Characterisation of a diverse set of maize germplasm for resistance to infection by Aspergillus flavus.
Abstract
Sources of resistance to Aspergillus flavus and accumulation of aflatoxin in maize have been identified elsewhere but no such materials are currently available for use by the maize improvement programme in Uganda. The main objective of this study was to develop maize genotypes with resistance to kernel infection by A. flavus and with diminished accumulation of aflatoxins. The specific objectives were to: (1) Evaluate inbred lines, testers and testcrosses for resistance to infection by A. flavus and to accumulation of aflatoxins; (2) determine the general and specific combining ability associated with resistance to kernel infection by A. flavus and reduced accumulation of aflatoxin; and (3) determine the relationship between A. flavus infection and agronomic traits of inbred lines, testers and hybrids. Screening for resistance involved artificial field inoculation and measurement of infection by A. flavus in the laboratory by determining rate of kernel infection and quantifying the amount of aflatoxin accumulation in the grain. Measurement of kernel infection in this study was done in three ways: Percent-Kernel-Infection (PKI), the Incidence of Severely-Infected-Kernels and Percent-Severely-Infected-Kernels (PSIK). Evaluation for agronomic performance was done at three experimental sites.
Results indicated a highly significant variation (P ? 0.001) in the Percent-Kernel-Infection among inbred lines and testers while testcrosses showed marginally significant variability (p ? 0.05). The Incidence of Severely-Infected-Kernels (ISIK) and Percent-Severely Infected-Kernels (PSIK) were not significant among hybrids but significant differences were observed among inbred lines and testers (P ? 0.001).
Line x Tester analysis of combining ability indicated that inbred lines (male parents) had significant GCA (P ? 0.01) for Percent-Kernel-Infection (PKI) while testers showed significant GCA for both Percent- Severely-Infected-Kernels (PSIK) (P ? 0.01) and ISIK (P ? 0.05). SCA was significant (P ? 0.01) for the Incidence of Severely-Infected Kernels (ISIK). A variance component ratio (Baker's ratio > 0.5) revealed that GCA was more predominant than SCA for PKI while SCA as more predominant that GCA for both PSIK and ISIK (Baker's ratios < 0.5). Heritability estimates form variance components were low for PKI (h2 = 19%, H2 = 37%) and PSIK (h2 = 15%, H2 = 36%) but broad sense heritability was high for ISIK (h2 = 13%, H2 =57%).
Hybrids showed significant variability for all agronomic traits across locations except anthesis-silking interval. Few traits showed significant genotype by environment interactions (G x E). Genotype by environment (G x E) interaction was highly significant (P ? 0.001) for MSVD, TLB, cob rots and husk cover ear borer damage and mildly significant (P ? 0.05) for cob borer damage and cob aspect).
Correlation analysis showed a significantly positive relationship to PKI to aflatoxin content of grain (r = 0.47, P ? 0.05; r2 = 0.22). Correlations with cob traits showed that a more dent type grain texture was correlated with more kernel infection (ISIK, r = 0.51, P ? 0.001; and PSIK, r = 0.23, P ? 0.05). PKI did not show significant relationship to grain texture. More complete husk cover was not correlated with A. flavus infection indicators, but was associated with a lower percent of general ear rots (ER%, r=0.49, P ? 0.01).
The major conclusion of this study is that resistance to Aspergillus flavus is mainly controlled by additive gene action and heritability among the germplasm is low. Additionally, the male (inbred lines) parents used in the crosses had a higher general contribution to variation in resistance to the fungus and to performance in agronomic traits.