sigma Chemical, St. Louis, MO). After 24-48 h incubation, zones of

inhibition around the test strain were measured.

A second technique involved incubation of the plates for 24 h, killing the test strains by

inverting glass plates over 2-3 mL of chloroform until all of the chloroform was evaporated,

aerating the plates for 1 h and overlaying the agar surface with 3.5 mL of 0.3% water agar (50

oC) which contained 200 Cll of a 5 x 107 CFU/mL cell suspension of the indicator strain (X.

euvesicatoria strain 91-106). A clear zone of inhibition around test colonies after 24 48 h was

considered indicative of antagonism and scored as bacteriocin-like (BcnA ) activity.

Cell-free extracts were screened for BcnA activity by growing the test cultures for 18 h in

NB followed by centrifugation to pellet cells. The supernatant was then sterilized using a low

protein binding Microcon filter (Amicon, Beverly, MA) with a 0.22 Clm pore size and analyzed

for antagonism by the well diffusion assay method (146). Five millimeter diameter wells were

cut into 20 mL NA plates. Wells were filled with 100 CIL of test filtrates and left for 18 h to

allow diffusion of the liquid into the medium. Plates were then overlaid with 3.5 mL of soft

agar containing 200 Cll of 5 x 10s CFU/mL cell suspension of the indicator strain 91-106. Plates

were examined after 24 h at 280C. Each test was replicated three times.









Evaluation of Immunity

In vitro a~ssa~ys. An immunity in vitro assay was conducted to determine the ORF

responsible for BcnA immunity. All ORFs downstream of ORFA, individually (ORF2, ORF3,

ORF4, ORF5) and in combination (ORF2 to 3 and ORF2 to 5). were subcloned into pLAFR119

and then mated into 91-106 to create 91-106 + pL:ORFA, 91-106 + pL:ORF2, 91-106 +

pL:ORF3, 91-106 + pL:ORF4, and 91-118 + pL:ORF5. Next they were evaluated for immunity

to 91-118::AbcnBC, which expresses BcnA. The test strains were shaken at 280C overnight in

NB. The cells were pelleted and resuspended in sterile tap water. Antagonism assays were

conducted using 91-118::AbcnBC as the producing strain and the mutants with deletion in all

ORFs downstream of ORFA as the test strain.

ORF5 was further tested to confirm its immunity. 91-106 containing either pLAFR119

(pL) or pLAFR119:ORF5 (pL:ORF5) were evaluated for sensitivity to 91-118::AbcnBC (Table

2-1). Strains were shaken at 280C overnight in NB tubes for 18 h. Cells were then washed,

resuspended in sterile tap water and standardized to produce 5 x 106 CFU/mL of 91-

1 18::AbcnBC, then incubated at 280C. After 6 h incubation, 5 x 105 CFU/mL of 91-106 strains

(with pL or pL:ORF5) were added to the flasks. Samples were assayed at 24 h intervals for 96 h.

Each experiment was conducted three times. Population data were transformed to logarithmic

values and standard errors were determined.

In planta a~ssays. The X. perforans and sensitive test I. euvesicatoria and transconjgant

strains (91-106 + pL, 91-106 + pL:ORF2, 91-106 + pL:ORF3, 91-106 + pL:ORF4 and 91-106 +

pL:ORF5) were inoculated at 5 x 107 CFU/mL and 5 x 106 CFU/mL, respectively. The 91-118

strains were inoculated into leaflets by infiltration 18 h prior to inoculation with the sensitive

strain. Six-week-old seedlings of the tomato cultigen Florida 47 were inoculated (15 leaflets









each plant) using a hypodermic syringe as described previously (68). Following inoculation,

plants were incubated at 240C to 280C. In order to determine populations of the sensitive test

strain and transconjugants in leaflets, 1-cm2 leaf disks were removed from inoculated areas,

macerated in 1 mL sterile tap water and dilution plated onto NA amended with the appropriate

antibiotic. Samples were assayed at 24 h intervals from 48 to 96 h. Each experiment was

conducted three times. Population data were transformed to logarithmic values and standard

errors were determined.

Results

Sequence Analysis of Genes Involved in BenA Activity

Previously a 12.1 kb fragment (pXV12.1) was shown to contain five ORFs (ORFA, ORF2,

ORF3, ORF4 and ORF5) potentially important for expression of and for immunity to BcnA (56).

Each putative ORF product was evaluated for presence of a signal peptide and localization

(Table 2-3). Based upon sequence analysis ORFA is predicted to be a water soluble protein,

with a hydrophobicity value of -0.56. It has a predicted location in the bacterial cytoplasm (0.56)

with no predicted signal peptides. The putative ORF2 product has an N-terminal signal peptide,

a hydrophobicity value of 0.28 and is predicted to be localized to the bacterial outer membrane

(0.926) or the bacterial periplasmic space (0.175). The ORF3 product has an N-terminal signal

peptide, a hydrophobicity value of 0.20, and an estimated localization to either the bacterial

periplasmic space (0.939) or the bacterial outer membrane (0.326). The ORF4 has an N-terminal

signal peptide, a hydrophobicity value of -0.36, and two transmembrane helices, from AAl48 to

170 (VTAVAPPPTPTFQPAILTLGAVL) and from AA 176 to 198

(PAAVSWVSPIMGSIVLAPVLYFA) The ORF4 product is predicted to be located in the

bacterial inner membrane (0.187). The ORF5 product has no signal peptide, a hydrophobicity









value of 0. 166, and is predicted to be a water soluble protein located in the bacterial inner

membrane (0.109).

BenA Activity Requires ORFA, ORF2, ORF3 and ORF4

To further analyze the function of BcnA, ORFA, ORF2, ORF3 and ORF4 were

individually disrupted (91-118::AORFA, 91-118::AORF2, 91-118::AORF3, 91-118::AORF4).

Each mutant was tested for bacteriocin activity against I. euvesicatoria strain 91-106. Three

mutants, 91-118::AORFA, 91-118::AORF2 and 91-118::AORF4, lost inhibition activity against

the T1 strain (Figure 2-4). 91-118::AORF3 had inhibition activity, but it was reduced compared

to 91-118.

Localization of BenA Activation

Bacteriocins are produced either in an active or inactive (pre-bacteriocin) form, which is

activated during its secretion. In order to determine the location of BcnA activation, different

cell fractions of a 24 h broth culture of 91-1 18 were tested for bacteriocin activity. Bacteriocin

activity was only found in the supernatant (Figure 2-5). No activity was observed in supernatant

from the less than 50 kDa fraction; however, supernatant from 50 to 100 kDa and above 100 kDa

had inhibitory activity (Figure 2-6), confirming previous results (154). In order to determine if

bacteriocin activity was associated with the cell fraction of 91-1 18, cells were disrupted by

sonication, intact cells removed by filtration through a 0.22 Clm filter and then the bacteriocin

activity was checked by plate assay. The cell fraction did not have activity (Figure 2-5).

Identification of the Immunity Gene

A 4.5-kb fragment downstream of ORFA was previously found to contain the immunity

gene (154). This fragment contains ORF2, ORF3, ORF4 and ORF5 (Figure 2-1). In order to

identify which gene was responsible for immunity I. euvesicatoria strains were created that

expressed each gene under a lac promoter in pLAFR119 (Table 2-1). The positive control









91-106 + pXV12. 1 and 91-106 + pL:ORF5 were not sensitive to 91-118::AbcnBC while 91-106

+ pL:ORF2, 91-106 + pL:ORF3 and 91-106 + pL:ORF4 were sensitive (Figure 2-7).

In an in vivo experiment, bacterial populations of 91-106 + pL:ORF5 and 91-106 +

pLAFR119 (empty vector) strains co-inoculated with 91-118::AbcnBC were determined. 91-106

+ pL:ORF5 reached concentrations of 105 to 106 CFU/cm2 tomato leaf tissue in the presence of

BcnA, whereas 91-106 + pLAFR119 was reduced to thousand-fold less at 100 CFU per mL after

9 h (Figure 2-8). Similarly, greenhouse experiments showed that 91-106 + pL:ORF5 was able to

establish population an average of 1.5 log higher than 91-106 + pLAFR119, when co-inoculated

with 91-1 18::AbcnBC in leaf tissue (Figure 2-9). In addition, 91-1 18::AORFA, 91-1 18::AORF2,

91-118::AORF3 and 91-118::AORF4 mutants all maintained insensitivity to 91-118::AbcnBC in

plate antagonism assays. These results clearly demonstrate that ORF5 confers immunity.

Sequence Analysis of BenB and BenC

It has been shown previously (154) that plasmid pLAFR3 carrying 5.8-kb (pLB5.8) and

5.1-kb (pLC5.1) DNA fragments, BcnB and BcnC, respectively, conferred bacteriocin activity to

sensitive X. euvesicatoria strains. In order to identify genes involved in BcnB and BcnC activity,

subclones of different regions of those DNA fragments were created in pLAFR1 19 (Figures 2-2

and 2-3). Each sub clone was expressed in I. euvesicatoria strains ME90 or 91-106 and the

ability to produce inhibition was tested on NA media using strain 91-106 as an indicator. BcnB

and BcnC subcloned to 3.0-kb and 1.7-kb DNA fragments, respectively, carried on plasmid

pLAFR119, were the smallest fragments that conferred bacteriocin activity (Figure 2-2 and 2-3).

The nucleotide sequence revealed two complete ORFs named bcnB and bcnC. BcnB shows

homology to endoprotease Arg-C with a predicted amino acid size and molecular mass of 466 aa

and 48, 487 MW. BcnB has no N-terminal signal peptide. BcnB is predicted to be located in the

bacterial outer membrane (0.933) or periplasm (0.258). BcnC shows homology to extracellular









metalloprotease with a predicted amino acid size and molecular mass of 401aa and 42, 471 MW,

respectively. BcnC has no N terminal signal peptide. BcnC is predicted to be located in the

bacterial inner membrane (0.351). The introduction of a stop codon (TAA, Table 2-2 in bold) at

the 5' end just downstream of the ATG start codon disrupted BcnB activity when expressed in

ME90 (ME90 + pL3.0mut) compared to the control (ME90 + pL3.0) (Figures 2-2 and 2-10).

Plasmid pLAFR119 has only lac promoter. Directional cloning of BcnB and C genes was

performed in pLAFR119. Plasmids pLB5.8 and pLC5.1 actively expressed BcnB and BcnC,

respectively, without aid of the pL lac promoter (Figure 2-2 and 2-3), suggesting that their native

promoters are functional. BcnB was subcloned to a 3.0 kb fragment. An ORF with

endoprotease Arg-C homology was determined to be responsible for BcnB' activity based on

analysis with of a stop codon (TAA) insertion in the forward direction (Figure 2-2). For BcnC, a

1.7 kb fragment of BcnC was directionally subcloned in pLAFR1 19 in both directions. The

reverse direction BcnC (pL1.7CR) gave very slight bacteriocin activity compared to under

direction of the lac promoter (Figure 2-3).

Purification and Characterization of BenB and BenC

Purification was conducted to evaluate BcnB and BcnC activity. Bacterial supernatant was

concentrated with Microcon YM-100 (Millipore, Billerica, MA) or TCA precipitation (135).

These concentrated samples were run on an SDS-PAGE gel and detected using Coomasie

Brilliant Blue 250 (Pierce Biotechnology, Rockford, IL) or Silver staining (BioRad Laboratories;

Hercules, CA). No bands were detected.

ME90 expressing ORFA was protease negative; however, ME90 expressing BcnB and

BcnC produced clearing zones (2.0 cm and 3.1 cm, respectively) typical of protease activity

(Figure 2-11). Size exclusion analysis was conducted using T3 strain 91-118::AORFA










supernatants. Protease activity was observed from total cell or filtrates of less than 50-kDa with

only minor activity above 50-kDa.

Type II Secretion Mutant Lost Secretion of Amylase and Bacteriocins

The role of the T2SS on delivery was determined for each bacteriocin by plate inhibition

assay. Confirmation of deletion of xpsD was performed via PCR and analysis of starch

hydrolysis media in XpsD mutants (Figure 2-12). The X. euvesicatoria T2SS xpsD mutant

expressing clones, 91-106 + pXV12. 1, 91-106 + pL5.8 and 91-106 + pL5.1, were unable to

produce a zone of inhibition in plate assays while wt 91-106 expression of each clone produced

typical zones for each bacteriocin.

Discussion

In this study, bacteriocins of X perforans were further characterized to determine their

activity and possible functions. Disruption of ORFA, ORF2 and ORF4 abolished BcnA activity,

suggesting that BcnA is part of a multiple component family of bacteriocins. ORF5 was shown

to encode the immunity function for BcnA, making normally sensitive X. euvesicatoria strains

insensitive. This information and the predicted localization in the inner and outer membrane of

the cell suggests that these ORFs make up the necessary parts of a three component system (the

toxin, immunity and a mechanism for delivery) of a typical Gram-negative bacteriocin outlined

by Riley and Wertz (128). Because of size selection (>100 kDA), ORFA is suggested to be the

toxin, ORF2, ORF3 and ORF4 proteins are responsible for delivery and/or possible processing of

BcnA (ORFA product) and ORF5 is the immunity function. All bioinformatics results (SOSUI

and PSORT) suggest ORFA is a soluble cytoplasmic protein. BcnA was only detected in

supernatants and not in detectable levels in the cell fraction of BcnA producing X. perforans

cells. These results suggest BcnA may be activated upon secretion. ORF3 is included in the

model because a mild reduction in antagonism was associated with ORF3 disruption. Based on









predicted localization to the periplasm and outer membrane, perhaps ORF3 aids in transfer of

pre-bcnA (once pre-bcnA is in the periplasm) to ORF2 on the outer membrane. ORF2, ORF3

and ORF4 appear to play secondary roles such as in transport, modification or secretion of

BcnA. The zone inhibition assay and growth rate in vitro and in planta experiments strongly

suggest that ORF5 is responsible for the immunity function. SOSUI predicted ORF5 would

localize to the bacterial inner membrane supported by a positive hydrophobicity value (0.2).

