Identification of a major QTL for adult plant resistance to coffee leaf rust (Hemileia vastatrix) in the natural Timor hybrid (Coffea arabica x C. canephora) G LADYS R OMERO 1 ,L UISA M. V  A SQUEZ 1 ,P HILIPPE L ASHERMES 2 and J UAN C. H ERRERA 1,3 1 Centro Nacional de Investigaciones de Caf e, CENICAFE-FNC, Reserva Plan Alto, AP 2427 Manizales, Caldas, Colombia; 2 Centre IRD, UMR RPB (IRD, CIRAD, Universit e Montpellier II), BP 64501, 34394 Montpellier C edex 5, France; 3 Corresponding author, E-mail: juanc.herrera@cafedecolombia.com Received April 25, 2013/Accepted September 19, 2013 Communicated by T. Debener Abstract Most of the commercial varieties of coffee (Coffea arabica L.) derived from the Timor hybrid (TH) have been shown to contain major genes for coffee leaf rust (CLR) resistance. To identify markers tightly linked to such genes, an F 2 mapping population derived from a cross between ‘Caturra’ (susceptible variety) and the TH-derived DI.200 line (highly resistant) was generated. Using expressed sequence information and a bioinformatics approach, both targeted region amplified polymorphism (TRAPs) markers and simple sequence repeat (SSR) markers were identified. Phenotypic evaluations in the field and under controlled con- ditions confirmed the existence of one quantitative trait locus for CLR resistance. Four candidate SSR markers were associated with high CLR resistance. They spanning a region of 2.5 cM designated Q CLR_4 located within chromosome 4 of the international C. canephora map. The presence of this region was confirmed in a set of elite lines and commercial varieties. The Q CLR_4 region corresponds to a new and genetically independent S H locus that could potentially be useful in gene pyramiding with other genes to enhance rust resistance in TH derivatives. Key words: QTL mapping — quantitative resistance — targeted region amplified polymorphism — expressed sequence tags — orange rust Coffee is considered as one of the world’s favourite beverages, the second most traded commodity after oil, and one crucial to the economies of several countries particularly in Latin America. Only two species are responsible for commercial production: Coffea arabica (Arabica coffees) and C. canephora (Robusta coffees). All coffee species are diploid, except C. arabica, which is allotetraploid (2n = 4x = 44) and derived from a recent (<50 000 years ago) interspecific hybridization between two dip- loid species: C. eugenioides and C. canephora (Lashermes et al. 1999, Cenci et al. 2012). Homoeologous genomes in C. arabica are designated E a and C a according to their parental origin. Although parental subgenomes exhibit low sequence divergence, C. arabica displays a diploid like meiotic behaviour (Lashermes et al. 2000). Plantations of C. arabica around the world are affected by several diseases, among them the most important is the coffee leaf rust (CLR), caused by the biotrophic fungus Hemileia vasta- trix Berk & Br, which is also considered to be the most devastat- ing disease for this culture. Although CLR can be controlled through the application of fungicides, resistance breeding is a more economical and environmentally friendly approach to con- trolling leaf rust. To prevent the spread of this disease, different breeding pro- grammes for rust resistance were initiated in many countries since 1970. Thanks to the high levels of rust resistance exhibited by a spontaneous interspecific hybrid between C. arabica and C. canephora, known as the Timor Hybrid, TH, several con- trolled crosses with susceptible C. arabica varieties, such as ‘Caturra’, ‘Villa Sarchi’ or ‘Catuai’, have resulted in the devel- opment of highly resistant and productive cultivars in Latin America, Africa, Asia and Oceania (Bettencourt 1981, Rodrigues et al. 2000). More recently, CLR disease is becoming more common in the whole altitudinal range of coffee-producing areas, mostly due to climatic variations that have favoured increasingly humid condi- tions (L€ aderach et al. 2010). Indeed, a devastating epidemic of CLR that affected those coffee regions planted with the suscepti- ble ‘Caturra’ variety occurred in 2008–2011 in Colombia, and more recently in some Central American countries (Cristancho et al. 2012, Cressey 2013). Several reports around the world have revealed the adaptive capacity of CLR disease that has resulted in the gradual loss of field resistance in most of the cultivated varieties. Race II was, for many years, the most prevalent rust race detected in the cof- fee-growing areas in Africa, Asia and America. Its predominance was attributed to the genetic uniformity of the majority of com- mercial varieties of C. arabica prevalent throughout these culti- vated areas (Rodrigues et al. 1975, Kushalapa and Eskes 1989). Today, race II frequency seems to have decreased considerably after the deployment of the first resistant varieties (e.g. ‘Iapar59’, ‘Colombia’, ‘Cauvery’, among others) around the world. Conse- quently, race-specific resistance showed by most of the ‘Caturra’ x TH derivatives has become less and less durable in the field because of the emergence of new compatible races favoured by their spatial and temporal permanence in the field and conditions that stimulate selection pressure (Rodrigues et al. 2000, Alvarado 2005, V arzea and Marques 2005). To date, at least five major resistance genes (S H 5 to S H 9) have been identified in the TH accessions (Bettencourt 1981). All of them seem to be race-specific and single-dominant genes, although their molecular characterization remains to be com- pleted. Over the last ten years, molecular studies on rust resis- tance have developed markers that are closely linked to some CLR resistance genes. For example, studying the S H 3 gene from C. liberica species, Prakash et al. (2004) and Mah e et al. (2008) developed the first locus-specific markers tightly linked to this region, while Lashermes et al. (2010) undertook its genetic and wileyonlinelibrary.com Plant Breeding 133, 121–129 (2014) doi:10.1111/pbr.12127 © 2013 Blackwell Verlag GmbH