The origins and evolution of Libertia (Iridaceae) : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Plant Biology at Massey University, Manawatū, New Zealand. EMBARGOED until 24 August 2028.

Abstract

Polyploidy is a mechanism that has generated plant species diversity. The condition arises from whole genome duplication resulting in organisms with multiple chromosome sets. Polyploidy has been proposed to increase the genetic potential of species diversity to evolve that, in the presence of other factors, has contributed to evolutionary innovation and lineage radiation throughout plant evolution, especially of the angiosperms. In this thesis, factors that may influence the contribution of polyploidy to plant diversity are explored through phylogenomic studies of 𝘓𝘪𝘣𝘦𝘳𝘵𝘪𝘢 Spreng. (Iridaceae), a non-model genus that comprises ca. 14 species ranging in ploidy level from diploid (2x) to dodecaploid (12x). To establish and contribute to diversity (i.e., be ‘successful’), polyploidy requires opportunities that arise from environmental change. In Chapter 2, the environmental context in which 𝘓𝘪𝘣𝘦𝘳𝘵𝘪𝘢 evolved was inferred through a biogeographic study. The genus occurs in New Zealand, Australia, New Guinea, and South America. Although this disjunct distribution is not unusual amongst plants, it is distinctive amongst the Iridaceae with only the sister genus of 𝘓𝘪𝘣𝘦𝘳𝘵𝘪𝘢, 𝘖𝘳𝘵𝘩𝘳𝘰𝘴𝘢𝘯𝘵𝘩𝘶𝘴 Sweet., similarly occupying both Australia and the Americas. Biogeographic analyses were conducted on a plastome sequence-based phylogeny of the Iridaceae. To address a challenge whereby widely used biogeographic models tend not to reconstruct areas that are not currently occupied, the connectivity parameter was adjusted amongst southern landmasses to assess whether Antarctica could be inferred. The model that was most stable and biologically reasonable suggested that Antarctica was likely involved in the early evolution of the 𝘓𝘪𝘣𝘦𝘳𝘵𝘪𝘢-𝘖𝘳𝘵𝘩𝘳𝘰𝘴𝘢𝘯𝘵𝘩𝘶𝘴 lineage. Ultimately, long-distance dispersal explained the disjunct distributions of 𝘓𝘪𝘣𝘦𝘳𝘵𝘪𝘢 and 𝘖𝘳𝘵𝘩𝘳𝘰𝘴𝘢𝘯𝘵𝘩𝘶𝘴. Additional dispersal and environmental changes within the southern landmasses likely underpinned in situ speciation of each genus. Taken together, the evolution of 𝘓𝘪𝘣𝘦𝘳𝘵𝘪𝘢 appears to have been shaped by environmental change that has likely provided opportunities for the success of polyploid lineages. In Chapter 3, species relationships were inferred to understand the sequence of, and contributing lineages involved in, polyploidy events that generated 𝘓𝘪𝘣𝘦𝘳𝘵𝘪𝘢 species diversity. Reconstructing relationships of non-model and high ploidy species has been challenging predominantly due to short-read sequences often not spanning polymorphisms that distinguish locus copies. Using a modified method, long-read sequence data were generated. A novel bioinformatic pipeline was applied firstly, to phase the long-read data into homeolog sequences for three polyploid species then, secondly, to use the polymorphism information of the phased long-reads to phase short-read data of other accessions and polyploid species. Phylogenetic tree and network analyses were employed which inferred four allopolyploidy events involving four putative progenitor lineages. All polyploid species appeared to share a subgenome that is closely related to a South American diploid species, explaining the close relationships and morphological affinities amongst the South American and polyploid Australasian 𝘓𝘪𝘣𝘦𝘳𝘵𝘪𝘢 species. Given the prevalence of polyploidy and uptake of target enrichment for phylogenetic studies, the methods developed in Chapter 3 may be of wide interest and application. In Chapter 4 the genetic diversity and structure of a hexaploid species, 𝘓𝘪𝘣𝘦𝘳𝘵𝘪𝘢 𝘱𝘦𝘳𝘦𝘨𝘳𝘪𝘯𝘢𝘯𝘴, was investigated to provide insights into polyploid evolution following establishment and to inform ideas concerning the longer-term consequence of polyploidy on plant diversity. 𝘓𝘪𝘣𝘦𝘳𝘵𝘪𝘢 𝘱𝘦𝘳𝘦𝘨𝘳𝘪𝘯𝘢𝘯𝘴 is endemic to New Zealand and displays all the reproductive assurance traits typically associated with polyploidy. Despite these, the species is endangered in nature and is being actively conserved. On the other hand, these traits make cultivation easy and the species a popular ornamental that is planted in urban settings throughout New Zealand. RAPDseq data were generated with a novel approach used to inform homeolog assembly, which involved the use of genetic diversity statistics of the phased homeolog sequences generated in Chapter 3. Population genetic analyses were conducted to investigate the level and distribution of genetic diversity as well as assess the identification of inland individuals as hybrids of the species with 𝘓. 𝘪𝘹𝘪𝘰𝘪𝘥𝘦𝘴 (12x). Results suggested that the genetic diversity of 𝘓. 𝘱𝘦𝘳𝘦𝘨𝘳𝘪𝘯𝘢𝘯𝘴 is low with only broad genetic structure of natural coastal populations supported across analyses. The identification of hybrid inland individuals was supported with no gene flow detected between natural coastal and inland individuals. Together, these results support the idea that reproductive traits that aid polyploid establishment may hinder further divergence in the longer-term. Results also suggested that current ex situ conservation and urban plantings are conserving genetic diversity of 𝘓. 𝘱𝘦𝘳𝘦𝘨𝘳𝘪𝘯𝘢𝘯𝘴 from the North Island, emphasizing the value of conservation efforts as well as the opportunities for conservation in urban settings. The three studies conducted in this thesis provide a multi-level account, in terms of time, space, and taxonomy, of the origins and evolution of 𝘓𝘪𝘣𝘦𝘳𝘵𝘪𝘢. The approaches used in Chapters 2-4 all have novel aspects that contribute to fields of biogeography, phylogenetics, and population genetics, respectively. They work towards enabling the reconstruction of the evolutionary histories of non-model polyploid taxa that have been considered challenging to study. Hence, 𝘓𝘪𝘣𝘦𝘳𝘵𝘪𝘢 is shown to be a deceptively modest system that is useful for solving analytical challenges associated with investigating the complex role of polyploidy in the evolution of plant diversity.