This may suggest ORF5 disrupts BcnA or prevents delivery of active BcnA from entry into the

cell. This is similar to what has been found for the immunity function of Colicin V (ColV) of E.

coli (169).

Col V is one of many known multiple component bacteriocins previously described (8, 99,

169). ColV was used as a reference for basic components of a Gram-negative bacteriocin (42,

43, 55, 56, 64, 169). ColV immunity was previously shown to prevent insertion of ColV into the

inner membrane of sensitive strains of E coli (1 69).

Based on localization analysis (PSORT) we predicted the localization of each ORF

involved in BcnA production. Based on this information and what is known for ColV, we have

developed a basic model for BcnA (Figure 2-13). For the BcnA model, all predicted locations

for ORFs involved in BcnA activity were based on predicted localization and deletion analysis.

The model suggests four steps: (I): pre-BcnA delivery into the periplasm with help of or

chaperoned by ORF4; (II) Processing of pre-BcnA and delivery of the active BcnA outside of the

cell by ORF2 and ORF3; (III) Entry of active BcnA into cells (unknown); (IVa) BcnA

suppressed by ORF5; and (IVb) BcnA inhibition (either in the periplasm or in the cytoplasm).

Previously BcnB activity was localized to a 5.9 kb fragment (153). Only two ORFs were

found within this fragment that contained homology to genes of known function. One was an









amino acid transporter and the other an endoprotease Arg-C. Both genes were isolated and

tested for bacteriocin activity. Only fragments containing the intact endoprotease like gene were

active. This ORF was confirmed using an inserted STOP codon (TAA) at the 5' end of the

fragment which in turn lost activity confirming that the endoprotease was responsible for the

bacteriocin-like activity. Endoprotease Arg-C is a family of serine endoproteases which cleaves

carboxyl peptide bonds of arginine residues. The enzyme has also been shown to cleave Lys-Lys

and Lys-Arg bonds (119).

BcnC was previously localized to a 1.7 kb fragment (61). Two possible ORFs were

located within this fragment one in the plus and one in the minus direction. Directional cloning

analysis shows that the plus directional ORF was responsible for BcnC activity. This gene had

high homology to an extracellular metalloprotease gene family. Metalloproteases are proteolytic

enzymes which use a metal for their catalytic mechanism. Most metalloproteases are zinc-

dependent, while some use cobalt (3).

BcnB and BcnC were tested for protease activity based on homology data. Our Eindings

show that both BcnB and BcnC exhibited protease activity as determined by casein degradation

analysis; however, BcnB produced smaller protease zones than BcnC. The results of the protein

size filtration data were consistent with the predicted size of BcnB (48 MW) and BcnC (42

MW) .

The ORFs responsible for BcnA activity were identified and their possible roles have been

hypothesized. Further research is needed to determine their specific roles. Only one gene was

determined to be necessary for expression of BcnB or BcnC. The protease assays have

determined their roles as proteases; however, further research is necessary to determine the target

of these proteases within sensitive strains.









Table 2-1. Bacterial strains and plasmids used in this study
Strain or plasmid Relevant characteristics Source or reference*
Xanthomonas euvesicatoria
E3-1 NalRSmR (61)
ME-90 RifRKmR (154)
91-106 (154)
91-106::AxpsD XpsD- CmR This study
X. perforans
91-118 RifR (154)
91-118::AORFA ORFA- RifR This study
91-118::AORF2 ORF2- RifR This study
91-118::AORF3 ORF3- RifR This study
91-118::AORF4 ORF4- RifR This study
91-118::AORF5 ORF5- RifR This study
Escherichia coli
DH500 F- recA BRL
C2110 NalR BRL
hPIR Host for pOKl; SpR OriR6K RK2 replicon UB
Plasmids
pBluescript-KS+ Phagemid, pUC derivative; AmpR Stratagene
pLAFR119 TcR rlx+ RK2 replicon BJS
pRK2073 SpR tr+ mob+ (28)
pRKP 10 CmR cassette (123)
pOK1 Suicide vector; SacB (63)
pL5.8 pLAFR119 Kpnl/EcoRI 5.8-kb BcnB+ fragment This study
pL3.0 pLAFR119 Kpnl/Xbal 3.0-kb BcnB' fragment This study
pL3.0mut pLAFR119 Kpnl/Xbal 3.0-kb BcnB~ fragment This study
with a TAA stop codon insertion
pLKH pLAFR 119 Kpnl/HinDIII 2.3-kb BcnB- fragment This study
pLHX pLAFR119 Xbal/HinDIII 2.4-kb BcnB- fragment This study
pL5.1 pLAFR119 HinDIIII/EcoRI 5.1-kb BcnC' fragment This study
* BRL, Bethesda Research Laboratories, Gaithersburg, MD; Stratagene, Stratagene Inc., La
Jolla, CA; BJS, B. J. Staskawicz, University of California, Berkeley, CA; UB, U. Bonas, Martin-
Luther-Universitait, Halle, Germany.




























































Xhol
Ecold

KpnI
Xhol


Table 2-2. PCR primer used in bcnA, bcnB and bcnC analyses for genetic manipulation

plasmid Primer name rsicon Primer sequence length GC Tm (C)


22 54.5 61


pLAFIR NP5


NP3
A5
A3
ORF5F
ORF5R
ORF3F
ORF3]R
ORF2F
ORF2]R
ORF4F
ORF4]R
B5' new
B32
BORFlF
BORFIR
BORF2F
BORF2R
BORF3F
BORF3R
BORF4F
BORF4R
BORF5F
BORF5R
BORF7F
BORF7R
B5Xhol
C5'
C3'
C5)Glol
xpsDF
xpsDR


CCCTTCACCAAGTTCGACGACA
GCGGGTGCCGTGCTCGTGTT
CCTCGAGATGCGCCACCCGTCGG
CCTCGAGCAGCAAAAGCTGATAGAGC
GGGGAAGCTTCAGGGTGGCGGCAAGGGA
GGGGAAGCTTGGGCTTCTCTGGAAGCGGAC
CCCGAAGCTTCCGGTTGACCTCTATGTAGATGGATGC
CCCGAAGCTTCCCAGTGCAAATGTAAGCCGCGAC
GGGGAAGCTTACACAGGACGGGACATGCACAG
GGGGAAGCTTACAACCTCCACATCTCGCACCG
CCCAAGCTTGCCGGATGCGACATTGTTGCGC
CCCAAGCTTGCTTGGTTCAAGCTCATCACC
CGGAATTCCAATCGCAAGAACGCGATG
CGGGTACCCTGGCCGAAGTAGGTGGAAAT
ATGGGCTTGTCGGCCACATAATCGTCACAA
TTGTGACGATTATGTGGCCGACAAGCCCAT
AACGAACGAAGGTTACACTGGCTCCACCAT
ATGGTGGAGCCAGTGTAACCTTCGTTCGTT
ATGAATCGCAAGTAAGCGATGTATCTGGCG
CGCCAGATACATCGCTTACTTGCGATTCAT
ATGGCTGCAAATTGATAATGCGCTCACGGT
ACCGTGAGCGCATTATCAATTTGCAGCCAT
CAGCCCGCGCGATTAGATGACCATTGCCAT
ATGGCAATGGTCATCTAATCGCGCGGGCTG
ATGGACGATCGCTAACCGTCGATCCGCTTC
GAAGCGGATCGACGGTTAGCGATCGTCCAT
CCTCGAGATGAATCGCAAGAACGCG
CGGAATTCCGTGAAGAACGTCTTCCTC
GGGGTACCCTTGTCGTCA TCGTTCTGCGCCGGAGTGTT
CCTCGAGGTGAAGAACGTCTTCCTC
ATGACGCCGCGCCTGTTTCC
CCCTTCCA~TC~~TCRAAGTOOA


BenA















BerB























BenC



T2SS


Xhol
Xhol
HuoDIII
HuoDIII
HmDIII
HlnDII
HlnDII
HlnDII
HmDHII
HlnDII
Ecold
KpnI












Table 2-3. Characterization of bacteriocin ORFs associated with the expression of BcnA, BcnB
and BcnC
Location
AAa Size (kDa) Signal Peptideb Localization" vale hydrophobicitib

BcnA ORFA 1012 111.0 none bacterial evtoplasm 0.56 -0.56
Outer membrane or 0.926 &
ORF2 124 13.6 ATLCIFLPLLCAKASAAPYVV NIGNIYTR 0.28
periplasmic space 0.175
Outer membrane or 0.326 &
ORF3 100 11.0 MRFYRISLLALIIFASPRASA. 0.20
periplasmic space 0.939
ORF4 290 31.9 MNKC SDAYGIYLRTLFY`FMhYTLFCTSASAQYIRY inner membrane 0.187 -0.36

ORF5 145 15.9 none inner membrane 0.109 0.17

Outer membrane or 0.933 &
BcnB bcnB 465 48.5 none .-0.07
periplasmic space 0.258
BcnC bcnC 400 42.5 none inner membrane 0.351 -1.44

a BcnA ORF predicted amino acid size (AA) were previously described (154).
b Prediction of signal peptide was performed with SOSUI (http://sosui.proteome.bio .tuat.ac jp/sosuiframe0 .html).
" Prediction of localization was performed using PSORTb v.2.0 (Intpl w\ int\.psort.org/).












Bcn-A operon
ORFA 2 3 4 5

:Kb) 1 4 5 6 7 8 9 lo Activity Immunity
[] AORF2 +
L] AORF3 +
O O AORF4 +
SO AOR F5
1 \\ 1 1 10S I p'S'12.1 + +
1 \2 I 91106+ AORF4- +


1 191~~~-10+45


91-106 + ORF2-5R
91-106 + ORF2-5 +
91-106 + ORF2-3 +
91-106 + ORF2
91-106 + ORF3
91-106 + ORF4
91-106 + ORF5- +


ORFA 3




pOK1:AORF2


\\


Figure 2-1. BcnA diagram showing position of individual ORFs and positions and fragment
constructs. The grey arrows indicate the directional expression of fragments under the
lac promoter of pLAFR1 19. The circle on ME90 + AORF4 indicates the location of a
transposon insertion. B. Diagram of deletion construct for ORF2.


~


T




















r
r


Amino acid transporter


bcn-b (endoprotease Arg-C)


(Kb) 0.5 1


1.5 2 .0 2.5 3 .0 3.5 4.0 4.5 5.0 5.5


plasm id
pL5.8
pL:3.0
pL:3.0M
pL:KH
pL:XH


BCN


1


stop


Figure 2-2. BcnB diagram showing position of individual ORFs and directional expression (grey
arrows) and activity (table) of fragment constructs under the lac promoter of
pLAFR119. The x indicates an artificial stop codon (TAA) insertion in frame of the
bcnB sequence.





bcn-c (Extrace


1.5 2.0 2.5 3.0 3.5 4.0 4.5

glycine-rich protein


llular metalloprotease)


(Kb) 0.5 1


plasmid
pL5.1
pL2.5
pL1.7CR
pL1.7


BCN


Figure 2-3. BcnC diagram showing position of individual ORFs and directional expression (grey
arrows) and activity (table) of fragment constructs under the lac promoter of
pLAFR119.











Test Strain WT T3 AORFA AORF2 AORF3 AORF4
inhibition zone +++ ++
Figure 2-4. Antagonism assays of bacteriocin-like activity against I euvesicatoria strain 91-106
were assessed using cell-free supernatants of individual ORF knockout mutants of
91-118.


9



















Test strain WT T3 AORFA AORF4
Inhibition zone +




Cell fraction I l~C31




Test strain WT T3 AORFA AORF4
Inhibition zone

Figure 2-5. Antagonism activation assays of bacteriocin-like activity againstI~ euvesicatoria
strain 91-106. Supernatant from Xk perforans was separated into supernatant and cell
fractions.


Supernatant





















Size range (kDa) All X < 50 50 < X < 100 X >100
Inhibition zone +++ ++ ++


Figure 2-6. Secretion assays of bacteriocin-like activity against I. euvesicatoria strain 91-106.
Supernatant from X. perforans was separated based on size exclusion technique into
three fragments, less than 50, 50 to 100 and over 100 kDa.



















challenge strain pLAFR119 pXV12.1 ORF2 & 3 ORF2 to 5
inhibition zone + +









challenge strain pL:ORF2 pL:ORF3 pL:ORF4 pL:ORF5
inhibition zone + + +

Figure 2-7. Immunity assays of bacteriocin-like activity against I euvesicatoria strain 91-106
using 91-118::AbcnBC as the producing strain. Bacteriocin sensitivity candidates
were screened to identify ORFs expressing immunity.




















