STANLEY ADEKEMI ESTHER

BREEDING INVESTIGATIONS FOR STRIGA HERMONTHICA RESISTANCE IN TROPICAL MAIZE (ZEA MAYS L.) IN NIGERIA

ABSTRACT

Maize (Zea mays L.) is a major staple crop that contributes to food safety and also serves as channel of making money to subsistence farmers in sub-Saharan Africa (SSA). Nevertheless, its production is constantly hampered by a plethora of biotic stresses such as the parasitic weed, Striga. Striga hermonthica is an economically important parasitic plant that affects most cereals with maize being the most susceptible. The use of Striga resistant lines has been identified as the basic part of any successful integrated control approach for Striga parasitism. This study was carried out to (i) estimate the phenotypic variation among diverse maize inbred lines with different resistant levels to Striga using agronomic traits, estimate the general and specific combining abilities (GCA and SCA) for grain yield and other Striga resistance traits under Striga infestation, (iii) estimate the genetic variation of the same panel using molecular markers, (iv) identify genomic regions and putative genes associated with Striga resistance among the lines, and (v) design and sequence molecular makers associated with Striga hermonthica in maize using low Striga germination stimulant (lgs) gene in sorghum as baseline. One hundred and fifty diverse tropical inbred lines were assessed in Striga infested and noninfested environments in two sites in Nigeria for two years. The least-square means (LSmeans), best linear unbiased estimates (BLUEs) and heritability estimates of these lines were generated through mixed linear model (MLM) implemented in SAS. The performance and genetic diversity of these lines were assessed using ten agronomic traits and molecular marker (SNPs). In addition, 10 lines were selected from the 150 inbred lines to make single cross hybrids. Five lines (female parent) were crossed with five inbred (male parent) to generate twenty-five testcrosses. The testcrosses alongside three checks were assessed in Striga infested and non-infested environments in two locations in Nigeria for two years. Furthermore, association analysis was performed between SNP markers and BLUEs generated under both conditions using Genetic Association and Prediction Integrated Tools (GAPIT) with a false discovery rate (FDR) at 5%. Also, a comparative study was done to mine information on Striga hermonthica resistance gene in sorghum low germination stimulant (lgs), a close relative of Zea Mays. The results from the combined experiments showed significant (p < 0.001) genotypes and environment effects on all traits in both environments except for ear aspect. Under Striga, approximately 57% of the inbred lines stimulated the emergence of few Striga plants and yielded more than the resistant check while about 29% of inbred lines stimulated the growth of more Striga plants but yielded more than the susceptible check. The hierarchical cluster analyses based on agronomic traits clustered the lines into five distinct groups depending on their response to Striga parasitism. The combining ability study showed that the hybrids and environments were significant (p < 0.001) for all measured variables over the test environments. The general combining ability (GCA) mean square was significant for all traits indicating that additive gene action was crucial for the inheritance of the traits. Lines IWDS46 and IWDDS55 had significant and positive GCA effects and were regarded as outstanding combiners for grain yield. Also, two hybrids (IWDS46 x TZLC and IWDS55 x TZLC) outperformed the resistant and commercial checks by 119% and 239.7% respectively; these hybrids also stimulated few Striga emergence and moderate Striga damage symptom rating. The SNP markers also clustered the lines into five distinct groups depending on their historical background, genealogy and maturity classes. The genome-wide association analysis identified twenty-eight SNPs linked with the Striga resistance-related traits (Striga damage symptom rating and emergence count at 8 and 10 weeks after planting (WAP), grain yield, ear aspect and ears per plant) and 7 SNPs under non-infested condition. Each major SNP elucidate between 11 to 16% of the phenotypic variation. Gene annotation using the Maize genome data base (MaizeGDB) revealed the presence of twenty putative genes and three uncharacterized proteins with functions related to strigolactone production and plant defense. From the comparative study, one of the designed markers was able to differentiate between the selected resistant and the susceptible lines.The clustering pattern of the inbred lines will assist in the development and deployment of Striga resistant hybrids to Striga prevalent zones. The identified inbred lines with good GCA can also be used as potential parental lines for producing high yielding Striga resistant hybrids that can be deployed to Striga endemic zones. The identified SNPs associated with Striga resistance-related traits will provide insight for the use of marker-assisted selection in breeding for S. hermonthica resistance in maize. In addition, the identified marker from the comparative study will be validated in other populations and they can be used in Striga breeding programs.