Identification of genes affecting the response of tomato and Arabidopsis upon powdery mildew infection

Many plant species are hosts of powdery mildew fungi, including Arabidopsis and economically important crops such as wheat, barley and tomato. Resistance has been explored using induced mutagenesis and natural variation in the plant species. The isolated genes encompass loss-of-function susceptibility genes and dominantly inherited genes encoding NB-LRR proteins, receptor-like kinases or proteins that do not have typical resistance protein domains. Cultivated tomato is susceptible to powdery mildew species Oidium neolycopersici, and exploiting the resistance genes present in wild tomato species is a favourable strategy to control the disease. In chapter 2, we give an overview of all the identified resistance genes in wild tomato species and their resistance mechanisms inferred from cytological and molecular data. Furthermore, resistance genes and their mechanisms are compared between tomato and other plant species, such as dicot Arabidopsis and monocots barley and wheat. This comparison illustrates that both common and species-specific mechanisms are involved with respect to resistance to powdery mildews in different plant species. Resistance gene Ol-1 originates from wild tomato species S. habrochaites. It confers race-non-specific resistance to tomato powdery mildew. To elucidate the resistance signalling pathway, we adopted a virus induced gene silencing (VIGS) approach to suppress genes which are differentially expressed when comparing genotypes with and without the Ol-1 introgression. In chapter 3, we showed that ALS (acetolactate synthase) activity is important for Ol-1-mediated resistance, as simultaneous silencing of two ALS genes attenuated the resistance level of NIL-Ol-1. ALS is a key enzyme in the biosynthesis of branched-chain amino acids, and a target of commercial herbicides. Reducing ALS activity via herbicidal treatment did not result in altered responses to powdery mildew infection in susceptible cultivar Moneymaker and resistant line NIL-Ol-4, indicating that ALS is not involved in basal defense nor in NB-LRR gene-mediated resistance. Whether the role of ALS in Ol-1-mediated resistance is associated with amino acid homeostasis is unknown and needs further investigation. Besides tomato, Arabidopsis is a host of powdery mildew O. neolycopersici. The large collection of Arabidopsis accessions and several mutant collections are valuable resources to identify novel resistance genes. In chapter 4, we first screened 123 Arabidopsis accessions for O. neolycopersici resistance and then studied the genetic basis of theresistance by segregation analysis in 19 F2 populations. The results showed that polygenic resistance is the main form of resistance. Accession C24 displays complete resistance with polygenic nature, as shown by QTL analysis of the F2 population derived from the cross between C24 and susceptible accession Sha. The recessively inherited locus on chromosome 1 was fine-mapped by recombinant screening, and analysis of candidate genes resulted in the isolation of the gene conferring resistance. It proved to be a mutant allele of EDR1, harbouring a deletion upstream of the kinase domain resulting in a truncated protein. Previously, an induced edr1mutationin Col-0 background was obtained. However, the edr1 mutation in our C24 source (referred to as C24-W) occurred in a different position. The resistance conferred by edr1 in C24-W was not associated with constitutively expressed pathogenesis-related genes. Remarkably, we observed that although C24-W carried the edr1 mutation this mutation was absent in other C24 sources. In addition, C24-W was smaller in size than C24 from other sources. Since the edr1 mutation confers resistance to tomato powdery mildew in Arabidopsis, we investigated whether this resistance system is conserved in tomato. The results showed that individual silencing of two tomato EDR1 candidate genes in susceptible cultivar Moneymaker did not result in decreased sporulation of tomato powdery mildew. In chapter 5, we screened an activation tag Arabidopsis mutant collection. In these mutants, tagged genes are overexpressed by the strong 35S enhancers resulting in a dominant gain-of-function phenotype. One mutant line, 3221, was identified due to its resistance to powdery mildew O. neolycopersici. Additional disease tests showed that 3221 displayed resistance to the downy mildew Hyaloperonospora arabidopsidis and the aphid Myzus persicae, but susceptibility to the bacterial pathogen Pseudomonas syringae pv tomato DC3000. The mutant line 3221 also showed reduced size and serrated leaves, and the altered morphology was associated with resistance. Inverse PCR and expression analysis revealed that the gene underlying the resistance was ATHB13, a HD-Zip transcription factor. Suppression ofATHB13 in 3221 by RNAi transformation resulted in the loss of resistance and altered morphology, while overexpression of ATHB13 in wild-type plants induced resistance and altered morphology. Microarray analysis of 3221 and the parental line Ws resulted in the identification of a large number of genes showing differential expression. Analysis of these results did not give a clear indication that the resistance phenotype in 3221 is due to the activation of classical hormone pathway genes involved in resistance. The possibility of utilizing ATHB13 for engineering pathogen resistance in tomato needs to be investigated in the future. Finally, in chapter 6 the results from the previous chapters are discussed in a broader context.