034









0 3 6 9
Time after inoculation (h)

Figure 2-8. In vitro populations of I euvesicatoria strain 91-106 transconjugants with
[pLAFR119 (m) and pLAFR:ORF5 (x)] and without [pLAFR119 (+),
pLAFR1 19:ORF5 (A)] co-inoculation of 91-1 18::AbcnBC. Error bars indicate the
standard error.



















OO






48 72 96
Time after inoculation (h)


Figure 2-9. Inplanta populations of I euvesicatoria strain 91-106 containing plasmid
pLAFR119 (0) or pLAFR119:ORF5 (0), respectively, when co-inoculated with
91-1 18:: AbcnBC (BcnA expression only) in leaflets of tomato cultigen Bonny Best.
Error bars indicate the standard error.


















Test Strain ME90 + pLAFR119 ME90 + pL5 8 ME90 + pLKH ME90 + pL3 0 ME90 + pL3 Omut
Inhlbition zone + +

Figure 2-10. Antagonism assays of bacteriocin-like activity against I euvesicatoria strain
91-106 were assessed using cell-free supernatants of individual ORF knockout
mutants of 91-1 18.


















Strain WT T3 ME90 + pLAFR ME90 + A ME90 + B ME90 + C
Protease activity +++ + +++


Size Range (kDa) All X < 50 50 < X < 100 X > 100
Protease activity +++ +++ + +

Figure 2-11i. Protease activity of bacteriocin candidates. A. Protease assay where 5 C1L of 5 x 10s
CFU/mL of each strain was plated onto 0.5% skim milk agar and incubated for 24 h
at 280C. B. Fractions from X. perforans were separated based on size exclusion
technique into three fragments, less than 50, 50 to 100 and over 100 kDa and
analyzed for protease activity.



















Test strain 91-106 91-106::AxpsD
Protease activity+


Test strain 91-106 91-106 + pL 91-106 + pXV12.1 9-0 xp

BcnA activity +++ -++ ++

Figure 2-12. Evaluation of type two secretion system mutant xpsD. A. Proteinase assay of
91-106 and 91-106::Arp~zsD. B. Antagonism assays of BcnA using cell-free
supernatants using I. euvesicatoria strain 91-106 as an indicator strain.











Step 3 <(cn)
Outside





Step 4ar\sr tep 4b Step, 2 ei l





J Step 1 Inside





Figure 2-13. BcnA Model for ORFs predicted involved in BcnA activity based on predicted
localization and deletion analysis. Step 1: pre-BcnA delivery into the periplasm by
ORF4. Step 2: Processing of pre-BcnA and delivery of the active BcnA outside of the
cell by ORF2 and ORF3. Step 3: Entry of active BcnA into cells (unknown). Step 4 a:
BcnA suppressed by ORF5. Step 4b: BcnA inhibition (either in the periplasm or in
the cytoplasm).









CHAPTER 3
ANALYSIS OF PATHOGENICITY MUTANTS OF XANTHOM~ONASPERFORANS AND
THEIR EFFECT ON BACTERIOCIN EXPRESSION

Fresh-market tomato production accounts for more than 50% of the harvested acres and

63% of the national yield in the southeastern U. S., with approximately 40,000 acres in Florida

(29). Bacterial spot of tomato, incited by Xanthomona~s euvesicatoria, is a devastating disease of

tomato in Florida, the Caribbean and worldwide (11, 108, 114, 118, 132). There are at least

three important management strategies to reduce severity and incidence of bacterial tomato

diseases: reducing initial inoculum, minimizing plant susceptibility (natural resistance) and

chemical control (copper-based chemicals and antibiotics). Although chemical control using

copper bactericides is routinely used to control bacterial spot, these efforts are often futile

because of the presence of copper-tolerant strains of the bacterium (90). Adding mancozeb, a

fungicide, to copper sprays was shown to improve control efficiency and was shown by Marco

and Stall (90) to control copper-tolerant strains. However, they also showed that this treatment is

insufficient when conditions favorable for disease development exist.

Recently, new strategies have emerged that could be utilized as alternative management

practices. These include using bacteriophages specific to the target bacterium (5), application of

plant activators that induce systemic acquired resistance (SAR) in the plant to bacterial

pathogens (89) and application of bacterial biological control agents (49, 67). These new

strategies focus on reducing initial epiphytic or internal populations of the potential pathogen.

Several studies on xanthomonads have shown a relationship between epiphytic populations of

the pathogen and disease severity (73, 81). Lindemann et al. (86) demonstrated a strong

correlation between a threshold level of epiphytic populations of Pseudomona~s syringe and the

occurance of disease in the field. Reducing epiphytic populations may play a key role in a

successful biological control strategy.









Recently Hert et al. (61) demonstrated that populations of I euvesicatoria strains were

reduced significantly by bacteriocin-producing X. perforans strains in the Hield. These results are

supportive of the potential of bacteriocins for controlling bacteriocin-sensitive strains and may

provide a new approach to biological disease control. Non-pathogenic Hrp mutants of X

perforans provided significant levels of control (88, 102) against bacteriocin sensitive strains;

however, there were still unacceptable levels of disease. In other studies, Hrp mutants were

created in a bacteriocin expressing Ralstonia solan2acearum strain for control of wild-type (wt) R.

solan2acearum strains (41, 49, 152). These studies revealed that slightly pathogenic hrp mutants

that were able to achieve higher levels of root and stem tissue colonization provided higher

levels of control than other hrp mutants with minimal colonization capabilities (49, 152).

Etchebar et al. (41) suggested that there was a relationship between the degree of colonization of

the xylem by the mutant and the level of control of the wt R. solan2acearum. I. perforans also

causes disease, therefore ifX. perforans is to be used as a biological control agent, it is necessary

to reduce the virulence of X perforans to levels that are not deleterious to the plant.

There has been progress in identifying genes involved in bacterial virulence of

Xanthomona~s species (3, 4, 44, 58, 64, 165). Identification of these genes as potential targets to

create strains attenuated in virulence has become much easier as a result of genome sequencing

of xanthomonads. Thieme et al. (150) sequenced the X. euvesicatoria genome and estimated that

480 putative pathogenicity factors and associated genes are found in the genome. These factors

were placed into six categories: (I) secretion systems, (II) flagellum, (III) secreted proteins (via

type III secretion system (T3 SS)), (IV) detoxification, (V) surface structure and adhesion and

(VI) quorum sensing. Based on previous studies, several candidate genes (opgH, avrBs2, hpaA,









hpaB, hpaC, xopA, xopD and gumD) appeared interesting for use in this study (18, 17, 21, 63,

76, 101, 106).

One known X. perforans pathogenicity-attenuating mutant served as our model candidate,

opgHxcy. The opgH mutant was previously shown to have an attenuated virulence phenotype

(101), exhibiting reduced disease severity and growth curve in susceptible tissue and triggering

delayed hypersensitive reaction (HR) in a resistant plant genotype. A second candidate gene, the

effector avrBs2, was chosen as another candidate as a result of its role in virulence of I

euvesicatoria in pepper (76). It has been shown that avrBs2 mutants are less virulent on

susceptible hosts (53, 76, 146).

Genes hpaA, hpaB and hpaC were shown to be involved in virulence. These genes play a

role in pathogenicity as an effector (hpaA) and as chaperones (hpaB and hpaC) of effectors of the

T3SS (18, 17, 63). HpaA appears to function as an effector molecule in I. euvesicatoria since

disruption of hpaA eliminates disease symptoms in tomato and pepper plants without affecting

the ability to elicit a hypersensitive response. HpaB and HpaC were shown to form an

oligomeric protein complex and interact with two classes of effectors (class A containing XopJ

and XopF 1 and class B containing AvrBs3 and XopC) and HrcV of the T3 SS (18).

Two genes that code for proteins that are designated Xanthomona~s outer proteins (Xop)

were also chosen as potential candidates that may affect virulence. XopA and XopD are secreted

by the T3SS and thus represent putative effector proteins. XopA is necessary for both inplanta

growth and full virulence (106).

GumD of the gum operon is involved in xanthan gum biosynthesis (7, 75). Xanthan gum

is a high molecular weight extracellular heteropolymer produced by xanthomonads and has been









implicated as a virulence factor based on deletion analysis in X. c. py. campestris. Xanthan gum

minus gumD mutants exhibited dramatic delay in disease symptoms (21).

In this study, mutations were created in the candidate genes described above to determine

their role in pathogeni city and pos sible use as pathogenically-attenuated b acteri ocin-expres sing

biological control agents.

Materials and Methods

Bacterial Strains, Plasmids and Culture Conditions

Strains of X perforans and I. euvesicatoria were grown on nutrient agar (NA) medium

(Difco Laboratories, Detroit, MI) at 280C (Table 3-1). Strains ofEscherichia coli were grown on

Luria-Bertani (LB) medium at 370C (Table 3-1) (97). All strains were stored in 20% glycerol in

sterile tap water at -800C. Bacterial cultures for plant inoculations were grown in nutrient broth

(NB) (Difco Laboratories, Detroit, MI) for 18 h at 280C with shaking (100 rpm). Cells were

pelleted by centrifugation (4,000 x g, 15 min) and resuspended in sterile tap water. Bacterial

suspensions were standardized to an optical density at 600 nm (OD600) = 0.3 (5 x 108 CFU/mL)

with a Spectronic 20 spectrophotometer (Spectronic UNICAM, Rochester, NY) and

subsequently diluted in sterile tap water to appropriate cell densities for individual experiments.

Antibiotics were used to maintain selection for resistance markers at the following

concentrations: tetracycline (Tc) 12.5 Clg/mL; rifampicin (Rif) 100 Clg/mL; spectinomycin (Sp)

50 Clg/mL; kanamycin (Km) 50 Clg/mL; chloramphenicol (Cm) 34 Clg/mL; streptomycin (Sm)

200 Clg/mL; and nalidixic acid (Nal) 50 Clg/mL.

Plant Material

Seeds of tomato (Lycopersicon esculentum) cy. Bonny Best were planted in Plugmix (W.

R. Grace & Co., Cambridge, MA). After 2 weeks, the emerged seedlings were transplanted to









Metromix 300 (W. R. Grace & Co.) in 10-cm plastic pots. Seedlings were grown in the

greenhouse at temperatures ranging from 25 to 350C.

Primer Design

Candidate genes (Table 3-2) were amplified using primers designed as follows. BLAST

search analysis was conducted to locate conserved regions flanking the candidate gene sequence

by scanning genomic sequences of closely related Xanthomona~s species (i.e. I. euvesicatoria

strain 85-10 (NC_007508.1), X. oryzae py. oryzicola (AY875714.3), X. oryzae py. oryzae strain

KACC10331 (NC_006834. 1), I. axonopodis py. citri strain 306 (NC_003919.1) andX.

campestris py. campestris strain ATCC 33913 (NC_003902.1). Primers were designed to

conserved regions for amplification of the corresponding regions in X. perforans strain 91-1 18

(Table 3-2). Candidates were confirmed by sequence analysis. Sequencing of the clones was

conducted at the ICBR sequencing facility (University of Florida, Gainesville, FL) with the

Applied Biosystems model 373 system (Foster City, CA).

Generation of Mutants

Candidate genes for creating attenuated mutants of X perforans (avrBs2, hpaA, hpaB,

hpaC, xopA and xopD) were disrupted using either restriction digestion or PCR-assisted deletion

mutagenesis. All candidate genes were amplified and deleted using PCR primers described in

Table 3-2. Each attenuated candidate was individually cloned into pGEM (Promega, Madison,

WI) or pTOPO (Invitrogen, Carisbad, CA) then an internal fragment was deleted using

restriction digestion or PCR assisted deletion mutagenesis and replaced with a chloramphenicol

resistance cassette.

An avrBs2 mutant was created as follows. A fragment containing avrBs2 was amplified

by PCR using primers avrbs2F and avrBs2R and cloned into pTOPO to create pTOPO:avrBs2.

For disruption of avrBs2, two primer pairs, avrBs2F, avrBs2DD2F, avrBs2DD2R and avrBs2R









were utilized to amplify portions of the 5' and 3' ends separately to delete a 174-bp fragment not

amplified in either set of primers (Figure 3-1A). Both fragments (5' and 3' products) were

cloned into pTOPO creating pTOPO:avrBs2F and pTOPO:avrBs2DR. Primers avrBs2DD2F

and avrBs2DD2R contain artificial Sall restriction sites. The pTOPO:avrBs2-3 was then

digested with Sall (artificially created by avrBs2DD2R) and pTOPO restriction site Spel, then

ligated into pTOPO:avrBs2DF, as illustrated in Figure 3-1A, to create pTOPO:AavrBs2. Next,

the artificial Sall restriction enzyme sites from avrBs2DD2F and avrBs2DD2R were utilized to

insert a Cm resistance gene cassette (124) to create pTOPO:AavrBs2. Finally, using pTOPO's

MCS restriction sites BamBBBBBBBB~~~~~~~~~HI and Apal, the deleted avrBs2 was ligated into pOK 1 to create

pOKl:AavrBs2. The final construct was then mated into 91-118 as described later in this

section.

