Impact of Recombinant Baculovirus Applications on Target Heliothines and Nontarget Predators in Cotton

Recombinant baculoviruses have been genetically engineered to reduce the time to kill infected pests, thus reducing crop damage. In this study, wild-type viruses and recombinant viruses expressing a scorpion toxin were applied to cotton in response to larval infestations of Helicoverpa zea and Heliothis virescens in 1997 and 1998. A chemical standard and an untreated control acted as comparison treatments. The goals of this field study were to (1) assess the efficacy of recombinant baculoviruses in protecting cotton from larval feeding damage; (2) assess the impact of recombinant virus introductions on predator densities and diversity; and (3) determine if cotton predators acquire baculovirus by consuming infected heliothines. When applications were timed at larval emergence, certain recombinant virus treatments protected cotton from damage better than wild-type virus treatments and as well as the chemical standard. Differences in efficacy between recombinant and wild-type baculoviruses were not apparent if treatments were applied 3 to 4 days after peak larval emergence. Predator densities and diversity were similar among recombinant and wild-type baculovirus treatments, whereas plots treated with the chemical standard had consistently smaller predator populations. From polymerase chain reaction analyses of predators in 1997 and 1998, 1.7 and 0.2%, respectively, of predators had consumed a virus-infected heliothine. Nine of the 26 predators carrying viral DNA were positive for recombinant virus. Additionally, 13 of the 26 predators were found to disperse 13.5 to 105 m 2 to 5 days after initial virus applications. Five of these dispersing predators (0.2% of all predators evaluated) carried recombinant viral DNA. These results suggest that the potential for the inadvertent spread of recombinant viral DNA via dispersing predators is low.