Creating a new allohexaploid Brassica crop:determination of fertility and stability in novel Brassica hybrids

Abstract

The Brassica genus is an economically important group of diploid (single genome) and allotetraploid (two-genome) species used as oilseeds, vegetables and condiments. Although no allohexaploid (three-genome) Brassica species exists in nature, the production of an allohexaploid Brassica crop could be beneficial for agriculture. However, production of a stable and fertile Brassica allohexaploid is currently hindered by challenges of infertility and unstable meiosis. To date, incremental progress has been made in creating an allohexaploid species using several different species combinations. Despite this, a gap in knowledge currently exists for genomic stability among Brassica allohexaploids, critical in the establishment of a successful species. Firstly, I investigated fertility and meiotic stability in 100 plants from the cross B. carinata × B. rapa (A2 allohexaploid population) and 69 plants from the cross (B. napus × B. carinata) × B. juncea (H2 allohexaploid population). Estimated pollen viability, self-pollinated seed set, number of seeds on the main shoot, number of pods on the main shoot, seeds per 10 pods and plant height were measured for both the A2 and H2 populations and a set of reference control cultivars. The H2 population had high segregation for pollen viability and meiotic stability, while the A2 population was characterised by low pollen fertility and a high level of chromosome loss. Both populations were taller, but had lower average fertility trait values, than the control cultivar samples. Additionally, I established that the genotypes of the parents and H1 hybrids are affecting chromosome pairing and fertility phenotypes in the H2 population. Next, I investigated fertility, meiosis and genetic variability in sets of self-pollinated progenies (the MDL2 population) resulting from first generation microspore-derived plants (the MDL1 population). These populations were derived from microspores of a near-allohexaploid interspecific hybrid plant from the cross (Brassica napus × B. carinata) × B. juncea. Fertility decreased from MDL1 to MDL2 population, with several fixed chromosome duplications and deletions present as well as novel genomic events seen to have occurred from the MDL1 to MDL2. Genetic non-identity between lines within various progeny sets in MDL2 was also observed, uncharacteristic of what would be expected of a microspore-derived (normally doubled haploid; DH) population. iii Finally, I made novel interspecific hybrids between wild C genome species, B. oleracea and B. juncea to utilise this germplasm for the creation of a diverse allohexaploid species. Hand pollinations between two genotypes of B. juncea (Xingyou 4 and B578) and B. oleracea (TO1000) and wild C genome species B. incana, B. montana and B. cretica were performed (747 total bud pollinations, average 62.25 per cross combination) in both cross directions. The combination with B. oleracea produced six triploid hybrids (2n = ABC) from 85 flower pollinations. Pollen fertility in triploid hybrids was low; between 2 -10% (averag...