The hpaA gene was amplified by PCR using primers hpaAF and hpaAR, containing

artificial HinDIII restriction sites and cloned into pGEM to create pGEM:hpaA. For disruption

of hpaA divergent PCR primers hpaADF and hpaADR were utilized to delete a 446-bp internal

fragment ofhpaA (Figure 3-1B). Artificial Smal restriction enzyme sites were added by primers

hpaADF and hpaADR, then utilized to insert a Cm resistance gene cassette (124) to create

pGEM:AhpaA. Finally, using pGEM' s MCS restriction sites Spel and Apal, the deleted hpaA

was ligated into the pOK1 to create pOKl:AhpaA. The final construct was then mated into 91-

118 as described later in this section.

For deletion of hpaB, a fragment containing hpaB was amplified by PCR using primers

hpaBF and hpaBR, containing artificial HinDIII restriction sites (not used in this study) and

cloned into pGEM to create pGEM:hpaB. For disruption of hpaB, a Cm resistance gene cassette

(124) was inserted into a native KpnI restriction site within hpaB to create pGEM:AhpaB.









Finally, using pGEM's MCS restriction sites Spel and Apal, the deleted hpaB was ligated into

the pOK1 to create pOKl:AhpaB. The final construct was then mated into 91-118 as described

later in this section.

A fragment containing hpaC was amplified by PCR using primers hpaCF and hpaCR and

cloned into pGEM to create pGEM:hpaC. For disruption of hpaC, divergent PCR primers

hpaCDF and hpaCDR were utilized to delete a 58-bp internal fragment of hpaC. Artificial

HinDIII restriction enzyme sites were added by hpaCDF and hpaCDR and utilized to insert a Cm

resistance gene cassette (124) to create pGEM:AhpaC. Finally, using pGEM's MCS restriction

sites Spel and Apal, the deleted hpaC was ligated into the pOK 1 to create pOK: A. hpaC. The

final construct was then mated into 91-118 as described later in this section.

A fragment containing xopA was amplified by PCR using primers XopAF and XopAR,

containing artificial HinDIII restriction sites, and cloned into pGEM to create pGEM:xopA. For

disruption of xopA two Aval restriction sites, located within pGEM:xopA were utilized. This

Aval deletion completely deleted xopA leaving flanking DNA for marker exchange. Finally,

using pGEM' s MCS restriction sites Spel and Apal, the deleted xopA was ligated into the pOK1

to create pOKl:AxopA. The final construct was then mated into 91-118 as described later in this

section.

A fragment containing xopD was amplified by PCR using primers xopDF and xopDR and

cloned into pGEM to create pGEM:xopD. For disruption of xopD divergent PCR primers

xopDDF and xopDDR were utilized to delete a 134-bp internal fragment of xopD. Artificial

HinDIII restriction enzyme sites were added by xopDDF and xopDDR were utilized to insert a

Cm resistance gene cassette (124) to create pGEM:AxopD. Finally, using pGEM's MCS









restriction sites Spel and Apal, the deleted xopD was ligated into the pOK1 to create

pOKl:AxopD. The final construct was then mated into 91-118 as described later in this section.

A fragment containing gunzD was amplified by PCR using primers GumDF and GumDR

and cloned into pGEM to create pGEM:gunzD. Two Ncol restriction sites within pGEM:gunzD

were used for disruption of gunzD. This Ncol deletion removed a 400bp segment of the gunzD

gene leaving 5' and 3' portions of flanking DNA for marker exchange. Finally, using pGEM's

MCS restriction sites Spel and Apal, the deleted gunzD was ligated into the pOK1 to create

pOKl:AgunzD. The final construct was then mated into 91-118 as described later in this section.

All candidates were confirmed by PCR using primers pOKlF and pOKlR (Table 3-2) for

sequencing of each candidate deletion at the DNA Sequencing Core Laboratory, as mentioned

previously. Once sequence analysis confirmed deletion of each candidate, suicide vector assisted

mutagenesis was performed as described previously (74). Candidates were screened using PCR

primers designed to amplify flanking regions of the cross-over region (Table 3-2).

Deletion mutant 91-1 18::AavrBs2 was also confirmed in resistant cultivars of tomato

(transgenic VS36 containing 3 5S::Bs2 (148)) and pepper (ECW-20R) for loss of HR. Leaves

were infiltrated with a bacterial suspension of 5 x 10s CFU/mL using a hypodermic syringe as

described previously (68) and scored for presence or absence of a hypersensitive response (HR)

after 24 h and observed for 72 h.

Growth Room Growth Curve Assays

Growth room assays were conducted to compare the growth curve of the deletion mutants

with that of parent strain 91-118. The strains were inoculated at 3 x 105 CFU/mL into leaflets of

6-week-old seedlings of the tomato cultivar Bonny Best. Leaflets were infiltrated (15 leaflets per

strain) using a hypodermic syringe and needle, as described previously (68). Following









inoculation, plants were incubated at 240C to 280C. Three samples were taken for each treatment

every 24 h for 96 h. Populations were quantified by macerating 1-cm2 leaf disks in 1 mL sterile

tap water and dilution plating onto NA medium amended with the appropriate antibiotic. Plates

were incubated at 280C and colonies were counted after 48 to 72 h. Population data were log

transformed and standard errors were determined. The overall growth curve was determined by

calculating the area under the population progress curve (AUPPC). The AUPPC is a

modification of the area under the disease progress curve (AUDPC) which has been used to

analyze population progress (136): standardized AUPPC = E [(x, + x, .1)/2](t, to ), where x is

population density in loglo CFU per cm2 and t is time in hours. The AUPPC values for the

strains were compared by analysis of variance and subsequent separation of sample means by

Waller-Duncan multiple range test using SAS version 9.0 (SAS Inc., Cary, NC). Each

experiment was conducted three times.

Greenhouse Disease Severity Assay

Greenhouse pathogenicity assays were conducted to compare symptom development

incited by the mutants and wt 91-118. In each test, four young Bonny Best plants (four-true-leaf

stage) were inoculated with each strain by dipping into sterile tap water suspensions containing

3 x 106 CFU/mL of bacteria and 0.025% Silwet L-77 (Loveland Industries, Inc., Greeley, CO)

for 15 s. Plants were maintained in the greenhouse during the evaluation period. The plants were

assessed for disease severity 14 to 21 days after inoculation. Disease assessments were made

based on leaf and stem ratings compiled from three separate greenhouse inoculation tests.

Growth Room Antagonism Assay

Antagonism assays were performed to determine the effect of wt and mutant X perforans

strains on a sensitive X. euvesicatoria strain E3-1. Internal and external/leaf surface

(phyllosphere) populations were evaluated using two different antagonism assay techniques.









Internal antagonism a;ssssssssssssssay. Six-week-old seedlings of the tomato cultigen Florida 47 were

inoculated with a 5 x 107 CFU/mL suspension ofX. perforans (15 leaflets per strain) using a

hypodermic syringe as described previously (68) followed 12 h later by injecting a 5 x 106

CFU/mL suspension of bacteriocin-sensitive X. euvesicatoria strain E3-1. Each treatment

consisted of three replications. Following inoculation, plants were incubated at 240C to 280C.

In order to determine populations of the sensitive strain (E3 -1sm"al) in leaflets, 1-cm2 leaf disks

were removed from inoculated areas, macerated in 1 mL sterile tap water and dilution plated

onto NA amended with Nal and Sm to qualify E3-1 populations. Samples were assayed at 24 h

intervals for 96 h. Each experiment was conducted three times. Population data were log

transformed and standard errors were determined. AUPPC values (calculated as described

above) were compared by analysis of variance and subsequent separation of sample means by

Waller-Duncan multiple range test using SAS version 9.0 (SAS Inc., Cary, NC).

Phyllosphere antagonism assay. Growth room phyllosphere antagonism assays were

conducted to determine if the gene deletions affected the levels of antagonism toward external

leaf populations of X. euvesicatoria strain E3-1 by comparing the antagonistic ability of the

mutants with the parent strain 91-118. Six-week-old Bonny Best tomato seedlings were dipped

into a suspension of the wt or mutant strain of X. perforans adjusted to 5 x 107 CFU/mL

amended with Silwet L77 (0.025%). Seven days later the plants were sprayed with a 5 x 107

CFU/mL suspension of X. euvesicatoria strain E3-1. Following spray inoculation, plants were

incubated at 240C to 280C. Leaf tissue was sampled every 24 h for 96 h to quantify E3-1

populations. Three leaflets were taken at each time point. Each leaflet was weighed, placed in a

polyethylene bag containing 10 mL of sterile tap water and shaken on a wrist action shaker

(Burrel Co., Oakland, CA) for 20 min. The leaf-wash was then dilution plated on NANalSm to









selectively determine the concentration of E3-1 colonies. Population data were analyzed

following logl0 transformation and standard errors were determined. AUPPC values (calculated

as described above) were compared by analysis of variance and subsequent separation of sample

means by Waller-Duncan multiple range test using SAS version 9.0 (SAS Inc., Cary, NC). Each

experiment was conducted three times.

Results

Sequence Analysis of Attenuated Mutant Candidate Genes

The eight candidate genes selected for disruption (Table 3-3), were amplified and cloned

from X. perforans strain 91-118 and sequenced (Appendix A-2, A-3, A-4, A-5, A-6 and A-7).

Sequence analysis of each gene was conducted to determine relatedness between X. perforans

AA sequence and other proteins using BLAST search protocol (Blastp) at the NCBI website

(http://www.ncbi .nlm.nih.gov/BLAST/). For hpaA, hpaB, hpaC and avrBs2 the nucleotide and

deduced amino acid sequences of these genes had very high homology (75% to 100%) (Table 3-

4) to the corresponding genes in other xanthomonads such as I. euvesicatoria strain 85-10

(NC_007508. 1), X o. oryzicola (AY875714.3), X. o. oryzae KACC10331 (NC_006834. 1), I. a.

citri strain 306 (NC_003919. 1), X. c. campestris strain ATCC 33913 (NC_003902. 1) and X. c.

glycines (AF499777. 1) (Appendix A-8, A-9, A-10 and A-11 and Table 3-4). XopA and XopD

had less AA homology than all other attenuated mutant candidate genes (Appendix A-12 and A-

13 and Table 3-4). XopA only had high homology to I. euvesicatoria strain 85-10 (100%) and

much lower homology with X c. glycines (47%). XopD only had high homology to I

euvesicatoria strain 85-10 (85%) and X. c. campestris strain ATCC 33913 (74%). XopD was

also found to have 86% nucleotide homology with P. syringae py. eriobotryae gene psvA

(AB018553).









Population Dynamics and Pathogenicity Assays

In growth room experiments, there were three separate groups according to overall AUPPC

values. The first group includes wt 91-118, 91-118::AhpaB and 91-118::AavrBs2. Although

91-118::AavrBs2 was not considered significantly different overall, it was significantly different

at the 96 and 120 h time points according to standard error. Populations of 91-1 18 and 91-

118AhpaB exhibited a normal growth curve over the 120 h sample period and, based on the

AUPPC (Table 3-5). The second group included 91-118::AxopA and 91-118::AgumD. These

mutants were not significantly different from each other and grew 0.5 to 0.75 loglo CFU/mL

lower than wt 91-118 throughout the experiment (Figure 3-2). 91-118::AhpaC was between

groups according to overall significant difference and was not considered significantly different

according to standard error at 48 and 72 h from 91-118::AxopA and 91-1 18::AgumD. The third

group consisted of only 91-1 18::AopgH. 91-1 18::AhpaC was considered significantly different

from 91-118::AopgH at 24, 48 and 72 h according to standard error. 91-118::AopgH consistently

grew 1 to 1.5 loglo CFU/mL lower than wt 91-118. 91-118:AxopA and 91-118::AavrBs2. The

largest reduction was observed for 91-1 18::AopgH which was consistently significantly lower

populations over the experiment.

Greenhouse disease severity experiments were conducted to determine the effects of the

mutations on the ability of 91-1 18 to cause disease in plan2ta. 91-1 18::AxopA and

91-118::AopgH mutants induced 1.5 to 3.2 times less disease than 91-118 (Figure 3-15 and Table

3-5). The largest reduction in disease severity was observed in 91-118::AopgH (Table 3-5).

The avrBs2 disruption mutant was confirmed on pepper and tomato genotypes expressing the

Bs2 corresponding R gene. A resistance response (HR) was only observed in tomato leaves

infiltrated with wt 91-118, infiltration with 91-118:AavrBs2 did not give an HR (Figure 3-3).










Antagonism Assays

Growth room (internal and external) antagonism assays were conducted to determine the

antagonistic ability of deletion-mutant candidates toward E3-1. Treatment with water prior to I.

euvesicatoria resulted in a normal growth curve over the 96-h sampling period in both

experiments based on the AUPPC (Figures 3-4 and 3-5).

All attenuated mutant candidates significantly reduced I. euvesicatoria populations in the

internal antagonism experiment (Figures 3-5). Wt 91-118 gave the most significant reduction in

E3-1 populations. There were 2 groups of mutants from this experiment, however all were not

significantly different from one another. 91-118::AhpaB, 91-118::AxopA and 91-118::AgumD

gave the most reduction of the mutants tested, however they were not significantly different from

any the other strains overall. Looking at standard error, however, they appeared to be

significant from hpaC at 96 h. The second group of mutants (91-1 18::AhpaC and

91-118::AavrBs2) was non-significantly different from the X. euvesicatoria 91-106 strain

treatment.

