Molecular evolution and genetic control of flowering in the Eucalyptus globulus species complex

Eucalyptus globulus (blue gum) is one of the most widely planted eucalypts in temperate parts of the world, including its own natural range in south-eastern Australia. Australia is therefore a custodian of an internationally important genetic resource and it is necessary to develop population-based strategies to manage this gene pool. In addition, an understanding of the flowering process in E. globulus has key implications in managing seed orchards, developing early-flowering trees to accelerate breeding programs, generating sterile genetically modified trees, and determining the risk of genetic pollution. This thesis consists of, firstly, a population genetic analysis of the E. globulus species complex and, secondly, an analysis of the flowering process in E. glob ulus, including quantitative genetics and gene expression analyses. Eucalyptus globulus is a species complex consisting of four taxa variously described as species or subspecies (bicostata, globulus, maidenii and pseudoglobulus). The cores of these taxa are geographically and morphologically distinct, but are linked by intergrade populations that are intermediate in morphology. The intergrade populations represent a significant proportion of the distribution of the species complex, but their diverse and intermediate morphologies confound taxonomic classification. To assess the genetic affinities within the complex, nine nuclear DNA markers (microsatellites) were used to genotype 1198 trees representing 33 morphological core and intergrade populations from across the natural range of the species complex in south-eastern Australia. Combined with morphological measurements, this analysis provided insights into some of the evolutionary processes that have shaped the patterns of genetic variation in the E. globulus gene pool, including drift and morphological convergence, and indicated that the patterns of variation in the complex are probably due to primary differentiation rather than recent secondary contact. The analysis also identified the possible geographic origin of the species complex, evaluated the conservation status of small, isolated populations and provided a framework to guide the development of seed transfer guidelines and identify the potential impact of gene flow from E. globulus planted within the natural range of the complex. Variation in flowering season is clearly a significant barrier to gene flow within the E. globulus complex. Surveys of E. globulus ssp. globulus provenances grown in a common environment near Hobart (Tasmania), conducted for the last six years, showed that flower bud opening occurred over a nine month period and was under strong genetic control (H2 = 0.69) with highly significant differences in flower opening time among provenances. Surprisingly though, provenances did not vary in the timing of macroscopic appearance of flower buds, which occurred in October (spring) in all provenances (H2 = 0.06). \Early\" flower opening genotypes therefore had a shorter flower bud development time than the flower buds of \"late\" flower opening genotypes. The E. globulus homologues of FLOWERING LOCUS T (F7) TERMINAL FLOWER] (TEL]) SUPPRESSOR OF OVEREXPRESSION OF CONSTANS1 (SOC1) LEAFY (LFY) APETALA1 (AP1) SHORT VEGETATIVE PHASE (SVP) SEPALLATA (SEP) and PISTILLATA (P1) were isolated and quantitative RT-PCR was used to show that the expression of the homologues of FT and LFY in the leaf and apex of E. globulus were associated with the annual transition from vegetative to reproductive growth (i.e. flower bud initiation) over a two year period. However there was no strong seasonal pattern of expression of TFL1 SOC1 API SVP SEP or PI in leaves apices and/or young flower buds. In a comparison of FT and LFY expression patterns in two clones each of an early and late flowering genotype no association between the expression of these genes and the timing of flower bud opening was shown. This indicates that FT and LFY could form part of the flower initiation pathway in Eucalyptus but do not regulate the observed differences in flower bud opening time."