Bioassays.

The pathogenicities of S. entomophila, S. proteamaculans, and their derivatives toward the larvae of a range of the species belonging to the family Scarabaeidae, specifically the New Zealand grass grub (C. giveni), the manuka beetle (P. festiva and P. setosa), the African black beetle (Heteronychus arator), the redheaded cockchafer (Adoryphorus couloni), the Tasmanian grass grub (Acrossidius tasmaniae), and the chafer beetle (Odontria sp.), were bioassayed as follows. Healthy third-instar larvae were individually fed carrot cubes (∼3 mm³ for grass grub larvae and ∼2 mm³ for manuka beetle larvae) that had been rolled in a lawn of bacterial colonies grown overnight on LB agar, resulting in approximately 10⁷ CFU/carrot cube. Twelve larvae were used for each treatment. Larvae were fed treated carrot on day 1 and were transferred to fresh trays containing untreated carrot on days 3 and 6. The occurrence of disease symptoms, including a change in coloration from amber through to brown/black and feeding activity, was monitored on days 3, 6, and 12. S. entomophila strain A1MO2 and, when required, S. proteamaculans strain AGR96X were used in all bioassays as positive controls, with negative-control larvae being fed untreated carrot.

To satisfy Koch's postulates for AGR96X, bacteria isolated from diseased larvae were reisolated and bioassayed against healthy larvae to confirm their pathogenicity, and the process was repeated. Dose-response bioassays against C. giveni larvae were performed by pipetting 5-μl aliquots of serial dilutions (corresponding to 1 × 10¹ to 1 × 10⁶ CFU) of AGR96X onto the surfaces of carrot cubes (3 mm³). Bioassays were then undertaken as outlined above, except that larvae were transferred to new trays containing freshly cut carrot cubes after 24 h. Negative controls were treated with 5 μl of LB broth. For the assessment of mitomycin C-induced Afp and AfpX samples, 5-μl aliquots of filter-sterilized semipurified samples were pipetted onto the surfaces of carrot cubes (3 mm³) and bioassayed as outlined above. Three independent bioassays were conducted for each treatment. Because of the accessibility, vigor, and consistent feeding activity of C. giveni larvae, this species was used in preference to manuka beetle larvae for assessments of the pathobiology of AGR96X.