In the external antagonism assay, all attenuation mutant candidates tested were similar to

wt 91-118 in antagonism, however they were significantly different overall (Table 3-6). At 48

and 72 h, however, 91-1 18::AopgH was not significantly different according to standard error.

Overall, wt 91-118 and all 91-118 mutants tested were significantly effective at reducing

populations of E3-1.

Discussion

The goal of this proj ect was to identify genes mutants that would provide a dramatic

reduction in disease symptoms while still maintaining the significant expression levels ofBcnA

and BcnC. Several pathogenicity related genes (hpaA, hpaB, hpaC, xopA, xopD, avrBs2, gumD)










were evaluated for their possible attenuating effects when in X. perforans. 91-118::AopgH

served as the model system. This opgH mutant had an attenuated phenotype in disease severity

and growth curve experiments, as observed previously (101), and maintained its ability to reduce

X. euvesicatoria significantly better than water and I. euvesicatoria controls. The opgH mutant

was selected as a model system for these experiments because it was also chosen for further

investigation in the form of field experiments in Chapter 4.

Growth curve analysis suggests mutants (91-118::AxopA, 91-118::AgumD, 91-118::AhpaC

and 91-118::AopgH) were effected in overall fitness within the plant by exhibiting a reduced

growth curve peak compared to wt 91-118. There are a few hypotheses that may explain why

we observed this reduction. One explanation could be associated with the effect of the mutants

on the overall fitness of the bacterium. A second hypothesis may be that there is recognition of

the pathogen by the plant due to the mutations created. It does not appear that these mutants

would eventually reach the levels based on the stationary phase of the curve at 96 to 120 h

supporting the second hypothesis. Further research is needed to solidify which or if both

hypotheses is correct.

Of the mutants created in this study, 91-1 18::AgumD and 91-1 18::AopgH exhibited the

overall characteristics we were looking. Both mutants caused significant reductions in growth

curve and disease severity while maintaining relatively high levels of antagonism in internal and

external antagonism experiments. Xanthan gum biosynthesis appears to be important in

pathogenicity for X. perforans as described previously in X. c. campestris (21). This gene may

be of interest for further investigation in designing a pathogenicity-attenuated biological control

agent.










Mutant 91-1 18::AxopA incited intermediate levels of disease severity and reduced growth

curve compared to wt 91-1 18. Deletion of xopA affected both internal and external antagonism

compared to wt X perforans, however, the bacteria still maintained relatively high levels of

antagonism. Overall XopA is necessary for full virulence and in planta growth as previously

described in I. euvesicatoria (106), however, disease levels caused by the xopA mutant were still

too high (>25%) to be a viable pathogenicity-attenuated mutant for use as a biocontrol agent.

91-118::AhpaC and 91-118::AavrBs2 were significantly different in overall growth rate,

however, they were the most affected in overall antagonism according to internal antagonism

experiments. Both only reduced E3-1 levels similar to a, non-bacteriocin producing, X

euvesicatoria 91-106 strain and were not significantly different from the other mutants tested or

the 91-106 treatments (Table 3-6).

Two candidates, 91-118::AhpaB and 91-118::AavrBs2, exhibited growth curve and disease

severity with overall similarity to wt 91-118. These candidates could, however, be distinguished

from wt, based on antagonism in internal antagonism assays. Although these genes were

previously shown to be involved in virulence in I. euvesicatoria (18, 17, 53, 63, 76, 146), they

do not appear to affect disease severity sufficiently to be feasible for creating a pathogenicity-

attenuated biological control agent. Furthermore, these mutants dramatically reduced bacteriocin

antagonism; however, further research is needed to determine if this reduction is due to overall

fitness of the bacteria or partial recognition and suppression of the bacterium by the plant as

hypothesized earlier.

Two genes, opgH and gumD, exhibited the desired attenuation and bacteriocin activity

when mutated. Although the other mutants did not reduce pathogenicity sufficiently alone, they

may provide a more dramatic effect when in combination with one another. HpaB and HpaC are










chaperones for effectors to the T3SS channel (HpaB and HpaC) (18, 17), and have been shown

to directly interact forming an oligomeric protein complex and interact with two classes of

effectors (class A containing XopJ and XopF1 and class B containing AvrBs3 and XopC) and

HrcV of the T3SS (18). Based on their known interaction in I euvesicatoria, HpaB and HpaC

may provide a greater reduction in disease when both are knocked out within the same bacterium

(18).

Although eight pathogenicity related genes were selected to study, there is an abundant

source of genes involved in pathogenicity that could be exploited to create further pathogenicity-

attenuated mutants. Many xanthomonads and other pathogenic bacteria have been recently

sequenced such as I. euvesicatoria strain 85-10 (150), I. axonopod'is py. citri strain 306 (27),

XanthomonasXX~~XX~~~XX~~XX campestris py. campestris (27), X. oryzae py. oryzae (85), P. syringae py. tomato

(15) and R. solanacearum (133). These sequences will provide essential information for

understanding how bacteria develop a pathogenic relationship with the host. For instance,

Thieme et al. has estimated that there are ~480 putative pathogenicity factors in I. euvesicatoria

strain 85-10 (150). This represents a large pool of putative pathogenicity genes, and therefore,

opportunity to utilize these genes to create pathogenicity-attenuated mutants. Although there are

a number of reports about mutants with reduced pathogenicity in I. euvesicatoria (17, 18, 54, 76,

105, 106, 107, 130, 146, 171), there are only a few reports of attenuated phenotypes in X.

perforans (101). This information was utilized to create a biological control strategy to allow a

bacteriocin-producing X. perforans strain to effectively colonize the plant and deliver

bacteriocins while causing little to no disease.

Other secretion systems may prove to be important in pathogenicity as well. I.

euvesicatoria has many substrates delivered via the type II secretion system such as cellulases,










P-glucosidases, pectate lyases, polygalacturonases and xylanases are proposed to exhibit plant

cell wall-degrading activity (150). Deletion of the type II secretion system may be too

detrimental to the bacterium; however, deletion of a number of the delivered cell wall-degrading

enzymes (eg. polygalacturnate lyase, ot-amylase, or endoglucconase) may provide a desirable

pathogenicity-attenuated phenotype. A putative type IV secretion system in I. euvesicatoria has

been shown to have homology to the Icm/Dot system of human pathogens (20). The essential

role of Icm/Dot type IV secretion system of Legionella species may suggest a possible role in

virulence in Xanthomona~s as well (20, 150). The type IV pilus is also thought to be involved in

movement by retraction to mediate adhesion to plant tissue (107).

Inactivation or over expression of the quorum sensing auto-inducers may also provide an

alteration in pathogenicity. Several genes encoding this system that are found in I. euvesicatoria

(rpfA to H) (150). Diffusible signal factors (DSFs) have been shown to be involved in regulation

of the synthesis of extracellular enzymes, exopolysaccharides and cyclic glucans (98, 163).

The genes mentioned here along with many others may provide optimal attenuation for our

system. Another possibility may be to create deletions in multiple pathogenicity related genes

(as discussed with hpaB and hpaC) to determine their combined mutant phenotypes. This may

be useful as a tweaking tool to create the optimal level of colonization and infection of the host

plant. One hurdle to overcome concerning characterization of these genes is functional

redundancy. Mutations in most effector genes do not show significant effects on bacterial

virulence when deleted, presumably because of functional redundancy of some effectors (9, 24,

3 1, 78). Overall, there is a great deal of potential for utilization of these pathogenicity related

genes for creation of a pathogenicity-attenuated biological control agent.









Table 3-1. Bacterial strains and plasmids used in this study
Strain or plasmid Relevant characteristics Source or reference*
Xanthomonas euvesicatoria
E3-1 NalRSmR (61)
91-106 (155)
ME-90 RifRKanR (155)
X. perforans
91-118 RifR (155)
91-118::ABopgH OpgH- RifR (101)
91-118::AavrBs2 AvrBs2- RifRCmR This study
91-118::AxopA XopA- RifR This study
91-118::AxopD XopD- RifRCmR This study
91-118::AhpaA HpaA- RifRCmR This study
91-118::AhpaB HpaB- RifRCmR This study
91-118::AhpaC HpaC- RifRCmR This study
91-118::AgumD GumD~ RifR This study
Escherichia coli
DH500 F- rec A BRL
C2110 NalR BRL
hPIR Host for pOKl; SpR OriR6K K2 replicon UB
Plasmids
pBluescript-KS+ Phagemid, pUC derivative; AmpR Stratagene
pLAFR3 TcR rlx+ RK2 replicon BJS
pRK2013 KmR tr+ mob+ (28)
pOK1 Suicide vector; SacB (63)
pOKl:AopgH OpgH~ SmR This study
pOKl:AavrBs2 AvrBs2~ SmRCmR This study
pOKl:AxopA XopA~ SmR This study
pOKl:AxopD XopD~ SmRCmR This study
pOKl:AhpaA HpaA~ SmRCmR This study
pOKl:AhpaB HpaB~ SmRCmR This study
pOKl:AhpaC HpaC~ SmRCmR This study
pOKl:AgumD GumD- SmR This study
* BRL, Bethesda Research Laboratories, Gaithersburg, MD; Stratagene, Stratagene Inc., La
Jolla, CA; BJS, B. J. Staskawicz, University of California, Berkeley, CA; UB, U. Bonas, Martin-
Luther-Universitait, Halle, Germany.











Table 3-2. PCR primers used in hpaA, hpaB, hpaC, xopA, xopD, avrBs2 and gumD analyses for
genetic manipulations
restriction GC Tm
Gene primer name Primer sequence length
site (94) (C)
hpaA hpaAF HanDIII AAGCTTGC:TCAAGCTGGTGGTG 22 54 556
hpa/JI HanDIII AAGCTTATCTAATCGTGCGCCTGC 24 50.0
hpaA2F ACGCAAACGAGCAGGAAG 18 55.6 5.
hpaA2R AGCAGGATCAGTGGAAGCAG 20 55.0
hpa/J3F S'mcd CCCGGGGTTTGGCTTCGATCTCC~ICTTCCTGCT 31 61.:354
hpaADR Small CCCGGGATCTCCTGCTTCCACTGATCCTGCT 31 61.3
hpaB hpalBF HanDIII GGGAAGCTTGCGACGCTGCGCGACA 26 65.4 60.0
hpaBR HanDIII GGGGAATA AGGCTGCCATGGAGGAGGG 29 62.1
hpaB2F CAGAAGTCCATGAACAACAAGATCACG 27 44.4 5
hpaB2R ATCTCCCGCCAAACCTGTATCG 22 54.5
hpaC hpaCF GGCATCGAGATCGCCCAG 18 66.7 5.
hpaCR CGCATACCGCAACCGCAG 18 66.7
hpaC2F GATAGCCAGCCACGCTTCCC 20 61.5 5.
hpaC2R CGCACAGCTCG;CGCTTCC 18 61.7
hpaCDF HanDIID GGGAAG CT~TGACAGAT~TGCGACCGAGTGGATAC 33 54 5 618:
hpaCDIT HinDIID GGGAAGCTTGATACGTAAGGGTGGGTCGGTTTG 33 54 5 67
xopA XopAF HinDIII GGGAAG;CTT.TGC TGGAAGAGGAAAAG;CG 28 53.6 52&
XopAR HanDIll GGGGATCATCCGCGCGTGCGA 24 62.5 60
XopA2F GAGAGGCTGAGGCTAGT 17 56.0 5.
XopA2R TCATTGAATACGTCGCACC 22 57.0
XopA3F AAGTGGATAACGGCAGTGAG 23 56.0 5.
XopA3R CGGAAAGCGACACAGCAG 18 57.2
xopD xopDF TGCTGCCTTTTTGATGGAC 19 47.4 52
xopDR TCCTGCCAACCCTACTTTAC 20 50.0
xopD2F TCCAAAAAGCAAGCCCAC 18 52.6 5.
xopD2R GACGAGCAAT GACCAATGAG 20 55.4
xopD3F GAGCCAACTTCAGAATGCG 19 55.2 5.
xopD2R GACGAGCAAT GACCAATGAG 20 55.4
xopDE)F HanDIID CCCAAGCTTCTGAAATCACTGCTTCACCCAGAC 33 51.5 598:
xoplf)R HinDIII CCCAAGCTTGGTTCTTCCTATTCGTCCCTGTTC 33 51 5 66
avrBs2D avrBs2F ATCGCCCGCATCGCCTTC 18 59.5 6.
avrBs2R CACGCAGTCGCCTCCACC 18 61.7
avrBs2D)D2F S'al GTCGACCTCGTAGGCATGATCGATGGAC 28 57.1 53&:
avrBs2DD2R S'al GTCGACGGAAACTACGTCAAGACCGACC 28 57.1 59
gumD gunD)F TCGTTCCTCTTCGTCGCAGC 20 60 6
gunD)R TCCCGTATGTTTCGGGCTCCT 20 60










