Whitefly resistance in tomato: from accessions to mechanisms

Tomato (<em>Solanum lycopersicum</em>) is affected by a wide range of biotic stresses, of which <em>Bemisia tabaci</em> is one of the most important.<em>Bemisia tabaci</em> affects tomato directly through phloem sap feeding, and indirectly through its ability to be the vector of a large number of viruses. Different methods are available for whitefly control, and although several biological control agents are used against whiteflies in greenhouse cultivation, chemical control still is an essential component in open field tomato production. Breeding for host plant resistance is considered as one of the most promising methods in insect pest control in crop plants, and especially it is a promising alternative in whitefly control. Resistance to whiteflies was found in several wild relatives of tomato like <em>Solanum peruvianum, S. pennellii</em>, <em>S. habrochaites, S. lycopersicum var. cerasiforme, S. pimpinellifolium </em>and<em>S. galapagense</em>. In spite of previous breeding efforts, whiteflies are still a problem in tomato cultivation. The aim of my research was to identify and understand resistance mechanisms targeting specific stages of the whitefly life cycle in order to provide breeders with tools for developing whitefly resistant varieties. I assessed the natural variation and whitefly resistance in <em>Solanum galapagense </em>and <em>S. cheesmaniae</em>, two wild tomato species endemic to the Galapagos Islands. Previously, <em>Solanum galapagense</em> and <em>S</em>. <em>cheesmaniae</em> were classified as two species based on a morphological species concept, but with molecular markers no clear separation could be made. So far, only a limited number of accessions/populations of <em>S</em>. <em>galapagense</em> and <em>S</em>. <em>cheesmaniae</em> have been evaluated for insect resistance and therefore it was unknown if the insect resistance coincides with the morphological species boundaries. Neither was there any knowledge about the relation between geographical and climatic conditions today on the Galapagos and the occurrence of the two species. We characterized twelve accessions of <em>S</em>. <em>galapagense</em>, 22 of <em>S</em>. <em>cheesmaniae</em>, and as reference one of <em>S</em>. <em>lycopersicum</em> for whitefly resistance using no-choice experiments. Whitefly resistance was found in <em>S</em>. <em>galapagense</em> only and was associated with the presence of relatively high levels of acyl sugars and the presence of glandular trichomes of type I and IV.It is likely that a minimum level of acyl sugars and the presence of glandular trichomes type IV are needed to achieve an effective level of resistance. Genetic fingerprinting using 3316 polymorphic SNP markers did not show a clear differentiation between the two species endemic to the Galapagos. Acyl sugar accumulation as well as the climatic and geographical conditions at the collection sites of the accessions did not follow the morphological species boundaries. Altogether, our results suggest that <em>S</em>. <em>galapagense</em> and <em>S</em>. <em>cheesmaniae</em> might be considered as morphotypes rather than two species and that their co-existence is likely the result of selective pressure. Plants possess several resistance mechanisms acting at different time points during the interaction with herbivorous insect. Before any contact with the insects, plants emit an array of volatile organic compounds that can act as attractant or repellent of insects.<em>Bemisia tabaci</em> use a set of plant-derived cues in the process of host plant selection. It recognizes mainly monoterpenes (p-cymene, γ-terpinene and β-myrcene, α-phellandrene) and sesquiterpenes (7-epizingiberene and <em>R</em>-curcumene). Previously the line FCN93-6-2, which was derived from a cross between a susceptible tomato cultivar (Uco Plata INTA) and <em>S. habrochaites</em> (FCN3-5) was proved to be non-preferred by the greenhouse whitefly <em>Trialeurodes vaporariorum</em>. We identified chemical cues produced by FCN93-6-2 and <em>S. habrochaites</em> that can affect the preference of the whitefly <em>B. tabaci</em> as well as the potential chromosomal region(s) of <em>S. habrochaites</em> harbouring the genes involved in the preference. Two <em>S. habrochaites</em> accessions (CGN1.1561 and in FCN3-5) and the line FCN93-6-2 were non-preferred by <em>B. tabaci</em> when the whiteflies could get in direct contact with the plant and also when the whiteflies were offered olfactory cues only. The non-preference was independent of trichome type IV and of the presence of methyl-ketones but associated to the presence of monoterpenes in lower concentrations. Functional validation of the candidate metabolites and of the different introgressions is still needed. Once the insect has landed on a plant, another set of resistance mechanisms enter into action. We have described a 3.06 Mbp introgression on top of Chromosome 5 (<em>OR-5</em>) from the wild tomato species <em>S. habrochaites</em> (CGN1.1561). For the identification of <em>OR-5</em>, we went from the selection of specific F₂ plants to the development of F₂BC₄S₁and F₂BC₄S₂ families. This introgression was sufficient to reduce whitefly fecundity without an evident effect on whitefly survival. The identification of mechanisms exclusively affecting whitefly fecundity and independent of trichomes type IV opens new doors for resistance breeding to whiteflies that may be especially interesting in greenhouse cultivation combination with natural enemies of the whitefly. As an additional layer of defences, plants can perceive stress signals and respond to them in a specific way through induction of their immune system. This induction can also be triggered by exposing the plants to priming agents like hormones, some xenobiotic chemicals, like benzothiadiazole (BTH), β-aminobutyric acid (BABA), and sugars. Although the effect of priming agents was shown in laboratory and field studies, little is known about the effect of the genetic background of tomato on the extent of the priming, e.g. do genotypes varying in their level of resistance to insects and pathogens respond in the same way to a priming agent. We assessed the effect of selected priming agents on the effectiveness of natural defence in tomato. A set of no-choice and choice bioassays was conducted using tomato genotypes varying in their level of basal resistance to <em>Bemisia tabaci</em> and pathogens. We observed that whitefly survival and oviposition were not affected by the priming treatment in no-choice assays. Overall, in choice assays, fructose treated plants were more preferred by whiteflies than control plants. A genotype specific effect of priming was seen for the line FCN93-6-2. On this tomato line, JA and BABA applications decreased the number of whiteflies, e.g. making them less preferred. In this thesis, I have gone from the screening of wild relatives of tomatoes to in depth characterization of resistance mechanisms. I have identified resistance mechanisms targeting specific stages of the whitefly life cycle, thus providing new tools for breeding durable whitefly resistance in tomato.