Exploiting genetic variation from unadapted germplasm-An example from improvement of sorghum in Ethiopia

Societal Impact Statement The productivity of sorghum in Ethiopia has been largely limited by rain-fed conditions because farmers tend to use local drought-tolerant but low-yielding landraces, as high-yielding and late-maturing landrace cultivars risk failure due to drought. Addressing such issues often requires a far-reaching approach to identify and incorporate new traits into a gene pool, followed by a period of selection to re-establish an overall adaptive phenotype. The sorghum backcross nested association mapping (BC-NAM) population developed in this study increases the genetic diversity available in Ethiopian elite adapted sorghum germplasm, providing new scope to improve food security in a region known for periodic devastating droughts. As the center of diversity for sorghum, Sorghum bicolor (L.) Moench, elite cultivars selected in Ethiopia are of central importance to sub-Saharan food security. Despite being presumably well adapted to their center of diversity, elite Ethiopian sorghums nonetheless experience constraints to productivity, for example, associated with shifting rainfall patterns associated with climate change. A sorghum backcross nested association mapping (BC-NAM) population developed by crossing 13 diverse lines pre-identified to have various drought resilience mechanisms with an Ethiopian elite cultivar, Teshale, was tested under three rain-fed environments in Ethiopia. Twenty-seven, 15, and 15 quantitative trait loci (QTLs) with predominantly small additive effects were identified for days to flowering, days to maturity, and plant height, respectively. Many associations detected in this study corresponded closely to known or candidate genes or previously mapped QTLs, supporting their validity. The expectation that genotypes such as Teshale from the center of diversity tend to have a history of strong balancing selection, with novel variations more likely to persist in small marginal populations, was strongly supported in that for these three traits, nearly equal numbers of alleles from the donor lines conferred increases and decreases in phenotype relative to the Teshale allele. Such rich variation provides a foundation for selection to arrive at a new "adaptive peak," exemplifying the nature of efforts that may be necessary to adapt many crops to new climate extremes.