Eastern Filbert Blight in Hazelnut (Corylus avellana): Identification of New Resistance Sources and High Resolution Genetic and Physical Mapping of a Resistance Gene

European hazelnut, Corylus avellana L., is the only economically important nut crop in the family Betulaceae. One of the threats to the hazelnut industry in the Pacific Northwest is the fungal disease eastern filbert blight (EFB) caused by the pyrenomycete Anisogramma anomala . Host genetic resistance to EFB identified in the obsolete pollinizer 'Gasaway' has been extensively used in the hazelnut breeding program at Oregon State University. Concern over deployment of a single resistance gene prompted a search for new sources of resistance. Eighty six accessions from ten countries were evaluated for their response to greenhouse inoculation with the pathogen. Nine accessions showed complete resistance. These new sources of EFB resistance have geographically diverse origins and will broaden the genetic base of our EFB-resistant hazelnut germplasm.;Map-based cloning of the EFB resistance gene from 'Gasaway' hazelnut was initiated by constructing a BAC library for 'Jefferson' which is heterozygous for resistance. The BAC library was constructed using the cloning enzyme MboI and the vector pECBAC1 (BamHI site). The library consists of 39,936 clones arrayed in 104 384-well microtiter plates with an average insert size of 117 kb and estimated coverage of 12 genome-equivalents.;Chromosome walking initiated with eight RAPD markers closely linked to resistance, and extended with two further rounds of walking identified a total of 93 BACs in the resistance region. A high resolution genetic map of the resistance region was created with 51 markers in a mapping population of 1488 seedlings. In parallel, a physical map was constructed. Analysis indicated that the resistance gene is located in a single contig of three BACs (43F13, 66C22 and 85B7).;Whole BACs identified in the resistance region (< 1cM) were sequenced using an Illumina IIx genome analyzer, with multiplexing and barcoded adapters to reduce the cost, and paired-end reads to facilitate de novo sequence assembly. De novo sequence assembly was carried out using the programs Velvet and SOPRA, and the resulting contigs were further aligned using CodonCode software, and generated contig length ranged from 356 bp to 99632 bp. Estimated coverage of the BACs ranged from 64 to 100%. The gene prediction program AUGUSTUS identified 233 genes from these sequences using Arabidopsis as the model. Of these, RNA-Seq data supported 73 genes at a 60% cutoff and 32 at 100% support. The predicted gene sequences were compared with sequences in GenBank using a BLASTP search (NCBI) and identified two putative genes encoding a p-loop NTPase and F-box super family. Genes in these two superfamilies have defense response properties. Future complementation and mapping studies are essential to confirm which gene confers resistance.