CORDIS - Wyniki badań wspieranych przez UE
CORDIS

Evolution of plant volatiles manipulation by vectored pathogens

Final Report Summary - PLANT VOLATILES (Evolution of plant volatiles manipulation by vectored pathogens)

Our project aims to disentangle the adaptive nature of the manipulation by plant pathogens of their host phenotype, including the volatile cues and the resistance they mount to herbivore insect vectors. Indeed, for vector-transmitted diseases the relationship between infected hosts and the insect vectors that feed on them will have a crucial impact on pathogen transmission. Recently, several studies have demonstrated that some pathogens, viruses in particular, induce volatile emission and / or resistance changes in their host plant that lead to their increased attractiveness for insect vectors. It is expected that these host phenotype alterations are beneficial for the transmission of these pathogens, but until now there has been very little attention paid to the question of how these strategies can evolve and be maintained in a complex ecological context where many factors, like the presence of competitor strains or species and the type of vector transmission, can affect the selective pressures acting on these traits.

In a first approach, we studied the functional mechanisms implicated in the pathogen-induced modifications of host resistance and volatile emissions by using tomato yellow leaf curl China virus (TYLCCNV), a whitefly-transmitted deoxyribonucleic acid (DNA) virus in which the function of different genes are well described and Arabidopsis thaliana and Nicotiana benthamiana as model hosts. We found that the gene beta-C1, the unique gene of TYLCCNV satellite beta and previously known to underlie several pathogenicity traits of the virus, is responsible for an alteration of jasmonic acid production in Arabidopsis and N. benthamiana plants transformed to express this gene. The jasmonic acid resistance pathway is notably efficient in reducing the performance of whiteflies, so this alteration might indirectly benefit the virus by enhancing the population performance of his vector. We also observed that the production of volatile linalool of the beta-C1-transformed plants was increased, suggesting that this gene could also be involved in the manipulation of volatile cues to the vectors. These first results demonstrate that changes in host phenotype that can benefit vector transmission of the virus can be directly attributed to a single gene and are not just a by-product of viral multiplication in host tissues. Indeed, in the lines used here only the individual gene was expressed independently of any virus infestation. Interestingly, this gene serves other functions for the virus, including an inhibition of plant ribonucleic acid (RNA) silencing resistance, and also induces leaf deformations that are typical symptoms of the TYLCCNV disease. This suggest that, in the very size constrained viruses, the genes for host phenotype manipulation can be co-opted from other tasks, or possibly act on core regulation mechanisms of the hosts with other functions.

In a second approach with two potyviruses infecting cucurbitaceae hosts, we investigated more ecological aspects of the virus-induced changes in host phenotype linked to aphid transmission. As competition between pathogens sharing the same vectors could be a factor influencing the selective pressures on this trait, we investigated the impact of co-infection by watermelon mosaic virus (WMV) and zucchini yellow mosaic virus (ZYMV) on their respective performance and exposure to aphid vector within the host plant, as well as its impact on the phenotype changes suffered by the host and the consequences for aphid preference and performance. Our results demonstrated that WMV is an inferior competitor within the host plant when faced with ZYMV in co-infections, but can still be successfully transmitted to new plants. Furthermore, only the ZYMV isolate was capable of inducing phenotypic changes in the host plants that resulted in a greater attractiveness for the aphids and their higher growth rate on infected hosts. Interestingly, the co-infected plants carrying the two viruses had the same changes as shown in single ZYMV infection, which suggest that WMV viruses in these plants can benefit from the increased aphid traffic induced by ZYMV even though they cannot themselves induce these changes in single infections. This is an important finding that shows that host phenotype manipulators are vulnerable to cheating by non-manipulative competitors. Additional experiments indeed show that different viral strains, in particular within ZYMV, differ in the amplitude of change in host volatile emissions they induce, thus potentially in their ability to attract aphid vectors. This would suggest that this trait could respond to selective pressures arising from the ecological context of the host populations.

Our findings have major implications for the understanding of pathogen evolution. We showed that, beyond the description of changes in plant volatiles previously reported in the literature, these parasite-induced changes might result from actual parasite transmission strategies and not be only a by-product of the disruption of plant physiological traits. The fact that there is variation for this pathogen trait, and that it is vulnerable to cheating by competitor parasites, is also an indication that this type of manipulation might only arise in ecological contexts that favour it, i.e. when co-infections are rare. Taking into account the adaptability of this trait could thus have a large impact of our understanding of plant pathogens epidemiology.

This project was also an opportunity to bring together several field of ecology: behavioural, chemical, environmental and pathogen or herbivore-host interactions. This interdisciplinary approach was applied again in two additional questions that arise from our main project. In addition to our study of viral co-infections in cucurbits, we investigated the impact of viral infection on plant susceptibility to Erwinia tracheiphila, a bacterial disease that is often fatal to the host plant. Indeed, it has been reported that virus infected cucurbit plants in the field are less likely to develop this bacterial disease, at least partly due to indirect interactions through a reduced visitation by E. tracheiphila beetle vectors. In order to complement this hypothesis, we performed controlled inoculation experiments and phytohormone analyses to test whether direct in planta effects acting on plant resistance could also play a role. We found that an early infection with ZYMV activated an enhanced salicylic acid response in the host plant when challenged with E. tracheiphila inoculation, but did not prevent the successful establishment of the bacteria. Only slight delayed effects on symptoms and plant death were observed, which cannot alone explain the wide epidemiological divergence occurring in the field. Our experiments thus confirmed that indirect effects of viruses on beetle vectors are the likeliest explanation for this phenomenon. This multidisciplinary approach also led to a good example of serendipity. While investigating A. thaliana volatile emissions, we observed that when plants were exposed to dimethyl sulfoxide (DMSO) - a common solvent used as a compound carrier in many applications - in an in-vitro or compost soil matrix, this chemical was released through the leaf surface. As this solvent can often be encountered by plants, since it is included in many pesticide compositions and can be found as a pollutant in some areas due to other industrial applications, we experimentally demonstrated this phenomenon of uptake and release of DMSO by A. thaliana and tobacco plants and investigated the possibility of ecological effects on caterpillar herbivory and feeding behaviour.
potyviruses1.jpg