Table 3-3. List of attenuation candidate genes and their published, corresponding mutant
phenotype
Candidate Organism Phenotype Reference
Genes
moderate reduction in in plant Minsavage et al.
opgH,,, X. perforans
growth. 2003
moderate reduction in in plant
hpaA X. euvesicatoria Huguet et al. 1998
growth.
moderate reduction in in plant
hpaB X. euvesicatoria Buittner et al. 2004
growth,
significant reduction in in plant
hpaC X. euvesicatoria Biittner et al. 2005
growth.
moderate reduction in in plant Noe~l et al. 2001 &
xopA X. euvesicatoria
growth. Noe~l et al. 2002
no reduction in in plant
xopD X. euvesicatoria Noe~l et al. 2002
growth.
avrBs2 X. euvesicatoria Significant reduction in in plant Kearney and
growth. Staskawicz 1990
delayed disease symptom
gumDn X: c. camlpestris dvlpe Chou etal. 1997










Table 3-4. Homology of X perforans genes chosen to be deleted to create pathogenicity-
attenuated mutants to genes of other closely related xanthomonads
% Homology (AA)
Xanthomonads HpaA HpaB HpaC AvrBs2 XopA XopD
X. euvesicatoria (85-10)" 98 100 99 98 100 85
X. oryzae. py. oryzicola 88 94 88 88
X. oryzae py. oryzae (KACC10331) 87 95 89 89
X. axonopodis py. citri (306) 75 93 86 95
X. campestris py. campestris (ATCC 33913) 75 85 55 76 74
X. campestris py. glycines 75 93 87 47


a Strains are designated in parenthesis









Table 3-5. hz planta growth and aggressiveness of X. perforans strain 91 118 mutants as
measured by area under the population progress curve (AUPPC) and percent disease
severity, respectively, following inoculation of Bonny Best tomato plants
Growth curve Disease severity
Strain
[AUPPC]a [% disease severitylb
X. perforans (91-118) 118.1 ao 38.7 a
91-118::AopgH 98.5 c 12.3 c
91-118::AhxopA 105.5 b 25.7 b
91-118::AhpaB 116.4 a 36.3 a
91-118::AhpaC 104.4 bc 33.7 a
91-118::AavrBs2 113.7 a 38.0 a
91-118::AgumD 106.4 b 13.7 c
a AUPPC from the growth curve of internal wt and mutant strains of X perforans inoculated at
5 x 106 CFU/mL over a 120 h period.
b Percent disease severity 14 days after dip inoculation of each bacterium at 5 x 106 CFU/mL
amended with 0.025% Silwet L-77.
" Values followed by the same letter are not significantly different based on Waller-Duncan
multiple range test (P = 0.05).









Table 3 -6. Growth room in planta internal and phyllosphere antagonism experiments measuring
I. euvesicatoria strain E3-1 populations when co-inoculated with water, I.
euvesicatoria strain E3- 1 or wt and mutants of X. perforans strain 91 118 measured
as area under the population progress curve (AUPPC)
Internal antagonism Phyllosphere antagonism
Strain
[AUPPC]a [AUPPC]
water control 115.3 ab
X euvesicatoria (91-106) 101.7 b 120.8 a
X perforans (91-118) 67.0 d 69.0 c
91-118::AopgH 86.7 c 79.1 b
91-118::AropA 87.2 c 84.7 b
91-118::AhpaB 89.2 c
91-118::AhpaC 94.6 bc
91-118::AavrBs2 95.7 bc
91-118::AgumD 86.7 c
a AUPPC from antagonism assay over a 96 h period based on recovered populations ofX.
euvesicatoria strain E3-1.
b Values followed by the same letter are not significantly different based on Waller-Duncan
multiple range test (P = 0.05).











A. salmel


(pTOPO:avrBs2DF ) SaHISaU Spd~ SaHI SaH Spd

SaHI Spd p TOPO: avrBs2DF p TOPO:AavrBs2 +
+ avrBs2DR CMR

pTOPO: avrBs2DR



B. HinDIII HinDIII HinDIII Sally Sally HinDIII Sally Sally


pGEMI:hpaA pGEMI:AhpaA pGEMI:AhpaA + CMIR



Figure 3-1. Illustration of deletion constructions. A. Deletion strategy for avrBs2. B. Deletion
strategy for hpaA, hpaC, xopA and xopD.













7

6




LL4


o3



0 24 48 72 96 120
Time after inoculation (h)

Figure 3-2. In planta growth of wild-type (wt) and mutant X perforans strains. Plants were
inoculated at 5 x 105 CFU/mL of 91-118::AopgH (m), 91-118::AhpaB (0),
91-118::AhpaC (0), 91-118::AhxopA (x), 91-118::AgumD (A), 91-118::AavrBs2 (A),
wild-type 91-118 (*) in tomato genotype Bonny Best. Error bars indicate the
standard error.












































Figure 3-3. Disease severity on Bonny Best leaflets 2 weeks after dip inoculation (5 x 106
CFU/mL + 0.025% Silwet L-77) with X. perforans strains. Top: wt 91-118 (left),
91-118::AhxopA (left center) and 91-118::ABopgH, (right center) and 91-118::ABhpaB
(right). Bottom: 91-118::AhpaC (left), 91-118::AavrBs2 (center) and 91-118::AgumD
(right) .
























Tomato Pepper
[VF36 + Bs2] [ECW-20R]

strain 91 -1 18 91 -1 18::AavrBs2 91 -1 18 91 -1 18::AavrBs2

HR HR -HR



Figure 3-4. Phenotype in leaves of Bs2 transgenic tomato VS36 and pepper (ECW-20R)

inoculated with 5 x 109 CFU/mL of X. perforans strains 91-1 18 (left) and

91-118:AavrBs2 (right). Phenotypes were recorded 24 h after inoculation. Browning

of the tissue is associated with a hypersensitive response (resistance).


1.:;*
''

c"~;'~:
-;...
---


li
L
x


















"56










0 24 48 72 96
Time after inoculation (h)

Figure 3-5. Growth room internal antagonism assay measuring I. euvesicatoria strain E3-1 in
leaflets. Plants were infiltrated with 5 x 107 CFU/mL of 91-1 18::AopgH (m),
91-118::AhpaB (0), 91-118::AhpaC (0), 91-118::AxopA (x), 91-118::AgumD (A),
91-118::AavrBs2 (A), wild-type 91-118 (*), I. euvesicatoria strain 91-106 (0) and
water (+), followed 18 h later by 5 x 106 CFU/mL of E3-1 in tomato genotype Bonny
Best. Error bars indicate the standard error.














8q



6








0 24 48 72 96
Time after E3-1 inoculation (h)

Figure 3-6. Growth room phyllosphere antagonism assay measuring I. euvesicatoria strain E3-1
in leaflets. Plants were dip inoculated with suspensions of 5 x 107 CFU/mL
(amended with 0.025% Silwet L-77) of: 91-118::AopgH (m), 91-118::AxopA (x),
wild-type 91-118 (*) and water (+), followed 7 d later by spray inoculation of 5 x 107
CFU/mL of E3-1 on tomato genotype Bonny Best. Error bars indicate the standard
error.









CHAPTER 4
EVALUATION OF XANTHOM~ONASPERFORANS MUTANTS IN CONTROLLING X.
EUVESICA TORIA IN GREENHOUSE AND IN THE FIELD

Bacterial spot of tomato is incited by four Xanthomona~s species; X. euvesicatoria, X.

vesicatoria, X. perforans and X. gardneri. The first three bacterial species were previously

known as tomato races T1, T2 and T3, respectively, based on their reaction on three tomato

genotypes: Hawaii 7998 (H7998), Hawaii 7981 (H7981) and Bonny Best (71, 72, 141). X.

gardneri has only been found in Yugoslavia, Costa Rica and Brazil (10, 122, 144).

Control of bacterial spot of tomato is difficult when high temperatures and high moisture

conditions exist. The disease has been demonstrated to cause significant damage to the crop

resulting in major losses. Pohronezny and Volin (118) estimated as high as 50% loss of

marketable fruit due to bacterial spot on tomatoes. There are currently no commercially

available tomato varieties resistant to bacterial spot. Scott and Jones (135) identified significant

resistance in H7998 in which I. euvesicatoria strains induce an HR. In 1993 Whalen et al. (166)

determined that I. euvesicatoria strains carry the avirulence gene avrRxy, which induces an

hypersensitive response (HR) on the genotype H7998 carrying the corresponding resistance gene

Rxy; however, X. perforans T3 strains carry avrXv3 which induces an HR in H7981 containing

the resistance gene Xv3 (100). Astua-Monge (4) characterized avrXv3 and found it to elicit an

HR in some tomato and pepper varieties. In 2000 a new avirulence gene avrXv4 was described

in X. perforans strains based on an HR in tomato genotype LA716 (Lycopersicum pinnellii)

carrying the Xv4 resistance gene (3, 4). I. perforans strains carrying this new avirulence gene

(avrXv4) and a non-functional AvrXv3 are designated as tomato race 4.

Bactericides, such as fixed coppers and streptomycin, have provided the primary means of

chemical control (90, 142, 143); however, streptomycin-resi stant mutants and copper-tolerant

strains became prevalent (143). Marco and Stall (90) reported widespread emergence of copper-









tolerant I. euvesicatoria strains and that addition of the fungicide mancozeb to copper sprays

improved disease control caused by copper-tolerant strains (25, 90). Chemical control alone is

insufficient to control the disease under optimal conditions for the pathogen. Additionally, the

use of copper compounds led to soil contamination in some instances (79).

Recently, there has been increased interest in integrated biological control strategies for

bacterial diseases, which are difficult to control with conventional management practices. Some

success has been achieved in this area through empirical selection of biocontrol agents, as

indicated by the commercialization of the products AgriphageTAI, a mixture of bacteriophages for

control of bacterial spot of tomato (46), GalltrolTM for control of crown gall, and BlightBanTM

A506 for control of fire blight and frost injury (89). However, achieving success using

biocontrol agents for many bacterial diseases has been difficult. This failure may in part be due

to the very narrow focus on the almost exclusive use of nonpathogenic, saprophytic bacteria as

biocontrol agents. While our understanding of the ecology of nonpathogenic saprophytes is

increasing, our knowledge is limited to labor-intensive protocols for identifying potential

biocontrol agents. New integrated biological control strategies are currently being sought

including the use of bacteriocins, attenuated plant pathogens and/or bacteriophages (28, 32, 46,

61, 67, 102, 108, 168) as part of an integrated biological control strategy.

One recent approach for biological control has been the use of bacteriocins (77, 152).

Bacteriocins are substances produced by bacteria that are inhibitory or harmful toward only

closely related bacteria (125). Bacteriocins and bacteriocin-like compounds encompass an array

of structurally different substances including enzyme inhibition, nuclease activity and pore

formation in cell membranes (125, 129, 128). Bacteriocins produced by Escherichia coli and

several Gram-positive bacterial species have been extensively characterized (65, 125, 147). For









bacterial spot of tomato at least three distinct bacteriocin-like activities (BcnA, BcnB and BcnC)

were identified in X. perforans that are antagonistic toward I. euvesicatoria strains (154). Hert

et al. previously (61) demonstrated that two of these bacteriocins previously discovered in X.

perforans (BcnA and BcnC) (154) could effectively control X. euvesicatoria populations in

greenhouse and field experiments.

Although early attempts for biocontrol of plant diseases using bacteriocin-producing

strains were made, few have been implemented (158, 159). For bacterial spot of tomato, Liu

(88) conducted biological control studies utilizing a non-pathogenic bacteriocin-producing X.

perforans strain to control disease incited by I. euvesicatoria strains. The non-pathogenic strain

reduced bacterial spot disease incidence and severity by 10 to 15 percent in the field when

applied prophylactically when compared to I. euvesicatoria alone; however, these levels were

still unacceptable levels of control (40% disease) (88).

For Ralstonia solan2acearum, efforts to obtain a biological control strategy utilizing

bacteriocin-producing non-pathogenic hrp- mutants gave low to moderate levels of control of

wild-type (wt) R. solan2acearum (152). However, control using a partially pathogenic hrp mutant

(hrc V), which is capable of higher levels of colonization of the root and stem tissue, achieved

better control (49). Research into colonization has been conducted to understand the possible

relationship between invasion efficiency of the biocontrol agent and its ability for disease

control. Etchebar et al. (41) suggested that there was a positive correlation between colonization

of the xylem by the hrp mutant and the level of control of the wt R. solan2acearum.

Based on previous studies in which non-pathogenic strains provided low levels of

biological control (41, 49, 88), it was hypothesized that moderate invasion by the biological

control agent using a partially pathogenic bacteriocin-producing strain of X perforans rather









than a non-pathogenic strain may increase the efficiency of control under field conditions. These

mutants may colonize the leaf tissue internally more effectively than non-pathogenic strains and

this could potentially result in more effective biological control. A known X. perforans mutant

gene opgHxcy was selected to create the pathogenicity-attenuated mutant (101). OpgH mutants

have a reduced disease severity and growth curve in susceptible tissue and delayed avirulent HR

phenotype in resistant plant tissue.

The obj ective of this study was to evaluate the ability of a pathogenically-attenuated

bacteriocin-producing strain (91-1 18::AopgH~bcnB) to reduce the populations and of the disease

caused by bacteriocin-sensitive strains of I euvesicatoria. Previous research showed (61) that

deletion of BcnB produces lower recovery of sensitive X. euvesicatoria strains than wt X.

perforans when co-inoculated with I. euvesicatoria.

