Enhancing adaptation of tropical maize to temperate environments using genomic selection

Authors: CHOQUETTE, NICOLE - North Carolina State University; WELDEKIDAN, TECLEMARIAN - University Of Delaware; Brewer, Jason; DAVIS, SCOTT - University Of Delaware; WISSER, RANDALL - Inrae; Holland, Jim - Jim

Submitted to: G3, Genes/Genomes/Genetics
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/14/2023
Publication Date: 6/27/2023
Citation: Choquette, N., Weldekidan, T., Brewer, J.C., Davis, S., Wisser, R., Holland, J.B. 2023. Enhancing adaptation of tropical maize to temperate environments using genomic selection. G3, Genes/Genomes/Genetics. 13(9):jkad141. https://doi.org/10.1093/g3journal/jkad141.
DOI: https://doi.org/10.1093/g3journal/jkad141

Interpretive Summary: Tropical maize can contribute useful genes to temperate maize breeding programs, but the photoperiod-sensitive late-flowering nature of most tropical maize hinders breeders from accessing these genes. Here we evaluated the potential of genomic selection to leverage flowering time evaluations in long daylength environments to aid selection in a short daylength 'offseason' environment where photoperiod-sensitivity is not expressed. Empirical results demonstrate this approach can be moderately effective and could contribute to faster adaptational selection.

Technical Abstract: Tropical maize can be used to diversify the genetic base of temperate germplasm and help create climate-adapted cultivars. However, tropical maize is unadapted to temperate environments, in which sensitivities to long photoperiods and cooler temperatures result in severely delayed flowering times, developmental defects, and little to no yield. Overcoming this maladaptive syndrome can require a decade of phenotypic selection in a targeted, temperate environment. To accelerate the incorporation of tropical diversity in temperate breeding pools, we tested if an additional generation of genomic selection can be used in an off-season nursery where phenotypic selection is not very effective. Prediction models were trained using flowering time recorded on random individuals in separate lineages of a heterogenous population grown at two northern U.S. latitudes. Direct phenotypic selection and genomic prediction model training was performed within each target environment and lineage, followed by genomic prediction of random intermated progenies in the off-season nursery. Performance of genomic prediction models was evaluated on self-fertilized progenies of prediction candidates grown in both target locations in the following summer season. Prediction abilities ranged from 0.30 to 0.40 among populations and evaluation environments. Prediction models with varying marker effect distributions or spatial field effects had similar accuracies. Our results suggest that genomic selection in a single off-season generation could increase genetic gains for flowering time by more than 50% compared to direct selection in summer seasons only, reducing the time required to change the population mean to an acceptably adapted flowering time by about one-third to one-half.