Eucalyptus genomics: Linkage mapping, QTL analysis and population genomic studies

This thesis reports on genomic studies conducted in five of the most commercially important species of Eucalyptus, all from subgenus Symphyomyrtus; E. globulus, E. nitens (section Maidenaria), E. grandis, E. urophylla (section Latoangulatae) and E. camaldulensis (section Exsertaria). These studies utilised Diversity Arrays Technology (DArT), a microarray-based genotyping system which provides high-throughput genome-wide genotyping. DArT markers, and in particular linkage maps constructed with these markers, formed the basis for most of the research conducted in this thesis. The DArT markers are highly transferable across species which facilitates the transfer of information between studies and as sequences are available for most of the 7680 markers contained on the eucalypt genotyping array it is possible to place these markers on the recently released E. grandis genome sequence. A 1060 DArT and microsatellite marker linkage map was constructed in a large E. globulus dwarf x tall ecotype out-crossed F2 family (n = 503) in Chapter 1. This is the highest density linkage map produced for E. globulus and together with other high density DArT maps constructed in E. grandis x E. urophylla it was used to assess genome similarity (e.g. synteny and colinearity) between species through comparative mapping. Despite the detection of two small putative translocations or duplications in the inter-sectional comparison, these species showed very high synteny and colinearity overall. This finding showed that it is possible to link the economically important E. globulus to the E. grandis genome sequence. The feasibility of constructing a multi-species composite linkage map for Symphyomyrtus was explored in Chapter 2. Seven independently constructed linkage maps (built in either E. globulus or E. grandis x E. urophylla mapping families) were integrated into a single map containing a total of 4135 markers (3909 DArT markers, 218 microsatellite markers and 8 candidate genes). This composite map will, (1) serve as a valuable reference map for Symphyomyrtus and related species, and (2) aid comparative research by enabling the relative positions of markers and quantitative trait loci (QTL) from different studies to be determined more easily. Quantitative trait loci analyses were used to examine the genetic architecture underlying differences between dwarf and tall E. globulus ecotypes in Chapter 3; with a particular focus on vegetative and reproductive phase-change traits. A total of 22 significant and 11 suggestive QTL were detected for the nine traits examined. A highly significant QTL was detected for vegetative phase-change (LOD 103, 62.8% phenotypic variation explained). A micro-RNA gene known to regulate vegetative phase-change was found near the peak of this QTL. It is hypothesised that a mutation in, or a deletion of, this gene may be responsible for the heterochronic evolution of early phase-change in dwarf ecotype populations and that this has allowed E. globulus to inhabit the exposed coastal environments in which the dwarf ecotype occurs. In Chapter 4, population genomic analyses were conducted using range-wide samples (84 to 93 individuals per species) of E. globulus, E. nitens, E. grandis, E. urophylla and E. camaldulensis. Genetic diversity parameters were calculated using a common set of 2207 markers. In each of ten pair-wise species comparisons, markers under putative directional selection were identified using FST outlier analyses. These outlier markers along with genetic diversity estimates were mapped on the composite map (Chapter 2) to produce a ‘genomic atlas’. This identified several genomic ‘hot-spots’ of species molecular differentiation and together with other genetic resources, such as the E. grandis genome sequence and QTL positional information, offers the potential to identify candidate genes underlying traits associated with species adaptation and speciation