Biological disinfestation of soil-borne bacterial wilt pathogen using a large population of virulent bacteriophages supported via the lytic cycle of non-phytopathogenic host bacteria introduced into soil

Field-grown tomato plants frequently suffer from infection by Ralstonia solanacearum, a bacterial wilt pathogen. We have not succeeded in producing bacterial wilt-resistant tomato cultivars via conventional breeding techniques. This work established an alternative biological method to disinfest bacterial pathogens in soil; i.e. the lysis of bacterial cells by the infection of virulent bacteriophages (phages) introduced into soil. For this purpose, we collected strains of R. solanacearum (tester strains) to screen various virulent phages. Soil samples were collected from 1210 fields (including paddy fields) located in several districts of Japan to obtain phages with different host ranges. The isolated phages were inoculated into 48 tester strains separately and eventually classified into 56 types based on their distinct host ranges. Lytic infection was confirmed using the standard plaque formation assay. Our procedure consists of several steps: 1) monitoring bacterial pathogens living in the fields used for tomato cultivation; 2) selecting phages that infect all of the bacterial isolates obtained from the test field; 3) multiplying the selected phages on a large scale using a 100-L jar fermentor to prepare a bacterial lysate containing 10 to 10 phage particles/ml; 4) diluting the lysate with 10 volumes of a fresh suspension of a non-phytopathogenic, phage-susceptible isolate of R. solanacearum to ensure subsequent multiplication of the phages in the soil; 5) soaking the soil with the diluted lysate and subsequently tilling the soil to promote encounters between the phages and target bacterial pathogens in the soil; and 6) confirming the successful elimination of the soil-borne pathogen from a field. The bacteria in soil were detected by suspending soil samples in sterilised water, centrifuging the suspension at low speed, and then spreading the supernatant on solid Tz-PCG medium selective for R. solanacearum. After a 2-day incubation, the resulting pinkish-red pathogen-specific colonies were counted to determine the number of bacteria in the soil samples. The practicability of the system was examined using an experimental 5 × 5-m block (depth 60 cm) in a paddy field to be switched to tomato cultivation. Soil samples were collected at points throughout the block. In all samples, specific bacterial colonies with different growth speeds or colony shapes appeared on the selective medium. These isolated bacteria were shake-cultured in liquid broth until the solution became turbid, and the turbid culture was divided for inoculation with different phage