PLASMID ADDICTION AS A NOVEL APPROACH TO DEVELOP A STABLE PLASMID VECTOR FOR XYLELLA FASTIDIOSA Project Leaders:

Understanding the progression of Pierce’s Disease (PD) has been limited by the lack of genetic and molecular tools that can be used to study the biology of Xylella fastidiosa (Xf). Although a number of potential plasmid vectors have been developed that are capable of replicating in Xf, none of these plasmids are stably maintained in Xf without antibiotic selection. To solve this problem, we have introduced two different types of stabilizing elements into the Xf plasmid vectors pXF004 and pXF005. These stabilizing elements include the plasmid addiction systems, hok/sok and parDE, and the active partitioning system, parA. Our preliminary studies indicate that the addition of the hok/sok addiction module to plasmid, pXF004, greatly increases its stability in Xf. INTRODUCTION Xylella fastidiosa (Xf) is a fastidious, xylem-limited, Gram-negative bacterium, which is the causative agent of numerous economically important plant diseases (Hopkins and Purcell 2002). Diseases that are important to the California agricultural economy include PD of grapevine, almond leaf scorch, alfalfa dwarf, and oleander leaf scorch. An important feature of the Xf infectious cycle is the ability of this pathogen to colonize and interact with the xylem tissue of plants and the foregut of insect vectors (Hopkins and Purcell 2002). Successful colonization of these hosts is dependent on the ability of Xf to subvert host defense networks and to acquire essential nutrients. Many research laboratories have been working to identify genes important for virulence and nutrient acquisition. However, rapid progress in this area is affected by the lack of genetic and molecular tools necessary to investigate the contribution of Xf genes to the infection process. In recent years, a number of plasmid vectors have been developed that are capable of replicating in Xf. These plasmids have different origins of replication and belong to different incompatibility groups (Qin and Hartung 2001, Vanamala, A. et al. 2002, Guilhabert and Kirkpatrick 2003, Guilhabert et al. 2005) However, in the absence of antibiotic selection, none of these plasmids are stably maintained in Xf. Therefore, one extremely important tool that is needed to advance studies investigating Xf virulence is a plasmid that is stably maintained by Xf throughout the infectious cycle. The goal of our project is to develop a plasmid that is stably maintained in Xf both in vitro and en planta in the absence of antibiotic selection. In our initial studies, we constructed these stable plasmids using the pXF plasmids, pXF004 and pXF005 (Guilhabert and Kirkpatrick 2003). These pXF plasmids contain the replicon from RSF1010 and confer resistance to kanamycin. They also are autonomously maintained with antibiotic selection and structurally unchanged by propagation in Xf. In fact, the only real problem with these vectors is that they are not maintained in Xf in absence of antibiotic selection. To circumvent this problem, we are evaluating whether stability can be achieved by introducing plasmid-addiction systems and plasmid partitioning elements into these existing Xf vectors. During the past year, we focused on plasmid addiction modules, which have been shown to dramatically increase plasmid stability in many Gram-negative bacteria (EngelbaergKulka and Glaser 1999, Zielenkiewicz and Ceglowski 2001, Hayers 2003). A plasmid addiction system is a two-component stable toxin-unstable antitoxin system. Examples of these systems include the hok/sok system of plasmid R1 and the parDE system of plasmid RK2 (Burkhardt et al. 1979, Saurugger et al. 1986, Gerdes 1988). When a bacterium loses the plasmid harboring either of these addiction systems, the cured cells loose the ability to produce the unstable antitoxin and the lethal effect of the stable toxin quickly kills the bacterium. Thus, a plasmid addiction system guarantees that all living bacteria maintain the plasmid throughout infectious cycle. Recently, we have initiated studies to examine whether or not active partitioning systems enhance plasmid maintenance. Specifically, we plan to test the plasmid partitioning system, parA, which consist of a centromere-like region adjacent to two co-regulated genes that encode an ATPase and a centromere specific DNA-binding protein, which is required for faithful plasmid segregation at cell division (Gerdes et al. 2000). Addition of this system to unstable plasmids has been demonstrated to increase plasmid stability in many Gram-negative bacteria (Zielenkiewicz and Ceglowski 2001).