Materials and Methods

Bacterial Strains, Plasmids and Culture Conditions

Strains of X perforans and I. euvesicatoria were grown on nutrient agar (NA) medium

(Difco Laboratories, Detroit, MI) at 280C (Table 4-1). Strains ofE. coli were grown on Luria-

Bertani (LB) medium at 370C (97). All strains were stored in 20% glycerol in sterile tap water at

-800C. Bacterial cultures for plant inoculations were grown in nutrient broth (NB) (Difco

Laboratories, Detroit, MI) for 18 h at 280C with shaking (100 rpm). Cells were pelleted by

centrifugation (4,000 x g, 15 min) and resuspended in sterile tap water. Bacterial suspensions

were standardized to an optical density at 600 nm (OD600) = 0.3 (5 x 108 CFU/mL) with a

Spectronic 20 spectrophotometer (Spectronic UNICAM, Rochester, NY) and subsequently

diluted in sterile tap water to appropriate cell densities for individual experiments. Antibiotics

were used to maintain selection for resistance markers at the following concentrations:









tetracycline (Tc) 12.5 Clg/mL; rifampicin (Rif) 100 Clg/mL; spectinomycin (Sp) 50 Clg/mL;

kanamycin (Km) 50 Clg/mL; chloramphenicol (Cm) 34 Clg/mL; streptomycin (Sm) 200 Clg/mL;

and nalidixic acid (Nal) 50 Clg/mL.

Generation of the 91-118:::AopgHb~bcn Attenuation Mutant

Triparental matings were performed using E. coli DH5a containing pRK2013Kn' as the

helper plasmid (Table 4-1), E. coli DH5a containing pXV442-255 (pXV442Tc with insertion of a

KmR cassette for inactivation of BcnB) as the donor and 91-1 18::AopgH as the recipient

(Recipient received from Gerald Minsavage). Marker exchange was achieved using standard

methods (134). The candidate colonies were screened for loss of BCN activity and confirmed for

insertion by Southern hybridization (using subclone BcnB as the probe) and PCR (with primers

BCN-1 and BCN-2) (61).

Plant Materials

Seeds of tomato (Lycopersicon esculentum) cy. Bonny Best were planted in Plugmix (W.

R. Grace & Co., Cambridge, MA). After 2 weeks, the emerged seedlings were transferred to

Metromix 300 (W. R. Grace & Co., Cambridge, MA) in 10-cm plastic pots. Seedlings were

grown in the greenhouse at temperatures ranging from 25 to 350C.

Growth Room Growth Curve Assays

Growth room assays were conducted to compare the growth curves of 91-1 18::AopgH and

91-118::AopgH~bcnB mutants with the wt parent strain 91-118. Strains were grown in NB for

18 h, harvested by centrifugation and resuspended in sterile tap water. Strains were inoculated at

3 x 105 CFU/mL into leaflets of 6-week-old tomato seedlings. Leaflets were infiltrated (15

leaflets per strain) using a hypodermic syringe and needle, as described previously (68).

Following inoculation plants were kept at 240C to 280C. Three samples were taken for each









treatment every 24 h for 5 d. Bacterial populations were quantified by macerating 1-cm2 la

disks in 1 mL sterile tap water and dilution plating onto NA medium amended with the

appropriate antibiotic. Plates were incubated at 280C and colonies were counted after 48 to 72 h.

Population data were loglo transformed and standard errors were determined. The overall growth

curve was determined by calculating the area under the population progress curve (AUPPC).

The AUPPC is a modification of the area under the disease progress curve (AUDPC) which has

been used to analyze disease progress (136): standardized AUPPC = E [(xi + xi- 1)/2](ti ti- 1),

where x is population density in loglo CFU per cm2 and t is time in hours. The AUPPC values

for the strains were compared by analysis of variance and subsequent separation of sample

means by Waller-Duncan multiple range test using SAS version 9.0 (SAS Inc., Cary, NC). Each

experiment was conducted three times.

Greenhouse Disease Severity Assay

Greenhouse disease severity assays were conducted to determine the effect ofX. perforans

mutant and wt 91-118 strains on symptom development. In each test, four young (four-true-leaf

stage) plants were inoculated with each strain by dipping into sterile tap water containing

bacterial suspensions of 91-1 18, 91-1 18::AbcnB, 91-1 18::AopgH, 91-1 18::AopgH~bcnB and I.

euvesicatoria strain 91-106 (3 x 106 CFU/mL of bacteria amended with 0.025% Silwet L-77

(Loveland Industries, Inc., Greeley, CO)) for 15 s. Plants were maintained in the greenhouse

during the evaluation period. The plants were assessed for disease severity 14 to 21 days after

inoculation. Disease assessments for wt and attenuated mutant candidates ofX. perforans strains

were made based on leaf and stem ratings compiled from three separate greenhouse inoculation

tests.









Growth Room Antagonism Assay

Antagonism assays were performed to determine the effect of wt and mutant X perforans

strains on the bacteriocin sensitive X. euvesicatoria strain, E3-1. Internal and external

(phyllosphere) populations were separately evaluated using two antagonism assay techniques.

Internal antagonism a;ssssssssssssssays. Strains were grown in NB for 18 h, harvested by

centrifugation and resuspended in sterile tap water. I. perforans and I. euvesicatoria strains

were inoculated at 5 x 107 CFU/mL and 5 x 106 CFU/mL, respectively. Six-week-old seedlings

of the tomato cultigen Florida 47 were inoculated (15 leaflets per strain) using a hypodermic

syringe as described previously (68). The mutant and wt X perforans strains were inoculated

into leaflets by infiltration 12 h prior to inoculation with the sensitive strain (E3-1). Each

treatment consisted of three replications. Following inoculation, plants were incubated at 240C

to 280C. In order to determine populations of the sensitive strain (E3-1smNal) in leaflets, 1-cm2

leaf disks were removed from inoculated areas, macerated in 1 mL sterile tap water and dilution

plated onto nutrient agar amended with the appropriate antibiotic. Samples were assayed at 24 h

intervals for 96 h. Each experiment was conducted three times. Population data were log

transformed and standard errors were determined. AUPPC values (calculated as described

above) were compared by analysis of variance and subsequent separation of sample means by

Waller-Duncan multiple range test using SAS version 9.0 (SAS Inc., Cary, NC). Each

experiment was conducted three times.

Phyllosphere antagonism a;ssayss~~~~ssss~. Growth room phyllosphere antagonism assays were

conducted to determine if the gene deletions affected the levels of antagonism toward external

leaf populations of I. euvesicatoria strain E3- 1 by comparing the antagonistic ability of the

mutants with the parent strain 91-118. Strains were grown in NB for 18 h, harvested by









centrifugation and resuspended in sterile tap water. Six-week-old Bonny Best tomato seedlings

were dipped into 5 x 107 CFU/mL suspension of the wt and mutant strains of X perforans

amended with Silwet L-77 (0.025%) 7 days prior to spray inoculation with a 5 x 107 CFU/mL

suspension of I euvesicatoria strain E3-1. Following spray inoculation, plants were incubated

at 240C to 280C. Leaf tissue was sampled every 24 h for 96 h for quantification of E3-1

populations. Three leaflets were taken at each time point. Each leaflet was weighed, placed in a

polyethylene bag containing 10 mL of sterile tap water shaken on a Wrist Action shaker (Burrel

Co., Oakland, CA) and shaken vigorously for 20 min. The leaf-wash was then dilution plated on

NANaism to selectively determine the bacterial population of E3-1. Population data were

analyzed following loglo transformation and standard errors were determined. AUPPC values

(calculated as described above) were compared by analysis of variance and subsequent

separation of sample means by Waller-Duncan multiple range test using SAS version 9.0 (SAS

Inc., Cary, NC). Each experiment was conducted three times.

Field Experiments

Fieldplot design. The field experiments were set up in a completely randomized block

design consisting of four replications. Raised beds were 0.91 m wide and were covered with

black plastic mulch. Plots were arranged in paired beds that were 1.83 m from center to center

and each set of paired beds was 7.32 meters apart. Plots within the paired beds were spaced 6. 1

m apart. Each plot containing 20 plants were spaced 457 cm apart.

Bacterial strains, inoculum production, inoculation and plant material. Field experiments

were performed using X. perforans mutant 91-118::AopgH~bcnB and I. euvesicatoria strain E3-

1 to evaluate antagonism of those strains to the X. euvesicatoria strain E3-1 (Table 4-1). Strains

were grown in NB for 24 h, harvested by centrifugation and resuspended in sterile tap water.









The bacterial suspensions were adjusted to 5 x 107 CFU/mL and the surfactant Silwet L-77 was

added to a final concentration of 0.025%. Plants were dipped into suspensions of the

91-118::AopgH~bcnB strain 24 h prior to spray inoculation with a 5 x 10' CFU/mL suspension

of E3-1. Six-week-old seedlings of the tomato genotype Florida 47 (Asgrow, Oxnard, CA) were

used in all experiments.

Incidence ofstrains in lesions. In 2004, field experiments were conducted at the North

Florida Research and Education Center (NFREC) in Quincy, FL to evaluate recovery of wt I.

euvesicatoria strain E3-1 and 91-1 18::AopgH~bcnB from symptomatic leaf tissue. The

experiment consisted of six treatments: (1) uninoculated control; (2) E3-1 + growers standard;

(3) E3-1 alone; (4) 91-118::AopgH~bcnB alone; (5) E3-1 and 91-118::AopgH~bcnB (applied bi-

weekly); and (6) E3-1 and 91-118::AopgH~bcnB (applied weekly). Plants in the grower

standard was treated on weekly rotations of acibenzolar-S-methyl (0.055 g/L) (Actigard 50WG;

Syngenta Crop Protection Inc., Greensborough, NC) or copper hydroxide (3.6 g/L) (Kocide

2000; Griffin Corp., Valdosta, GA) plus mancozeb (2.5 g/L) (Manzate 75DF; Griffin Corp.,

Valdosta, GA) every two weeks. Symptomatic leaf tissue was collected every 2 weeks

beginning 35 days after transplanting. Ten to twenty leaflets were randomly collected in each

plot and bacteria were isolated from thirty lesions. Individual lesions were macerated in 75 CIL

of sterile deionized water and the suspensions were streaked on NA amended with 134 Clg/mL of

pentachloronitrobenzene (PCNB) (126) and 50 Clg/mL of cycloheximide to eliminate fungal

contaminants from the samples. Individual colonies were plated onto two media to differentiate

X. perforans and I. euvesicatoria (NA amended with the appropriate antibiotics for

91 1 1 8::opgH~bcnBRif Km and E3 1"l sm). The overall strain incidence was expressed by

calculating the area under the incidence progress curve (AUIPC). The AUIPC is a modification










of AUDPC: standardized AUIPC = E [(xi + xi-~ 1)/2](ti ti-~1), where x is the arcsin of the percent

recovery (to normalize the data) and t is days after inoculation. The AUIPC values were

compared by analysis of variance and subsequent separation of sample means by Waller-Duncan

multiple range test using SAS version 9.0 (SAS Inc., Cary, NC).

Incidence ofphyllosphere populations. In 2005, field experiments were conducted at two

locations (NFREC and the Citra Research Farm) to evaluate recovery of E3-1 and

91-1 18::AopgH~bcnB from the surface of asymptomatic leaf tissue. Asymptomatic leaf tissue

was sampled every 2 weeks beginning ~20 days after transplanting (DAT). Seven leaflets were

collected from each plot. Each sample was weighed, placed into a polyethylene bag (Becton

Dickinson, Rutherford, New Jersey) containing 5 to 10 mL of sterile tap water and shaken at 200

rpm for 30 to 45 minutes. Serial ten-fold dilutions were made in sterile tap water. A 50 Cll

aliquot of each dilution was plated two NA plates, one amended with 134 Clg/mL of PCNB

and/or 50 Clg/mL of cycloheximide with addition of antibiotics for selection of E3-1 (Sm and

Nal) and the second for the 91-1 18::AopgH~bcnB mutant (Rif and Km). After incubation at

280C for 4-5 days, colonies typical ofXanthomona~s were counted and populations were

calculated. Data was analyzed for statistical significance by using the AUIPC. AUIPC values

(calculated as described above) were compared by analysis of variance and subsequent

separation of sample means by Waller-Duncan multiple range test using SAS version 9.0 (SAS

Inc., Cary, NC). Each experiment was conducted three times.

Results

91-118 and Mutants Reduce E3-1 in Growth Room and in the Greenhouse

Growth room experiments were conducted to determine the effects of the mutations to

grow in planta. The 91-1 18::AopgH~bcnB mutant reached populations 1 to 1.5 log units lower









than wt 91-118 (Figure 4-1). The 91-118 population exhibited a normal growth curve during the

120-h sampling period and, based on the AUPPC results (Table 4-2), and had significantly

higher populations over time than both 91-118::AopgH mutants. The 91-118::AopgH~bcnB

mutant was not significantly different from 91-118::AopgH suggesting that there was no

pronounced effect on growth curve based on the AbcnB deletion.

