Attempts to remove gametocidal genes co-transferred to common wheat with rust resistance from Aegilops speltoides

Abstract

Rust resistance genes (introgressions S24 and S13) transferred to hexaploid wheat from two Aegilops speltoides accessions could not be used commercially due to associated gametocidal (Gc) genes. Crosses to wheat followed by rigorous selection for increased fertility were employed in an attempt to separate the unmapped S24 stem rust resistance from the Gc gene(s). However, improved fertility of the better selections could not be maintained in subsequent generations. Since the S13 introgression (leaf, stripe and stem rust resistances) mapped to chromosome 3A, allosyndetic pairing induction was used in an attempt to remove the Gc gene(s). This produced putative primary recombinants with improved fertility and plant type, the best of which had exchanged a small region of Ae. speltoides chromatin, yet was still associated with (reduced) Gc effects. This selection (04M127-3, which appears to have the Su1-Ph1 suppressor) was then crossed with wheat. Surprisingly, the 04M127-3 gametocidal effect differed drastically from that of the original introgression allowing the recovery of 35 recombinant, leaf rust resistant progeny. Microsatellite and DArT markers showed that each secondary recombinant had exchanged most of the Ae. speltoides chromatin. Although the data suggested that a complex multigenic interaction may govern the gametocidal response, preliminary indications are that the Gc effect had largely been removed and it now seems possible to completely separate the gametocidal genes from the S13 leaf rust resistance gene (here designated Lr66). The associated (S13) stripe rust and stem rust resistance genes were lost during recombination.