Sexton, T 2011, 'Candidate gene SNP discovery, genotyping and association with wood quality traits in Eucalyptus pilularis (blackbutt)', PhD thesis, Southern Cross University, Lismore, NSW.
Copyright T Sexton 2011
This thesis uses association mapping to identify links between quantitative wood quality traits and Single Nucleotide Polymorphisms (SNPs) within wood formation genes. One priori expectation is that if a large panel of SNPs from candidate genes thought to influence associated wood properties are screened, some will be found that are associated with variation of phenotypes within the population and could be used for Marker assisted selection (MAS). Selecting trees based on their DNA sequence offers an alternative to traditional phenotypic selection for traits like wood quality, where breeding programs are largely constrained by the long generation times and the delay before measurements can be made reliably. Although MAS is being used in domesticated animals and non-woody crops, it is yet to be realised in forest trees where low linkage disequilibrium, high cost of marker development and implementation has hindered its uptake.
The focal species of this investigation was Eucalyptus pilularis Smith (blackbutt) subgenus Eucalyptus (informally Monocalyptus), and is one of the most important species for solid wood production in Australia. As part of the development of suitable SNP markers for E. pilularis, two other species were included for comparative purposes, a closely related Eucalyptus pyrocarpa (large fruited blackbutt) and the more distantly related Eucalyptus globulus Labill (southern blue gum). These species are representative of the two largest Eucalyptus subgenera the Monocalyptus and Symphyomyrtus respectively, which include the majority of commercially important species.
Initial methodologies designed to target candidate genes using polymerase chain reaction confirmed previous microsatellite studies, where primers developed in the commercial Symphyomyrtus species would amplify genes in species represented in the distantly related Monocalyptus subgenera. Analysis of SNPs shared between the closely related E. pilularis and E. pyrocarpa provided evidence that three specific wood formation genes that may play a role in local adaptation of these species to different environments. Comparative analysis of SNPs shared between the Monocalyptus and Symphyomyrtus subgenera found that between 11-13% of SNPs were shared (trans-subgeneric SNPs). These trans-subgeneric SNPs are likely to have persisted in separate lineages for tens of millions of years prior to the split of theses subgenera and may be indicative of variation important for adaptation, which has been maintained by "balancing selection". Such trans-subgeneric SNPs may be important for long-term resilience of the Eucalyptus genus to re-occurring biotic and abiotic stresses.
Association testing of two genes thought to be under balancing selection (pectin methylesterase 6 and 7) in E. pilularis revealed that SNPs in these genes could explain variation observed in a wide array of physical wood properties. This provides the first direct DNA based evidence that some wood formation candidate genes that control variation in quantitative traits, are also under balancing selection, and therefore could be of adaptive importance. Because of the large number of trans-subgeneric SNPs identified in these genes, many informative SNPs are also likely to transfer to other species. This will facilitate the development of SNP marker in other commercial species within the diverse Eucalyptus genus. Associations were also identified in a gene sharing few trans-specific SNPs, the MYB2 transcription factor, where two non-synonymous SNPs unique to E. pilularis, appeared to control variation in the dimensional stability of drying timber. This thesis demonstrates the power of association studies to identify correlations between genotype and phenotype, even where the individual effect of a SNP accounts for only a minor proportion of variance in the trait. Further association testing with SNPs identified in this project, and other association studies in Eucalyptus will yield larger panels of informative SNP markers, that together will explain a large proportion of the heritable variation of phenotypes and therefore enable the uptake of MAS in Eucalyptus breeding programs.