Analysis of resistance gene analogs in 'Calcutta 4'

Commercial banana varieties are cultivated in over 120 countries, generating an annual global production of around 100 million tonnes. Brazil represents the third largest producer, with a production area of 560 000 hectares and an annual production of 6 million tonnes. In 1998, the fungal pathogen Mycosphaerella fijiensis, the causal organism of black leaf streak disease, which causes premature fruit ripening, necrotic lesions leading to leaf area decomposition and yield losses of up to 50%1 was reported for the first time in Brazil in the Amazon region2. Subsequently, the pathogen has spread to 6 states in the north of the country. Specific recognition of plant pathogens, leading to programmed cell death, is controlled by resistance genes (R-genes). A number of R-gene subfamilies frequently possess conserved NBS (nucleotide binding site) and LRR (leucine rich repeat) domains3. The objectives of this study were to identify sources of resistance in 'Calcutta 4' through the analysis of resistance gene analogs (RGAs) using degenerate primers designed from highly conserved amino acid domains (NBS and LRR), and later via full length sequencing of positive BAC clones. To date, specific PCR amplification from genomic DNA targeting P-loop and GLPL amino acid motifs within the NBS domain has generated 15 sequence clusters that, based on BlastX analysis with a non-redundant NCBI protein database, show homology to resistance genes and RGAs from Musa acuminata, Oryza sativa, Elaeis guineensis, Theobroma cacao, and Arabidopsis thaliana. To date, PCR amplification, using a second set of degenerate primers designed from conserved amino acid motifs within non-Tir type NBS (P-loop, kinase 2, RNBS, GLPL) and LRR domains from monocot protein databases, has resulted in amplification of PCR products of expected size that are currently being sequenced. All RGAs will subsequently be characterized via reverse northern blots, in order to determine whether they represent gene regions, and candidate sequences will be labelled for use as probes against the 'Calcutta 4' BAC library. (Texte intégral