University of Tasmania
Browse
Whole_thesis_030112.pdf (2.3 MB)

Evidence for natural selection acting on genes affecting lignin and cellulose biosynthesis in Eucalyptus globulus

Download (2.3 MB)
thesis
posted on 2023-05-26, 02:22 authored by Hadjigol, S
Eucalyptus globulus (Myrtaceae) is a forest tree species that is native to South-eastern Australia, including the island of Tasmania. It is the main eucalypt species grown in pulpwood plantations in temperate regions of the world and is being domesticated in many breeding programs. The improvement of its wood properties is a major objective of these breeding programs. As many wood properties are expensive to assess, there is increasing interest in the application of molecular breeding approaches targeting candidate genes, particularly those in the lignin and cellulose biosynthesis pathways. To assist in the identification of genes and allelic variants likely to have important phenotypic effects, this study aimed to determine whether there was a signature in the genome indicating that natural selection had caused differentiation amongst the races of E. globulus in candidate genes for wood properties. Differentiation among races based on single nucleotide polymorphisms (SNPs) within candidate genes was compared to differentiation based on microsatellite (SSR) markers. The rationale behind this approach is that if the differentiation observed in the gene-related SNPs was significantly different from that based on putatively selectively neutral markers, then this is evidence that selection maybe affecting the candidate gene. In order to do this, the genetic affinities within E. globulus (368 trees representing 42 localities partitioned into eight races from across the natural range of the species) were studied using 30 gene-based SNPs and 18 neutral nuclear SSR markers. STRUCTURE analysis based on these SSR markers showed that individuals fell into two distinctive groups (lineages). One group comprised individuals from King Island and mainland races from the Otways and Strzelecki Ranges; the second group comprised all the Tasmanian races plus the Furneaux Islands. The pattern of differentiation between races found using the neutral SSR markers was similar to that found previously, although the average FST was lower than in previous studies (FST = 0.05; 95% CI 0.041-0.063). The SNP dataset (98 SNPs from 20 genes) from the same set of samples was provided by Dr. Saravanan Thavamanikumar of the University of Melbourne. Twenty-four SNPs were excluded because their minor alleles were too rare (total minor allele frequency < 10%), also eight SNPs that were not in Hardy-Weinberg Equilibrium (HWE) were eliminated. A further 36 were excluded because positive linkage disequilibrium (LD) between SNPs within genes was found in at least one of the races. While virtually no LD was found between SNPs in some genes, the LD level varied markedly between races and between genes. Of the 30 SNPs included in the analysis, the FST values of most were within the 1-99% inter-percentile range observed for the SSR data, and the average FST (0.09; 95% CI 0.058-0.133) was not significantly different. However, 6 SNPs had FST values that were higher than the upper 99% percentile of the FST distribution for SSRs. The SNPs exhibiting signals of selection occurred in two candidate genes in the lignin (4CL; LIM) and three in cellulose (KOR ‚Äö- 2 SNPs; SUSY3) biosynthetic pathways and in one in a Protein kinase-like gene (PKL1). This suggests that natural selection has promoted adaptive differentiation between races and is congruent with quantitative genetic analysis of wood chemicals (cellulose and lignin content and S:G ratio of lignin) which have also been found to vary more between races than expected by chance. Despite evidence for selection acting on several SNPs, similar groupings of individuals were obtained from STRUCTURE analysis with both data sets.

History

Publication status

  • Unpublished

Rights statement

Copyright 2012 the author

Repository Status

  • Open

Usage metrics

    Thesis collection

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC