Sequences and signals : evolutionary histories of New Zealand skinks : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Genetics at Massey University, New Zealand
The application of DNA sequencing to studies of the New Zealand biota is illustrated by investigations into the evolutionary relationships of skinks in the genus Leiolopisma. DNA sequences from a region of the mitochondrial 12S rRNA gene were determined for 20 taxa by use of the polymerase chain reaction. A vertebrate secondary structure model for this part of the gene was developed using comparative sequence analysis and calculations of RNA folding energies. Approximately one third of the molecule does not vary between the vertebrates examined, and there are similar patterns of sequence variability among the vertebrates. The secondary structure model was subsequently used to assist phylogenetic analyses of the skink sequence data set. Analyses of the mitochondrial DNA sequence information, using newly developed and more sophisticated algorithms, did not produce a fully resolved phylogenetic tree for all the skinks, though relationships between some taxa are less ambiguous. The lack of resolution does not appear to be due to limitations in the analytical methods, nor to the patterns of nucleotide substitutions in the skink 12S rRNA sequences. The skink sequence data set is unusual in that most of the taxa have similar numbers of nucleotide substitutions when compared to each other. These results are interpreted as reflecting a rapid divergence of the Leiolopisma group of skinks. Simulation studies support this interpretation. Three hypotheses are presented to account for the patterns of sequence differences between the skinks. One proposes a Gondwanan divergence of the Leiolopisma group, about 80 million years ago. Under this hypothesis the distribution of the skinks on islands in the Pacific and Indian oceans can be explained, in part, by continental drift. A second hypothesis suggests that New Zealand Leiolopisma are derived from a Miocene (15-25 million years ago) evolutionary radiation in New Zealand. This hypothesis however is inconsistent with observations of the sequence similarities between Mauritian, Australian, and New Zealand skinks. A third hypothesis, proposing several independent colonizations of New Zealand by Leiolopisma, is also not as well supported by the available sequence data. However, a close relationship between the 12S rRNA sequences of one New Zealand species L. infrapunctatum and the Australian Lampropholis guichenoti suggests that at least two skink immigrations to, or emigrations from, New Zealand may have occurred. Predictions of the three hypotheses and strategies to test them are discussed. Some of the conclusions derived from analyses of the mitochondrial DNA sequences conflict with those obtained from allozyme information, though there are points of agreement. Comparison of the allozymc and sequence data do also revealed a case of hybridization between two sympatric species, L. n. polychroma and L. maccanni, at a site in Southland. Analyses of both data sets indicate that the morphological similarity of Leiolopisma species obscures a large amount of genetic diversity, and the evolutionary histories of New Zealand Leiolopisma are older and more complex than previously considered. Further genetical and ecological studies of Leiolopisma are required, but this thesis emphasizes both the suitability and necessity of molecular genetic approaches for evolutionary investigations in New Zealand.