Genomics of multiple-insect resistance in maize
MetadataShow full item record
Maize, a food security crop, is attacked from the fields to the storage facilities by a plethora of insect pests on all parts of the plant. Most studies that were conducted to understand maize defense mechanisms to incorporate resistance in new varieties focused on single insect attacks. The phenotypic correlations and co-localizations of quantitative trait loci (QTL) for maize resistance to stem borers (SB), the fall armyworm (FAW), and storage pests (SP) on the one hand, and with cell wall constituents (CWC) and benzoxazinoids, on the other hand, suggested the existence of single and multiple-insect resistance (MIR) mechanisms. Therefore, this research aimed to investigate the feasibility of MIR in maize through the application of genomic-aided breeding (GAB) technologies for mapping and selection by establishing the genetic architecture and genomic predictability of maize resistance to multiple insect pests. A QTL meta-analysis for resistance to stem borers, FAW, and storage pests along with CWC and Benzoxazinoids revealed 42 stem (SIR), 24 leaf (LIR), and 20 kernel (KIR) insect resistance meta-QTL (MQTL) with CWC and Benzoxazinoids involved in most, suggesting a defensive role played by CWC and BXs in resistance. A genome-wide association analysis conducted on a diverse association mapping panel (AMP) planted in three locations during three seasons and genotyped with 34509 diversity array technology (DArT) single-nucleotide polymorphism (SNP) markers revealed sixty-two quantitative trait nucleotides (QTNs) for resistance to FAW and MW, majorly located within or near the insect resistance genomic regions (IRGRs). Seventy-nine genes were located around the QTNs, and 64 were functionally related to plant defense mechanisms through an array of In-silico interspecific comparative functional analyses. A network-based candidate gene identification that used these 64 genes as guides identified an additional set of 107 candidate genes differentially expressed under various stress conditions located within the IRGRs. These MQTL, QTNs, and genes could be incorporated in molecular breeding or gene editing and transgenic programs targeting single or combined insect resistance. Also, benchmarking of genomic prediction (GP) strategies was conducted using 17 GP algorithms, and two training (TS) and breeding set (BS) determination approaches were performed for FAW and MW resistance traits. The prediction accuracies (PA) were high for FAW (up to 86%) and MW (up to 82%) resistance traits, even with moderately-sized TS. The GP models performed differentially on both FAW and MW resistance traits. However, the determination of the TS size and its relationship with the BS were more influential. These results give hope for the application of GS in breeding for resistance to insect pests in Africa.