O objetivo deste trabalho foi identificar marcadores moleculares relacionados à resistência do cafeeiro (Coffea arabica) à ferrugem (Hemileia vastatrix). Foram identificadas sequências de DNA potencialmente envolvidas na resistência do cafeeiro a doenças, por meio de análise "in silico", a partir das informações geradas pelo Projeto Brasileiro do Genoma Café. A partir das sequências mineradas, foram desenhados 59 pares de iniciadores para amplificá-las. Os 59 iniciadores foram testados em 12 cafeeiros resistentes e 12 susceptíveis a H. vastatrix. Vinte e sete iniciadores resultaram em bandas únicas e bem definidas, enquanto um deles amplificou fragmento de DNA em todos os cafeeiros resistentes, mas não nos suscetíveis. Esse marcador molecular polimórfico amplificou uma região do DNA que corresponde a uma janela aberta de leitura parcial do genoma de C. arabica que codifica uma proteína de resistência a doenças. O marcador CARF 005 é capaz de diferenciar os cafeeiros analisados em resistentes e susceptíveis a H. vastatrix.

The objective of this work was to identify molecular markers related to the resistance of coffee (Coffea arabica) to rust (Hemileia vastatrix). DNA sequences potentially involved in coffee disease resistance were identified, using "in silico" analysis, from data obtained by the Brazilian coffee genome project. After data mining, 59 primer pairs were designed to amplify the sequences identified. The 59 primers were tested on 12 resistant and 12 susceptible coffee plants to H. vastatrix. Twenty-seven primers resulted in unique and well-defined bands, while one of these amplified a DNA fragment in all resistant plants, but not in the susceptible ones. This polymorphic molecular marker amplified a region of DNA that corresponds to a partial open reading frame of C. arabica genome that encodes a disease resistance protein. The marker CARF 005 can be used to differentiate between resistant and susceptible coffee plants to H. vastatrix.

Sequences potentially associated with coffee resistance to diseases were identified by in silico analyses using the database of the Brazilian Coffee Genome Project (BCGP). Keywords corresponding to plant resistance mechanisms to pathogens identified in the literature were used as baits for data mining. Expressed sequence tags (ESTs) related to each of these keywords were identified with tools available in the BCGP bioinformatics platform. A total of 11,300 ESTs were mined. These ESTs were clustered and formed 979 EST-contigs with similarities to chitinases, kinases, cytochrome P450 and nucleotide binding site-leucine rich repeat (NBS-LRR) proteins, as well as with proteins related to disease resistance, pathogenesis, hypersensitivity response (HR) and plant defense responses to diseases. The 140 EST-contigs identified through the keyword NBS-LRR were classified according to function. This classification allowed association of the predicted products of EST-contigs with biological processes, including host defense and apoptosis, and with molecular functions such as nucleotide binding and signal transducer activity. Fisher's exact test was used to examine the significance of differences in contig expression between libraries representing the responses to biotic stress challenges and other libraries from the BCGP. This analysis revealed seven contigs highly similar to catalase, chitinase, protein with a BURP domain and unknown proteins. The involvement of these coffee proteins in plant responses to disease is discussed.