Cucumber mosaic virus (CMV) has three genomic RNA segments, RNAs 1, 2 and 3, that encode five proteins. In some hosts, CMV induces biochemical or physiological changes that may influence virus transmission by aphids. In Arabidopsis thaliana ecotype Col-0 and cucurbits, CMV diminishes host suitability for aphids, inhibiting prolonged feeding and encouraging migration of viruliferous aphids from infected plants to healthy plants. In tomato (Solanum lycopersicum), a model species and economically important crop, viral modification of vector-host interactions for non-persistently transmitted viruses is understudied. I investigated CMV-induced effects on settling preference of two aphid species (Myzus persicae, a generalist, and Macrosiphum euphorbiae, a Solanaceous specialist) at 3, 9 and 21 days post-inoculation (dpi) and at 1 and 24 h following aphid release, using aphid free choice and trapping assays in light and dark conditions and under light of varying wavelengths. Further experiments were carried out to examine whether specific CMV viral proteins control aphid-tomato interactions. Finally, the role of plant signalling pathways in CMV-aphid interactions in tomato and tobacco was investigated. In tomato, the effects of Fny-CMV infection on aphid settling varied as disease progressed. While pre-symptomatic (up to 3 dpi), infected tomato plants were more preferred by Myzus persicae than by Macrosiphum euphorbiae. As symptoms became apparent, between 9 and 21 dpi, more aphids settled on mock-inoculated plants than on virus-infected plants. In trapping assays in the dark, aphids did not prefer either type of plant, despite virus-infected plants emitting greater quantities of volatile organic compounds in the dark than in the light. In olfactometry assays, in which visual and contact cues were not available, aphids of both species were more attracted by odours emitted by virus-infected plants than to those of mock-inoculated plants. Thus, CMV stimulates multiple cues, but visual cues are prioritised in aphid-tomato interactions. Specific CMV gene products, in particular the 1a, 2a and 2b proteins, contribute to virus-induced changes in plant-aphid interactions, sometimes in a host-specific fashion. The effects of CMV pseudorecombinant viruses on settling preference of Myzus persicae and Macrosiphum euphorbiae on tomato was assessed. Six pseudorecombinant viruses were made by systematic exchange of RNAs of LS-CMV, a mild strain on tomato, with those of Fny-CMV, a severe strain. It was found that RNA 2 regulates the ability of Fny-CMV to repel aphids in tomato. Further experiments with a recombinant CMV virus, where the Fny 2b protein was replaced with LS 2b, and the Fny-CMVΔ2b deletion mutant, suggested that the Fny-CMV 2b protein but not the LS-CMV 2b protein, influences settling of both aphid species on tomato.
I tested whether salicylic acid (SA) influences CMV-aphid-tomato interactions. NahG-transgenic tomato plants unable to accumulate SA were used. Fny-CMV infection of NahG-transgenic tomato plants affected the settling of Myzus persicae and Macrosiphum euphorbiae differently. NahG-transgenic plants infected with Fny-CMV attracted Myzus persicae but not Macrosiphum euphorbiae. More aphids of both species settled on mock-inoculated untransformed tomato plants than on Fny-CMV infected NahG-transgenic plants. Thus, the effects of SA-dependent signalling on tomato-aphid interactions are limited, with only a small influence on the interaction with Myzus persicae. Previous work showed that on tobacco plants infected with Fny-CMV and LS-CMV Myzus persicae survival and fecundity are enhanced but on plants infected with Fny-CMVΔ2b, aphid survival and fecundity are decreased. Using CO11-silenced transgenic tobacco plants I showed that the aphid resistance induced by Fny-CMVΔ2b infection is dependent upon defensive signalling mediated by jasmonic acid.