ANALYSIS OF XYLELLA FASTIDIOSA TRANSPOSON MUTANTS AND DEVELOPMENT OF AN EN PLANTA XYLELLA FASTIDIOSA PLASMID VECTOR Project Leader: Cooperator:

We have developed an autonomously replicating Xylella fastidiosa (Xf)/E. coli plasmid that efficiently transforms Xf; unfortunately this plasmid was not stably maintained in Xf cells en planta or without antibiotic selection in vitro. Another plasmid, containing an Xf native plasmid, a Kan cassette and cloned in pUC18 was also constructed and shown to be unstable without antibiotic selection. 1000 random Tn5 mutants were again screened for virulence in grapevines growing in the greenhouse. An expectedly high percentage (35%) did not develop typical pierce’s disease (PD) symptoms; the insertion sites of the Tn5 in these apparently avirulent mutants are being sequenced. Approximately 3% of the random mutants were hypervirulent as compared to the wild type parental strain. Insertion sites of Tn5 in these mutants showed 2 were in putative LPS genes and 1 was in a hemagglutinin-like gene. The phenotype of these putative hypervirulent mutants is being confirmed in a second round of grapevine inoculations. INTRODUCTION During the past 4 years one of the objectives of our research on Pierce's disease (PD) has involved the development of transformation and transposon mutagenesis systems for the bacterium that causes PD, Xylella fastidiosa (Xf). We developed a random transposon based mutagenesis system for Xf in 2001 (Guilhabert et. al., 2001). Recently, we developed an E.coli/Xf plasmid shuttle vector based on the plasmid RSF1010 that replicates autonomously in Xf (Guilhabert and Kirkpatrick, 2003), however this plasmid is only stably maintained in Xf cells that are kept under selection using the antibiotic, kanamycin. Therefore, this shuttle vector will be useful for in vitro studies of Xf gene function; however it cannot be used to study the function of Xf genes in the plant host. We continue to evaluate other plasmids that can be stably maintained in Xf cells inoculated into plant hosts. The complete genome sequence of a citrus (Simpson et al., 2000) and a grape (Van Sluys, et al., 2002) strain of Xf have been determined. Analysis of their genomes revealed important information on potential plant pathogenicity and insect transmission genes. However, approximately one-half of the putative ORFs that were identified in Xf encode proteins with no assignable function. In addition, some of the putative gene functions assigned on the basis of sequence homology with other prokaryotes may be incorrect. For these reasons we felt that it was important to develop and assess the pathogenicity of a library of random Tn5 mutants in order to identify any gene that may influence or mediate Xf pathogenicity. Our group, as well as other PD researchers, is also evaluating specific mutants in Xf genes that are speculated, based on homology with other gene sequences in the database, to be involved with pathogenicity. However, screening a random transposon (Tn) library of Xf, a strategy that has led to the identification of important pathogenicity genes in other plant pathogenic bacteria, may identify other novel genes, especially those that regulate the expression of pathogenicity/attachment genes that will be important in the disease process. Using Tn5 mutagenesis, there is a high probability that we can knock out and subsequently identify Xf genes that mediate plant pathogenesis. Proof that a particular gene is indeed mediating pathogenicity and/or insect transmission would be established by re-introducing a cloned wild type gene back into the Xf genome by homologous recombination, or more ideally, introduce the wild type gene back into Xf on the plant stable shuttle vector we propose to develop. OBJECTIVES 1. Develop a Xylella fastidiosa (Xf)/E. coli plasmid shuttle vector that is stable en planta. 2. Screen a library of Xf transposon mutants for Xf mutants with altered pathogenicity, movement or attachment properties. RESULTS AND CONCLUSION Objective 1 We are cloning the RSF1010 origin of replication into pUC18, creating a plasmid harboring a polylinker, a LacZ selection and expressing the pUC18 antibiotic cassette, carbenicillin that was showed to be expressed in an Xf background (Qin and Hartung, 2001). Such vector will allow selection for plasmid maintenance using an antibiotic that is different from the kanamycin resistance gene carried on the transposome that we use to create Tn5 Xf mutants and provide the necessary tool to complement and prove the function of a Tn-tagged Xf gene. Another approach to create an Xf plasmid shuttle vector is to modify native Xf plasmids with a selectable marker. Such a strategy was successful in developing a stable plasmid vector for citrus (CVC) infecting strains of Xf (Qin and Hartung, 2001) however this plasmid (pER10) did not replicate in grape strains of Xf (Guilhabert and Kirkpatrick, 2003). A 1.3 kb size plasmid from a PD Xf strain was cloned in pUC18 and sequenced. Nucleotide and amino acid sequence analysis revealed conserved sequences that are typical of initiator (Rep) proteins involved in rolling-circle type DNA replication as well as a