Animal colouration is associated with a multitude of ecologically adaptive traits known to drive biological diversification, from predator avoidance to physiological regulation. As such, it is an ideal system in which to study the evolutionary patterns and processes that generate and maintain biological diversity. Within the terrestrial vertebrates, amphibians display some of the greatest complexity and variation in terms of colour patterning, with the salamander genus Salamandra particularly renowned for its colour diversity. Typically, Salamandra species present bright, highly variable yellow-black patterns consisting of spots and/or stripes, which are thought to hold an aposematic (warning) function related to their toxic secretions. In addition to this, individual species and populations have evolved melanic, fully yellow and fully brown colourations, with gradations seen in-between. Importantly, there are also indications of parallel colour pattern evolution, making Salamandra an attractive system for studying the repeated evolution of adaptive phenotypes. However, the genus currently lacks phylogenetic resolution, and the molecular mechanisms underlying amphibian colouration are poorly understood. In this thesis, I aim to fill both of these knowledge gaps through the use of next-generation sequencing (NGS) techniques, which offer both unpredicted opportunities to resolve systematically challenging relationships and allow us to study the genetic basis of ecologically adaptive phenotypes in wild non-model organism. In Chapter 2 we reconstruct the controversial interspecies phylogeny of Salamandra using three largely independent phylogenomic data sets. First, using restriction site associated DNA sequencing (RAD-Seq), I genotyped representatives of all six currently recognised Salamandra species (and two outgroup species from its sister genus Lyciasalamandra). This was combined with nuclear protein-coding sequences derived from RNA-Seq and full mitochondrial genomes. Analyses of concatenated RNA-Seq and RAD-Seq data retrieved well supported, fully congruent topologies that placed: (1) S. infraimmaculata as sister to all other species in the genus; (2) S. algira sister to S. salamandra; (3) these two species sister to a clade containing S. atra, S. corsica and S. lanzai; and (4) the Alpine species S. atra and S. lanzai as sister taxa. The phylogeny inferred from mitochondrial genomes differed from this in its placement of S. corsica, as did species tree analyses of RNA-Seq and RAD-Seq data. However, the general congruence among topologies recovered from the RNA-Seq and RAD-Seq data sets gives us confidence in our methodologies and results. In Chapter 3, I perform more in-depth phylogenomic analyses, using RAD-Seq to genotype 231 salamanders from across the taxonomic and geographic breadth of Salamandra. Both Bayesian and maximum likelihood based analyses of concatenated RAD-loci (comprising 187,080–294,300 nt of sequence data) returned well-supported, largely congruent topologies that supported the monophyly of all six currently recognised species. However, the placement of S. corsica was again unclear, and data filtering parameters were found to have a great impact on downstream analyses. Further, I identified undescribed diversity within the North African species (S. algira) and find that 43% of S. salamandra subspecies do not meet a criterion of monophyly. Following this, I use the phylogenetic hypothesis generated to assess the parallel evolution of reproductive (parity) mode and two colour phenotypes (melanism and stripe formation) through ancestral state reconstruction analyses. I find that pueriparity (giving birth to fully metamorphosed juveniles) has independently arisen in at least four lineages, melanism in at least five, and a striped phenotype in least two, all from a common yellow-black spotted larviparous (larvae depositing) ancestor. Finally, in Chapter 4, I leverage and highly colour-variable lineage of the European fire salamander (S. salamandra bernardezi) to identify genetic associations with colour, test for selection on colouration, and test the relationship between colour phenotype and toxicity (the functional basis of aposematism). I show that, within a geographically restricted region, colour phenotypes form a gradient of variation, from fully yellow to fully brown, through a yellow-black striped pattern. Population genetic analyses suggest a sympatric evolutionary origin for this colour variation, and I found no association between a salamanders colour pattern and the metabolomic profile of its toxic secretions, which calls into question the adaptive significance attributed to these striking colourations. Following this, I identified significantly differentially expressed genes between skin colours using transcriptomic (RNA-Seq) analyses and genomic loci associated to representative colour phenotypes (yellow, brown and striped) using RAD-Seq approaches. I also found signals of selection on genomic loci between representative colour phenotypes, several of which overlap with genomic analyses. Overall, my results provide greater phylogenetic resolution for the genus Salamandra than ever before, revealing the need for taxonomic revisions and confirming the convergent (or parallel) evolution of both reproductive and colour phenotypes. My data also represents a significant contribution to our understanding of the genetic basis of amphibian colouration, providing a valuable resource for future comparative research on vertebrate colour evolution.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:744130 |
| Date | January 2018 |
| Creators | Burgon, James D. |
| Publisher | University of Glasgow |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://theses.gla.ac.uk/9029/ |
Page generated in 0.0112 seconds