The continued expansion of the world population and the likely onset of climate change combine to form a major crop breeding challenge. Genetic advances in most crop species to date have largely relied on recombination and reassortment within a relatively narrow gene pool. Here, we demonstrate an efficient wheat breeding strategy for improving yield potentials by introgression of multiple genomic regions of de novo synthesized wheat. The method relies on an initial double top-cross (DTC), in which one parent is synthetic hexaploid wheat (SHW), followed by a two-phase selection procedure. A genotypic analysis of three varieties (Shumai 580, Shumai 969 and Shumai 830) released from this program showed that each harbors a unique set of genomic regions inherited from the SHW parent. The first two varieties were generated from very small populations, whereas the third used a more conventional scale of selection since one of bread wheat parents was a pre-breeding material. The three varieties had remarkably enhanced yield potential compared to those developed by conventional breeding. A widely accepted consensus among crop breeders holds that introducing unadapted germplasm, such as landraces, as parents into a breeding program is a risky proposition, since the size of the breeding population required to overcome linkage drag becomes too daunting. However, the success of the proposed DTC strategy has demonstrated that novel variation harbored by SHWs can be accessed in a straightforward, effective manner. The strategy is in principle generalizable to any allopolyploid crop species where the identity of the progenitor species is known.
