Insect vector manipulation by a plant virus and simulation modeling of its potential impact on crop infection.

There is widespread evidence of plant viruses manipulating behavior of their insect vectors as a strategy to maximize infection of plants. Often, plant viruses and their insect vectors have multiple potential host plant species, and these may not overlap entirely. Moreover, insect vectors may not prefer plant species to which plant viruses are well-adapted. In such cases, can plant viruses manipulate their insect vectors to preferentially feed and oviposit on plant species, which are suitable for viral propagation but less suitable for themselves? To address this question, we conducted dual- and no-choice feeding studies (number and duration of probing events) and oviposition studies with non-viruliferous and viruliferous [carrying beet curly top virus (BCTV)] beet leafhoppers [Circulifer tenellus (Baker)] on three plant species: barley (Hordeum vulgare L.), ribwort plantain (Plantago lanceolata L.), and tomato (Solanum lycopersicum L.). Barley is not a host of BCTV, whereas ribwort plantain and tomato are susceptible to BCTV infection and develop a symptomless infection and severe curly top symptoms, respectively. Ribwort plantain plants can be used to maintain beet leafhopper colonies for multiple generations (suitable), whereas tomato plants cannot be used to maintain beet leafhopper colonies (unsuitable). Based on dual- and no-choice experiments, we demonstrated that BCTV appears to manipulate probing preference and behavior by beet leafhoppers, whereas there was no significant difference in oviposition preference. Simulation modeling predicted that BCTV infection rates would to be higher in tomato fields with barley compared with ribwort plantain as a trap crop. Simulation model results supported the hypothesis that manipulation of probing preference and behavior may increase BCTV infection in tomato fields. Results presented were based on the BCTV-beet leafhopper pathosystem, but the approach taken (combination of experimental studies with complementary simulation modeling) is widely applicable and relevant to other insect-vectored plant pathogen systems involving multiple plant species.