Discovery and characterization of the major late blight resistance complex in potato: genomic structure, functional diversity, and implications

Potato is the most important non-cereal crop in the world. Late blight, caused by the oomycete pathogen Phytophthora infestans, is the most devastating disease of potato. In the mid-191h century, P. infestans attacked the European potato fields and this resulted in a widespread famine in Ireland. Late blight remains the No.l constraint to potato production and causes a yearly multi-billion US$ loss globally. In Europe and North America, late blight con trol heavily relies on the use of chemicais, which is hardly affordable to farmers in developing countries and also raises considerabie environmental concerns in the developed countries.<o:p></o:p></span>Use of host resistance is ecologically the most sustainable way of disease management. Disease resistance (R) genes exist in a wide range of wild species of the genera Solanum. One wild species, Solanum demissum, became the donor of most characterized R genes due to its crossability with the cultivated species S. tuberosum. However, these race-specific genes (Rl-Rll) didn't enable satisfactory protection under the current deployment scheme. Breeders either turned to other wild species or adopted a so-called 'R-gene-free' approach to explore quantitative resistance. Unfortunately, neither alternative offered a solution to late blight control and did not result in commercial release of cultivars with durable resistance.<o:p></o:p></span>At the end of the 20th century, biological research entered the genomics era, landmarked by the Human Genome Project and by the Arabidopsis and rice genome sequencing initiatives. Genomics also became the new frontier of research in potato and P. infestans. This scientific development is deepening and broadening our understanding of the biology of the host and the pathogen and is facilitating isolation of key genes involved in the interaction. The genetically modified organism (GMO) strategy allows a much more efficient application of these genes than time-consuming conventional breeding. This thesis deals with the isolation, characterization, and deployment of host R genes with the expectation to achieve an ecologically and economically sound control of late blight.<o:p></o:p></span>The potato-P.infestans</span> interaction follows the gene-for-gene model, that is, resistance only occurs when a host R gene and its corresponding avirulence (Avr) gene in the pathogen are both present. Several disease testing methods have been developed for determination of the gene-for-gene interaction between potato andP. infestans. In vitro inoculation was developed as a quick, space-effective, and accurate assay (Chapter 2). The method exploits the amenability of potato for tissue culture and the suitability of the in vitro environment for late blight disease development.<span class=GramE>lts</span>specificity and reliability was confirmed by comparison with the well-established detached-Ieaf assay. Currently, in vitro inoculation is routinely used in phenotyping of segregating populations, resistance testing of transformants for functional complementation, and screening of new R genes in a wide range of wild germplasms.<o:p></o:p></span>The investigation of host resistance was focused on the R3 complex locus on the distal part of chromosome I\\. The R3 complex locus segregated in a potato mapping population, which was used to construct the potato ultra-high dense (UHD) map saturated with over 10,000 amplified fragment length polymorphism (AFLP) markers. Using a population of 1748 plants, we constructed a high-resolution genetic map at the R3 complex locus. The combination of fine-mapping and accurate disease testing with specific P. infestans isolates resulted in the unexpected discovery that the R3 complex locus is composed of two functionally distinct genes, R3a and R3b, which are 0.4 cM apart and have both been introgressed from S. demissum (Chapter 3). Each gene was localized into a genetic interval of 0.25 cM, providing the starting point for map-based cloning.<o:p></o:p></span>Plant R gene families undergo fast evolution, resulting in considerable intra­and inter-specific variation. Plant disease resistance (R) loci frequently lack synteny between related species of cereals and crucifers but appear to be positionally well conserved in the Solanaceae. Comparative genomics provides a tooi to utilize the exponentially increasing sequence information from model plants to clone agronomically important genes from less studied crop species. We were keen to investigate whether this tooi can enable new R gene cloning by a case study. The comparative study revealed that the potato late blight R3 locus and the tomato Fusarium wilt 12 locus were derived from an ancestrallocus involved in plant innate immunity. We adopted alocal RGA approach using DNA sequences of the 12 gene to isolate the R3a gene (Chapter 4).<span class=GramE>l2</span> and R3a share 88% and 83% identities at the DNA and protein level, respectively. R3a is a member of the R3 complex locus. Comparative physical mapping disclosed that the potato R3 complex locus underwent a significant expansion after divergence from tomato without disruption of the flanking colinearity. This expansion resulted in an increase in the number of R genes and functional diversification.<span class=GramE>lnterestingly</span>, the differential evolution of the ancient R locus in the two closely related species is well correlated with the contrasting evolutionary potentials of the pathogens with which 12 and R3 interact.<o:p></o:p></span>Fusarium oxysporum is a soilborne fungus with low genotype diversity, whereas the late blight pathogen P. infestans is notorious in its ability to move and mutate. An intriguing question is why<span class=GramE>S. demissum</span>.<span class=GramE>a</span>species that appears to contain only easily-broken R genes, can display durable resistance at the population level in its natural environment. The polymorphism of parasite recognition capacity in a host population will restrict most isolates of the parasite to grow on most hosts. Allelism is an efficient way of creating recognition polymorphism. We made another unexpected discovery that the R3 complex locus has very high allelic diversity and that the R5 -R 11 resistant specificities all contain a distinct allelic version (Chapter 5). Sequence exchange between alleles and diversifying selection are the major driving forces of this allelic recogntion polymorphism to P. infestans. Remarkably, the genomic structure of the R3 complex locus favors the creation of new resistance specificity by reshuffling of elements from the two clusters (R3a and R3b). The multiple allelism of the R3 complex locus may be a natural mechanism of S. demissum to suppress late blight epidemics and should be mimicked in resistance breeding. We suggest that R-gene polyculture via the GMO approach should be the future paradigm of R gene deployment in late blight controI.<o:p></o:p></span>Potato breeding for late blight resistance was one of the earliest mankind practices in combating plant pathogens by means of genetic improvement but the disease has not been controlled by resistance breeding so faro The non-durable nature of S. demissum R genes apparently disappointed most breeders and the blame of easily-broken R genes even led to the unsuccessful R-gene-free approach. However, from an evolutionary and ecological point of view, single R genes can never defeat pathogens such as P. infestans with extremely high evolutionary risk. We propose that the community rethinks its strategy of R gene deployment in late blight disease management. R gene monoculture is obviously not recommended. R gene pyramiding is currently practiced by breeders but this strategy is basically the same as R gene monoculture since it creates uniformity in host resistance specificity in the field, which will be eventually broken by the fast-evolving pathogen. R gene polyculture should be the strategy of the future. However, we should soberly realize that host resistance might be high enough for survival of the plant population but it alone might never offer a protection that meets the economical threshold. Integrated pest management should include R gene polyculture as the central element, cultivation measures, and limited chemical applications To provide this central element, we need to clone more R genes. Our discovery of the multiple allelism of the R3 potato late blight resistance complex and molecular characterization of one of its allelic versions will offer a possibility to a dozen R genes in the near future and then to deploy them in a potato field using marker-free GMO techniques