1Bureau of Sugar Experiment Stations, PO Box 86, Indooroopilly, Qld 4068, Australia; 2Cooperative Research Center of Tropical Plant Pathology, University of Queensland, Brisbane, Qld 4072, Australia; 3Bureau of Sugar Experiment Stations, PO Box 566, Tully, Qld 4854, Australia; 4Joint Center for Crop Improvement, University of Melbourne, Parksville, Vic 3052, Australia; 5Department of Botany, University of Queensland, Brisbane, Qld 4072, Australia

Background and objectives
Ratoon stunting disease (RSD), caused by Clavibacter xyli subsp. xyli (Cxx), is the most economically important disease of sugarcane in the world. Cxx is a fastidious, slow-growing, Gram-positive, rod-shaped bacterium. There are no distinguishing external symptoms permitting easy recognition of RSD as the cause of stunting or selection for disease resistance. Very little has been done to breed for resistance, but hot-water treatment of the vegetative planting material, 50C for 4 h, eliminates Cxx unless there is a very high bacterial population in the stalks. Unfortunately, some varieties will not germinate after such treatment, and in some varieties hot-water treatment makes the cane more susceptible to sugarcane smut disease. In order to more clearly understand the host-pathogen interaction, potentially to develop novel forms of resistance to RSD, transposon mutagenesis was used to tag genes from Cxx involved in pathogenicity.

Materials and methods
The transposon Tn4431 [1], containing a promoterless lux operon from Vibrio fischeri and a tetracycline resistance gene, was chosen for the mutagenesis of Cxx. Mutants were isolated based on resistance to tetracycline and were characterized for light production in culture and the ability to colonize sugarcane. Colonization was used as the indicator of pathogenicity because of the lack of reliable symptom expression. To inoculate, the tops were chopped off of 4- to 6-week-old plantlets and a suspension of each Cxx mutant was placed on the cut surface. The plantlets were grown from 10 to 15 weeks and then the xylem juice was extracted and assayed.

Samples were assayed for the presence or absence of Cxx using evaporative-binding enzyme-linked immunoassay (EB-EIA) [2]. Samples that were negative were re-assayed for colonization a minimum of two additional times with increasing replications from three plants in the primary screen to 15 plants in the final screen. DNA flanking the transposon insertion was cloned from a confirmed avirulent mutant using inverse PCR. This fragment was used as a probe to isolate clones from a wild-type Cxx cosmid library for further sequence analysis and complementation studies.

Results and conclusions
Approximately 1000 transposon mutants were generated and screened for loss of pathogenicity. Three mutants were unable to colonize sugarcane after three rounds of inoculation into sugarcane variety Q110. The DNA flanking the transposon insertion was amplified from one of these mutants. Sequence analysis of the translation of this fragment revealed strong homology to a cell surface protein from Mycobacterium tuberculosis. Southern blot analysis confirmed that the inverse PCR product was the flanking DNA. The inverse PCR product was then used as a probe to isolate the wild-type gene from a cosmid library of a wild type Cxx (QPF110). Two unique cosmid clones were identified. Complementation studies using these two cosmid clones are in progress.

1. Shaw JJ, Settles LG, Kado CI, 1988. Molecular Plant-Microbe Interactions 1, 39-45.
2. Croft BJ, Greet AD, Leaman TM, Teakle DS, 1994. Proceedings of the Australian Society of Sugar Cane Technologists 16, 143-51.