Terpene synthase gene family organisation in <em>Melaleuca alternifolia</em> and its influence on chemotype

Julia Voelker

doi:10.25918/thesis.273

Tea tree is an ideal model to study terpenoid genetics within the family of Myrtaceae due to the deep understanding of terpenoid biochemistry, genomic tools and population resources accumulated over the more than 30 years it has been used as a plantation crop for essential oil production in Australia. Tea tree has six recognised chemotypes that are defined by the foliar concentrations of the three key monoterpenoids terpinen-4-ol, 1,8-cineole and terpinolene, but only the high terpinen-4-ol chemotype is grown commercially for its medicinal properties. Chemotypes in natural populations have a distinct geographic arrangement in upland and coastal regions. An adaptive role for chemotypes remains unclear, although recent studies showed a correlation of host preference by insects with terpenoid chemistry. The fundamental steps of terpenoid biosynthesis are well understood, and some TPS and other key enzymes have been identified and characterised in tea tree. Many plants encode a vast array of TPS, each catalysing the synthesis of one or more terpenoid constituents, with some species from the related genus Eucalyptus possessing the largest TPS gene families known in plants. Genome studies have shown that these TPS tend to be organised into gene clusters, due to an evolutionary process involving tandem duplication and subsequent functional specialisation. The resultant gene organisation makes TPS clusters prime candidates for supergene formation. Long-read genome sequencing was carried out to provide a foundation for reliable gene discovery, and study of genomic organisation in tea tree. A draft genome of 362 Mb size was assembled that exceeded the quality (N50 = 1.9 Mb) of any previously available sequence for the taxon. A total of 58 putatively functional full-length TPS genes, and 21 pseudogenes were identified and manually annotated. As in other Myrtaceae, TPS genes tended to occur in tandem arrays. The largest cluster of 13 TPS was located within 400 kb on a single scaffold of 731 kb length. This cluster included four TPS that encoded a sabinene hydrate synthase, producing the precursor of terpinen-4-ol, one terpinolene synthase, and two 1,8-cineole synthases. The observation of a tight clustering of four structural genes, previously shown to influence the levels of the key monoterpenoids defining chemotypes in tea tree, satisfied the first part of the supergene concept. A genome-wide association study (GWAS) was then used to associate marker genotypes with terpenoid chemotypes in tea tree. A set of 148 geographically diverse individuals spanning the upland and coastal ecotypes, and all of the six known chemotypes, were selected for genotyping, and approximately 22,000 single-nucleotide polymorphisms (SNPs) were identified. The GWAS found SNPs significantly associated with variations in 1,8-cineole, terpinen-4-ol, and terpinolene concentrations. Large proportions of their variation were explained by five SNPs near the previously identified TPS cluster, with some individual SNPs explaining 19-30% of phenotypic variance. Closer examination uncovered the presence of chemotype-dependent null alleles at a further three SNPs located within the TPS cluster, suggesting one or more structural variants in this region, which may contribute to reduced recombination, and thus provides a possible basis for the strong linkage disequilibrium observed on this scaffold. Confirmation of SNP allele associations with terpenoid phenotypes was found in a separate domesticated population subject to artificial selection for a single chemotype, where all SNP alleles surrounding the TPS cluster were fixed. These observations strongly suggest a supergene, where multiple physically proximal TPS genes behaving as a single linkage block account for the discrete chemotypes in tea tree. TPS supergenes are likely facilitated by tandem duplication and subsequent specialisation, leading to gene clusters with related functionality. Multiple adaptive alleles at neighbouring loci could then be captured as haplotypes, if structural variation arises that impedes recombination between TPS in this region. The shared biochemical origins of the terpenoid components and their interdependence are likely key to the concerted changes in discrete chemotypes and this mode of supergene evolution. Such supergenes, where relatively few alleles from several TPS loci are tied together in a limited number of haplotype blocks, may allow concerted, saltatory changes in complex terpenoid phenotypes in tea tree and other plants. This thesis argues that a genetic model with three haplotypes, encompassing the four characterised TPS genes, and invoking substrate competition for a common precursor, could account for all six chemotypes in M. alternifolia, and is consistent with the available bi-parental cross segregation data.

Terpene synthase gene family organisation in Melaleuca alternifolia and its influence on chemotype

Files and links (1)

Metrics

Abstract

Details

Terpene synthase gene family organisation in Melaleuca alternifolia and its influence on chemotype

Files and links (1)

Metrics

Abstract

Details

Southern Cross University Social media