Analysis of genetic variability of grain mould resistance in grain sorghum

[摘要] English: In an effort to characterize the relationship between fungal pathogenicity and host genetic resistance, sorghum panicles of 11 genotypes were inoculated with five fungi frequently isolated from sorghum grain. Panicles were inoculated at anthesis with Fusarium graminearum, Fusarium thapsinum, Curvularia lunata, Phoma sorghina and Alternaria alternata spores. There were highly significant differences in the levels of fungal pathogenicity on different sorghum genotypes. These differences accounted for 58.4% of observed variation in ergosterol concentration. Genotype by pathogen (G x P) interactions accounted for 33.5% of the observed ergosterol concentration variation. The implication is that different genotypes reacted differently to different fungi. The genotypic reactions of the hosts accounted for 8.1% of the observed ergosterol concentration variation. Overall, fungal pathogenicity is the most important factor to consider in the evaluation of germplasm for grain mould resistance. Possible sources of resistance could be identified by use of biplot analysis of G x P interactions. Visual scoring for grain mould has limited value without identifying causal fungi. Fusarium thapsinum and Phoma sorghina were the most abundant fungi across all genotypes. Gene action and heritability for grain mould resistance in sorghum were investigated using a selection of 9 random pollen parents with varying levels of grain mould resistance to a different set of three random seed parents. Differences in ergosterol concentration were used as a measure of level of grain mould resistance among all genotypes. Use of ergosterol concentration as a measure of grain mould severity did not correlate with visual field scoring. Other traits measured include plant yield, plant height, kernel hardness, field grade score, days to flowering, glume color and seed color. The combined analysis of variance showed no genotypic variance for grain mould resistance. The expression of grain mould resistance was also not stable with significant genotype x location interaction. The analysis from Potchefstroom showed significant differences among genotypes whereas Cedara-1 and 2 showed no differences. Additive genetic variance was greater than dominance variance for all traits except grain mould resistance. A significant heterosis of -20.15% was observed for grain mould resistance indicating the importance of use of hybrid seed. LM124 (white seeded female) and LM130 (brown seeded female) produced resistant hybrids. Due to very high environmental variance, grain mould heritability could not be detected. Yield indicated the highest heritability of 0.41 at Cedara-1, plant height 0.39 at Cedara-2 and kernel hardness 0.38 at Potchefstroom. The significance of genotype by environment interaction was assessed over three environments. Differences in ergosterol concentration were used as a measure of level of grain mould resistance among all genotypes. Significant G x E interaction was detected after analysis of variance across all three locations. Single site analysis was then done to better explain the nature of the G x E interaction. Potchefstroom is the only location that showed significant genotypic responses to grain mould infection. Low grain mould pressure at Cedara-1 and Cedara-2 caused very low genotypic responses. A biplot was then used to indicate all genotypic performances across the three locations in a graphical design. Fungal species infecting sorghum kernels were isolated, counted and identified. Low genotypic responses at the Cedara locations seems to have been caused by prevalence of a less aggressive fungal species, Mucor spp. Weather variables did not correlate significantly with ergosterol concentration. Mycotoxin levels of aflatoxin, deoxyvalenol (DON), and zearalenone together with ergosterol concentration levels across 39 sorghum genotypes were measured. Varying levels of ergosterol and mycotoxins were observed across the three locations. There was no correlation between ergosterol concentration and any of the mycotoxins which indicates that mycotoxins concentration is not related to total fungal biomass. Mycotoxin concentration must therefore be related only to the biomass of the fungal species that is producing the mycotoxin among all fungal species infesting the grain. Farmers in Potchefstroom should be wary of aflatoxins and zearalenone while farmers in Cedara should choose varieties that tend to be low in DON and aflatoxins while not ignoring zearalenone.

[发布机构] University of the Free State