Reliability of diagnostic techniques for Erwinia amylovora, the causative agent of fire blight disease.
A total of 20 putative strains of Erwinia amylovora originating from 11 samples of host plants with symptoms of fire blight were analyzed in detail using commercial polyclonal antibodies in immunochemical tests. Fourteen strains reacted negatively in all tests; 6 strains reacted positively with a polyclonal antibody for PTA-ELISA (plate-trapped antigen-enzyme linked immunosorbent assay) at a concentration corresponding to A620 = 0.1, while at A620 readings of 0.01 and 0.001 the results were negative. Five strains reacted positively with a polyclonal antibody for indirect immunofluorescence test at all tested concentrations. Three of those strains were positive in the PCR test with AMSbL and AMSbR primers designed for detection of E. amylovora. In hypersensitivity test in tobacco and in immature pear fruit assay, all putative strains were negative while a known reference strain of E. amylovora gave a typical hypersensitive-reaction response. On a medium with 5% sucrose the reference strain of E. amylovora produced levan while putative strains did not. After modification of the PCR protocol, 3 putative strains reacted as negatives. Optimization of PCR test was achieved by finding the optimum annealing temperature and time for primers. The recommended annealing temperature (49 degrees C) for these primers was increased to 55 degrees C and the annealing time was reduced from 2 min to 30 s. Using the microbial identification system Biolog those 3 strains were identified as Pantoea dispersa (1 strain) and Pantoea agglomerans (2 strains). The strains are supposed to be white variants of the species P. dispersa and P. agglomerans occurring less frequently than the yellow variants. Since there were positive reactions in our immunochemical tests these strains could cause false positives in routine screening of plant samples.