Inheritance of resistance to Magnaporthe grisea in upland rice

The main objective of this study were: 1) to determine the pathogenicity of rice blast isolates collected in three key rice producing districts of Uganda on standard host differentials, 2) to estimate the heritability, mode of gene action and gene number of blast resistance in F2 populations. To do this, a survey was carried out in the rice growing districts to determine the prevalence of rice blast disease caused by Magnaporthe grisea. Using a set of twelve differentials and eight parental lines, isolates collected during the survey were characterised. The experiments were carried out in a screen-house using artificial inoculation. To determine the combining ability, heritability and number of genes controlling rice blast resistance in segregating populations, a half-diallel scheme was employed. F2 progenes and their parents were artificially inoculated with the isolate 1 (Race B) in an 8 by 5 Alpha-lattice design with 4-replications. The results revealed that rice blast disease incidence and severity varied from one district to another. Beside the blast disease, rice farmers were also faced with other production constraints including Rice Yellow Mottle Virus (RYMV), lodging, drought, and nitrogen stresses. Thus, varieties grown in these districts require incorporation of multiple traits for their efficient use. Three isolates collected from Eastern and Central districts that were successfully isolated into pure cultures, were classified as races A, B and C following the procedures, of Inukai et al. (1994). However, further analysis using molecular tools will need to be employed to validate these findings. Griffing's diallel analysis showed that broad sense heritability was intermediate (0.57) and narrow sense heritability was low (0.17), indicating that non-additive effects were more predominant (71%) than additive effects (29%) in controlling the resistance of rice blast in F2 progenes; suggesting that selection can be conducted in later generations for improving rice blast resistance. Both general combining ability (GCA) and specific combining ability (SCA) effects were significant, indicating that genetic improvement can be achieved by selection followed by inter crossing. Parental lines IR64, NERICA 10 and Koshihikari showed a negative and significant GCA, they were considered as good combiners for rice blast resistance. Chi square analysis and frequency distributions showed that blast resistance could be explained by 1 or 2 major genes. Resistance genes were recessive for most of crosses and progenies segregated in accordance with the ratio 1:3 (R:S), and 1:15 (R:S), suggesting that the gene action for nine crosses and duplicate dominant epistasis for blast susceptibility in thirteen crosses. Further studies should focus mapping of additional race structure of the pathogen in Uganda with collection, isolation and characterization of many isolates. Molecular analysis will be a useful tool to validate such race patterns. Further studies to understand the mode of gene action for resistance to rice blast using better techniques, more isolate collections and more sources of resistance are recommended.