Development of molecular markers for introgression of resistance to turcicum leaf blight in Sorghum

Abstract

Sorghum {Sorghum bicolour (L.) Moench (2n=2x=20)}, a C4 grass that diverged from maize about 15 million years ago, is the fifth major cereal crop in the world after wheat, rice, maize and barley. It has relatively small genome of 750 million base pairs. Sorghum production especially in the tropics is affected by several pests and diseases. Turcicum leaf blight (TLB) caused by the pathogen Exserohilum turcicum (Pass) K.J. Leonard and E.G. Suggs (teliomorph: Setosphaeria turcica [Luttrell] Leonard and Suggs) is one of the threats to sorghum production. It is one of the most destructive foliar diseases of sorghum. Development of resistant varieties is the most economically viable solution for disease management for cereals in general. However, the design of the well targeted disease management strategies that involve deployment of resistant genotypes requires detailed characterisation of a pathogen’s pathosystem. E. turcicum attacks both maize and sorghum. The maize Exserohilum turcicum pathosystem has been characterised, and host species specialisation occurs. However, a comprehensive review of the published literature shows that sorghum resistance to TLB has received limited research attention and the E. turcicum - pathosystem has limited studies. The objectives of this study were to (1) determine the mode of TLB inheritance in sorghum; (2) develop and validate SSR and RAPD markers linked to the TLB resistance loci and (3) use the polymorphic SSR markers to map QTL for resistance in sorghum to TLB. The study was carried out in Uganda at Makerere University Agricultural Research Institute Kabanyolo (MUARIK). Three populations derived from a cross of MUC007/009 (resistant) and Epuripuri (susceptible) an elite sorghum variety were used together with two parents and four checks GAO6/106 (Moderately resistant), Lulud (Susceptible), MUC007/010 (Resistant) and GAO6/18 (Moderate Susceptible). A total of 304 F2 segregating population, 278 F2:3 and 246 F2:4 segregating families were used. The experiments were set up following a completely randomised design with no replication to evaluate F2 and F2:3 and alpha lattice design to evaluate F2:4 population. Generation mean analysis was used to determine the contribution of additive, dominant and epistatic genetic effects and also to confirm the genetic ratio analysis for the population distribution under a greenhouse and field conditions. Disease severity was assessed using percentage of leaf area affected on individual plant basis using a scale of 0 to 75, where 0%= no disease and >75 % of leaf surface diseased. Assessment commenced at stage 4 (the growing point differentiation) 51 days after planting and continued on a weekly basis of disease severity and they were used to compute area under disease progress curves (AUPDC). To standardise area under disease progress curve the AUDPC, values were divided by the total period of epidemics. Data were subjected to analysis using GenStat Discovery Edition 12 to establish any association between AUDPC disease severity, lesion type and dates to flowering. Chi square (χ2) analysis was used to test goodness of fit of the mode of TLB inheritance data to expected segregation ratios. Disease severity of F2 plants in the greenhouse condition indicated a normal distribution indicative of quantitative inheritance or minor gene effects. Under the field conditions, disease severities of F2:3 and F2:4 matched a normal distribution also suggesting quantitative inheritance. Though the performance of the resistant parent MUC007/009 and the susceptible parent Epuripuri was not different under the greenhouse environment, it was highly significant different (P<0.001) under the field conditions. There was transgressive segregation towards the resistance under both environments for F2, F2:3 and F2:4 progenies. However all populations (F2, F2:3 and F2:4) from this cross, showed negative correlation between flowering dates and AUDPC. The early maturity lines had higher disease severity. In this study there was a clear difference between greenhouse and field environments. Similar reports have been made elsewhere. In this study the resistant parent MUC007/009 and the susceptible parent Epuripuri expressed distinctly different lesion types under both greenhouse and field environments. The resistant lesion type and the susceptible lesion type were use to screen the F2:3 and F2:4 families. The two distinct lesion types segregated according to the 1:2:1 ratio indicative of dominant gene inheritance. Partitioning of genetic effects into additive, epistatic and dominance components in this study shows that this type of resistance is attributed to additive and epistatic effects. These data are consistent with other studies in maize which also show that resistance to E. turcicum is quantitative in nature. The limited role of dominance effects under both greenhouse and field environments further demonstrates the bigger role of additive and epistatic effects. This study involved screening eighty random amplified polymorphic DNA (RAPD) and ten simple sequence repeat (SSR) markers for polymorphism between the two distinct parents. Eight RAPD markers were polymorphic between the resistant parent MUC007/009 and the susceptible parent Epuripuri. These markers are recommended to be used for mapping of resistance to sorghum TLB. Out of ten SSR markers, three were polymorphic between the resistant parent MUC007/009 and the susceptible parent Epuripuri. The three polymorphic SSR markers are distributed along the sixth linkage group of sorghum consensus map. However the SSR marker Xtxp95 showed high significant association between the allelic groups and the resistant lesion type, while the SSR markers Xtxp57 and Xtxp247 did not show any significant association. The polymorphic information content (PIC) for SSRs was 0.490 (Xtxp57), 0.496 (Xtxp247) and 0.499 (Xtxp95). The SSR marker Xtxp95 was linked to the resistant lesion type and the analysis showed that 23.74% in F2:3 and 14.09% in F2:4 of the variability in the resistant lesion type is associated with the SSR marker Xtxp95 segregation. The results from this study show that the genomic region flanked by plant colour locus and Xtxp95 marker may harbour a locus for sorghum TLB lesion type.