Evidence continues to accumulate supporting the general hypothesis that plant viruses manipulate the performance and host selection behaviors of their vectors that can promote virus transmission. Work in this area has focused on intraspecific virus transmission, but most plant viruses have multiple host species and their transmission among these species will affect virus prevalence and persistence in many natural and managed settings. Selection should favor strategies to optimize virus persistence and prevalence in these settings, with implications for their evolution and management. We conducted a preliminary study to test the hypothesis that plant viruses manipulate multiple hosts to broaden the realized host range of their vector among these hosts. We tested the effects of infection by Bean leaf roll virus (BLRV) and Pea enation mosaic virus (PEMV) on preference and performance of five genotypes of pea aphid, Acyrthosiphon pisum, with different performance and preference profiles on six legume host plant species. After plants were infected with PEMV, pea aphid performance, measured as colony growth rate, was more similar across all six hosts than when plants were not virus-infected. In contrast, after plants were infected with BLRV, pea aphid performance differed more among hosts than when plants were not infected. There were no effects of infection by either virus on aphid preference in a six-choice settling experiment. The results have several possible interpretations, but suggest multi-host manipulation by plant viruses occurs, differs among virus species, and has implications for virus spread and persistence in multi-host environments.
