Genetic and phenotypic variation of Onithochiton neglectus across a heterogeneous marine environment

Abstract

It is important to understand how the interaction between species and their environment shape the spatial distribution of biological variation, and to identify the underlying evolutionary forces responsible for maintaining diversity in varied biological systems. The marine environment offers many possibilities for developing such knowledge, due to the dynamic interplay of complex abiotic features and high biodiversity. In New Zealand, the marine environment is influenced by a range of circulation patterns and environmental gradients. This environment can impose both intraspecific barriers to connectivity and strong selective pressures, leading to different modes of population genetic structure through genetic drift and local adaptation. Onithochiton neglectus is a chiton endemic to New Zealand, with highly variable morphology, and populations broadly distributed over this latitudinal environmental gradient. Given its parental brooding of larvae, and limited dispersal, low gene flow is expected among populations, unless facilitated by rafting on the holdfast of the bull kelp Durvillaea sp. Due to such characteristics, assessments of the biological variation among O. neglectus populations should prove particularly helpful to our understanding of the role of environmental interactions on the spatial distribution of variation. Using O. neglectus as a model, I described patterns of genetic and phenotypic variation across the New Zealand marine environment, and attempted to identify processes underlying such patterns. To describe the levels of differentiation among populations and the influence of kelp rafting on population connectivity, I undertook a combined analysis of morphological and genetic variation across the range of the species. To do so, I used geometric morphometric analyses to assess shell shape, including measurements such as perimeter, circularity, height and width of shell valves. To assess the genetic patterns, I used single-locus nuclear (ITS) and mitochondrial (COI and 16s) sequences, in addition to a large number of Single Nucleotide Polymorphisms (SNPs). The SNPs were generated through genotyping-by-sequencing, and putatively neutral and adaptive SNPs (outliers) were discriminated using genome scans for selection. The patterns of population differentiation recovered by shell shape, singe-locus markers, neutral SNPs and putatively adaptive SNPs were all compared, and differences and similarities among them were highlighted. Overall, my main findings suggest that O. neglectus has three highly differentiated genetic clades across New Zealand, corresponding to a North clade (populations from northern North Island), a Central clade (populations from central New Zealand) and a Southern clade (populations from the South Island, the Sub-Antarctic Islands, and the Chatham Islands). The Southern clade is consistently recovered by single-locus molecular markers, neutral SNPs, outlier SNPs, and by shell shape data. In contrast, the North and Central clades were recovered only by single-locus molecular markers and SNPs. These two northern clades have very similar morphology, even though there is strong genetic differentiation between them. Population structure is also much stronger within these clades than within the southern clade. These results support previous evidence that connectivity among O. neglectus southern populations is effectively enabled by kelp rafting, and similarity among distant southern populations is greater than expected for a brooding species. Given the high genetic differentiation among clades, and the distribution of the species across a heterogeneous environment, populations are likely to possess a degree of local adaptation. Comparisons of phenotypic differentiation in shell shape traits (measured using PST) with neutral genetic differentiation in SNPs (FST), revealed that PST was greater than expected in many of the PST-FST comparisons. These indicated a potential adaptive role for shell shape, which could be correlated with O. neglectus association with Durvillaea sp. Using genome scans for selection and genotype-by-environment association analyses, those loci that diverged from the background neutral genetic pattern (outliers) were identified, as were environmental factors associated with them. Due to the strong background hierarchical population structure among the three genetic clades, it was ultimately not possible at the New-Zealand-wide scale to confidently identify adaptive variation, or associated environmental factors. However, population differentiation in 230 outliers identified within the Southern clade are more likely to be reflecting responses to non-neutral process. Some of these outliers were associated with latitude, longitude, precipitation rate, surface pressure, and sea surface height relative to the geoid, suggesting that environmental heterogeneity might have a selective role on these populations In summary, I generated large genetic and morphological datasets for O. neglectus populations, and described geographic patterns of genetic and phenotypic variation in this system. They revealed unexpected relationships within a species that may be in the process of further speciation. The results show different patterns of population structure at varying scales, and a morphological subdivision, both likely due to the effects of kelp rafting. They also reveal discordance between neutral and non-neutral markers, which is likely due to natural selection, at least among southern populations. The findings in this thesis will help integrate our understanding of complex patterns of variation across marine species. Additional investigations of this type will ultimately lead to a better understanding of the evolutionary responses of marine species to environmental variability.