Greenhouse disease severity experiments were conducted to determine the effects of the

mutations on the ability of 91-1 18 to cause disease in planta (Figure 4-2). Disease severity

incited by 91-118::AopgH~bcnB and 91-118::AopgH strains (12% and 14 %, respectively) was

significantly lower than that of 91-1 18 (39 %) (Table 4-2).

Both internal (Figure 4-3) and phyllosphere (Figure 4-4) antagonism assays under growth

room conditions were performed to determine the antagonistic activity of 91-118 and

91-118::AopgH~bcnB and 91-118::AopgH strains toward the E3-1 strain. For both assays

mutants 91-118::AopgH and 91-118::AopgH~bcnB were moderately antagonistic, whereas wt

91-118 and 91-118::AbcnB provided the greatest reduction in E3-1 populations. The water

control treatment consistently had significantly higher population levels than all other treatments

(Figures 4-3 and 4-4 and Table 4-3). Treatment of a wt I. euvesicatoria strain (91-106) prior to

I. euvesicatoria strain (E3-1Nalsm) reduced populations by ~0.5 loglo CFU/mL in both assays

compared to the water control. Differences in the levels of antagonism was observed between

internal and phyllosphere antagonism assays.

Field Study

The ability of the attenuated mutant, 91-1 18::AopgH~bcnB, to reduce E3-1 populations

was assessed in the Hield. Controls included non-inoculated control, a E3-1 alone control and

91-118 alone control plot. In 2004, symptomatic leaf tissue was sampled at the NFREC in









Quincy, FL and I. euvesicatoria populations were significantly reduced by both

91-118::AopgH~bcnB control treatments (weekly and two week application) (Figure 4-5). E3-1

was recovered from less than 5 percent of the samples in 91-1 18::AopgH~bcnB plots compared

to 26 percent from E3-1 alone (Table 4-4).

In 2005 at Quincy (Figure 4-6) and Citra (Figure 4-7) phyllosphere populations were

sampled. In both locations E3-1 populations were significantly reduced by the

91-118::AopgH~bcnB mutant when applied weekly throughout the growing season (Figures 4-6

& 4-7). In Quincy, E3-1 was recovered from 30% of the samples in plots where E3-1 was

applied alone (Table 4-4). In the treatment where 91-118::AopgH~bcnB was applied every two

weeks, the frequency of recovery of E3-1 populations was not significantly different from plots

where E3-1 was applied alone. Weekly application of 91-1 18::AopgH~bcnB, however,

significantly reduced recovery of E3 -1 populations compared to plots where the E3-1 was

applied alone (approximately 65% reduction) (Table 4-4).

In Citra (2005), both 91-118::AopgH~bcnB weekly and biweekly treatments significantly

reduced recovery of E3 -1 populations (Figure 4-7). The AUEPC of 91-1 18::AopgH~bcnB

weekly and biweekly treatments had significantly reduced E3-1 incidence compared to the

grower standard (37%) and E3-1 alone (54%) plots (Table 4-4).

Discussion

In this study, we sought to create a pathogenically attenuated X. perforans mutant to: (I)

express two of the three previously described bacteriocins (based on previous field analysis (61)

and (II) maintain itself at a level to maintain antagonism toward I. euvesicatoria strains while

causing minimal disease. We decided to examine the previously described osmorelgulated

periplasmic glucan gene, opgHxcy (101). In greenhouse experiments the growth curve of both









attenuated mutants 91-118::ABopgH and 91-118::AopgH~bcnB exhibited reduced growth curves

compared to 91-118. The 91-118::AopgH~bcnB mutant also caused significantly less disease

than wt which was similar to a previously study by Minsavage et al. (101). The 91-118::AopgH

still maintained significant levels of antagonism toward E3-1 populations as observed in the

antagonism assay. Overall, mutants 91-1 18::AopgH and 91-118::AopgH~bcnB were

significantly effective in reducing I. euvesicatoria populations in the greenhouse. Interestingly,

91-118::AopgH and 91-118::AopgH~bcnB were more effective in reducing external E3-1

populations than internal populations. Lindemann et al. (86) previously concluded that there is a

direct correlation between phyllosphere populations and occurrence of disease. The existence of

this correlation together with the reduction in I. euvesicatoria phyllosphere populations during

antagonism experiments suggest that a biological control strategy for I. euvesicatoria by

91-118::AopgH~bcnB may be effective at reducing disease by reducing phyllosphere populations

below the threshold level necessary to cause lesion development.

In 2004, two hurricanes during the season introduced high external populations of wt X

perforans, which reduced both E3-1 and 91-118::AopgH~bcnB populations. Although naturally

occurring populations of X perforans were introduced into the plots, early sampling data along

with AUDPC data suggest that weekly and biweekly treatments with 91-118::AopgH~bcnB

significantly reduced I. euvesicatoria populations. Similar trends were also observed in our

2005 field data evaluating phyllosphere levels. In 2005 experiments 91-118::AopgH~bcnB

effectively reduced I. euvesicatoria populations by up to 85 percent. In both years, weekly

application of 91-1 18::AopgH~bcnB at 5 x 106 CFU/mL significantly reduced I. euvesicatoria

populations compared to plots receiving I. euvesicatoria alone. In two of three experiments, bi-

weekly application was found to significantly reduce E3-1 populations. This reduction in X










euvesicatoria populations is similar to previous field data using 91-118::AbcnB (61). Hert et al.

(61) found that co-inoculation of 91-1 18::AbcnB and an I. euvesicatoria in the field yielded less

than 5 percent recovery of the X. euvesicatoria strain over all seasons tested. These results

suggest that the 91-118::AopgH~bcnB attenuated mutant has potential as a biological control

agent of reducing I. euvesicatoria populations.

The reduction in I. euvesicatoria populations by 91-1 18::AopgH~bcnB in the antagonism

assays does appear to be less inhibitory compared to 91-118::AbcnB. This suggests that

bacteriocin expression was also affected by the 91-1 18::AopgH mutation. The repeated

treatment of 91-118::AopgH~bcnB in the field experiments was sufficient for maintaining

91-118::AopgH~bcnB levels and consequently sufficient bacteriocin levels to suppress I.

euvesicatoria populations. Weekly application of 91-1 18::ABopgH~bcnB was more effective in

reducing I. euvesicatoria populations than previously observed in similar experiments using hrp-

X. perforans mutants (88). The 91-118::AopgH~bcnB mutant appears to colonize within the leaf

tissue more effectively than non-pathogenic strains and is similarly to what was observed in a

previous study with R. solan2acearum hrcY mutants (41).

Although the opgHxcy mutant was effective in suppressing I. euvesicatoria populations,

there is potential for identifying other gene targets that can help improve biological control

efficacy. Several other pathogenicity factors and associated genes have previously been

described in I. euvesicatoria with other genes associated with the hrp system (hpaA, hpaB,

hpaC), avirulence genes (avrBs2, xopA, xopD) and pathogenicity factors (gumD) (17, 18, 63,

106, 10O5, 167). In the previ ous chapter (Chapter 3) we te sted the se pathogeni city-associ ated

genes for their phenotype in X. perforans and the potential use for creating further pathogenicity-

attenuated biocontrol agents. When several of the genes were mutated in X. perforans strain









91-118 there was an associated reduction in pathogenicity and growth curve in plan2ta. These

mutants or other pathogenicity associated genes may improve our pathogenicity attenuated

biological control model system by allowing better internalization and subsequent competition

between I. euvesicatoria and X. perforans populations without detrimental effects to the plant.

Sequencing of the X. euvesicatoria genome has provided significant new possibilities for

developing pathogenicity attenuated candidates. In 2005, Thieme et al (150) published the X.

euvesicatoria genome sequence and estimated over 480 putative pathogenicity factors and

associated genes. These genes were grouped into 6 categories: (I) secretion systems, (II)

flagellum, (III) secreted proteins (via type III secretion system), (IV) detoxification, (V) surface

structure and adhesion and (VI) quorum sensing. Genomic sequencing of bacteria provides an

opportunity to exploit these genes for our utilization.

Further research is needed to optimize this system to create a weakly aggressive biological

control agent that is as antagonistic as wild-type. Recent information from related bacteria and

genomic sequences can be used to provide opportunities to improve our understanding of how

pathogenic bacteria colonize and subsequently infect the host. Continued exploration of new

innovative ideas will help us to utilize this knowledge in effective ways.









Table 4-1. Bacterial strains and plasmids used in this study
Strain or plasmid Relevant characteristics Source or reference

Xanthomonas euvesicatoria
E3-1 NalRSmR (61)
91-106 (155)
IVE-90 RifRKanR (155)

X. perforans
91-118 RifR (155)
91-118::ABbcnB BcnB- RifR KmR (61)
91-118::ABopgH OpgH- RifR (101)
91-118::ABopgH~bcnB OpgH- BcnB- RifR KmR This study

Escherichia coli
DH500 F- rec A BRL
C2110 Nale BRL
hPIR Host for pOKl; SpR oriR6K RK2 replicon UB
Plasmids
pOK1 Suicide vector; SacB (63)
pBluescript-KS+ Phagemid, pUC derivative; AmpR Stratagene
pLAFR3 Tcr rlx' RK2 replicon BJS
pRK2013 helper plasmid; Kmr tra' (28)
BRL, Bethesda Research Laboratories, Gaithersburg, 1VD; Stratagene, Stratagene Inc., La Jolla,
CA; BJS, B. J. Staskawicz, University of California, Berkeley, CA; UB, U. Bonas, Martin-
Luther-Universitait, Halle, Germany.









Table 4-2. In planta growth and aggressiveness of X. perforans strain 91 118 mutants as
measured by area under the population progress curve (AUPPC) and percent disease
severity, respectively, following inoculation of Bonny Best tomato plants
Growth curve Disease severity
Strain
[AUPPC]a [% disease severitylb
X. perforans (91-118) 118.3 a 38.7 a
91-118::AopgH 104.2 b 12.3 b
91 -1 18::AopgH ~bcnB 111.6 b 14.1 b
a AUPPC from growth curve of internal X. perforans strains inoculated at 5 x 106 CFU/mL over
a 120 h period.
b Percent disease severity 14 days after dip inoculation of each bacterium at 5 x 106 CFU/mL
amended with 0.025% Silwet L-77.
" Values followed by the same letter are not significantly different based on Waller-Duncan
multiple range test (P = 0.05).









Table 4-3. Growth room in planta internal and phyllosphere antagonism experiments measuring
I. euvesicatoria strain E3-1 populations when co-inoculated with water, I.
euvesicatoria strain E3- 1 or wt and mutants of X. perforans strain 91 118 measured
as area under the population progress curve (AUPPC)
Internal antagonism Phyllosphere antagonism
Strain
[AUPPC]a [AUPPC]
water control 115.3 ab
X euvesicatoria (91-106) 101.7 b 120.8 a

X perforans (91-118) 67.0 d 69.0 c
91-118::AopgH 86.7 c 79.1 b
91-1 18::AopgH ~bcnB 90.8 c 86.3 b
a AUPPC from antagonism assay over a 96 h period based on recovered populations ofX.
euvesicatoria strain E3-1.
b Values followed by the same letter are not significantly different based on Waller-Duncan
multiple range test (P = 0.05).










Table 4-4. Incidence and recovery ofX. euvesicatoria strain E3-1 in the field when treated with
X. perforans mutant strain 91-118::AopgH~bcnB. Incidence and recovery were
measured by area under the incidence progress curve (AUIPC) and percent recovery,
respectively
2004 2005
Citra Citra Quincy
% % %
AUIPCa AUIPC AUIPC
Treatment Recoveryb Recovery Recovery
Uninoculated control 0 co 0 425 b 13 157 b 7
Grower standard 563.8 a 21 1273 a 37 757 ab 15
E3-1 alone 614.5 a 26 1435 a 54 1430 a 30
91 118::AopgHAbcnB alone 0 c 0 446 b 5 92 b 2
91 118::AopgHAbcnB (2)+E3-le 181.0 b 5 263 b 12 890 ab 25
91 118::AopgHAbcnB (1)+E3-1f 158.2 b 4 260 b 8 477 b 11
a AUIPC from fields evaluating E3-1 populations for each season for each treatment.
b The % recovery is the average percent recovery of E3-1 populations for the season for each treatment.
" Values followed by the same letter are not significantly different based on Waller-Duncan multiple range test (P =
0.05).
d Grower standard plots were treated with Copper + Manzate and Actigard biweekly throughout the season.
e The 91 118:: AopgH~bcnB (2) + T1 plots were treated with AopgH~bcnB every two weeks throughout the season.
f The 91 118:: AopgH~bcnB (1) + T1 plots were treated with AopgH~bcnB weekly throughout the season.














a,7


6


5


LL4


03



0 24 48 72 96 120
Time after inoculation (h)


Figure 4-1. D planta growth of wild-type and mutant X perforans strains. Plants were
infiltrated with 5 x 105 CFU/mL of 91-118::AbcnB (0), 91-1 18AopgH (m), 91-
118::AopgH~bcnB (A) and wt 91-118 (*) in tomato genotype Bonny Best. Error
bars indicate the standard error.






























Figure 4-2. Disease severity on Bonny Best leaflets 2 weeks a tr diip inoculation 5 x 106
CFU/mL + 0.025% Silwet L-77) with X. perforans strains wild-type 91-118 (left),
91-118AopgH (center) and 91-118::AopgH~bcnB (right).