Genome-wide SNP analysis reveals patterns of population differentiation and adaptation of red rock lobster Jasus edwardsii

Declines in global marine finfish catches, which accounts for ~15% of the animal protein consumed by humans, has caused a 6-fold increase in total reported catch of invertebrates since 1950. This has led to the over-exploitation and decline of many marine invertebrate fisheries. The red rock lobster (Jasus edwardsii) fishery is New Zealand’s most economically valuable inshore fishery. The current management strategy relies on the assumption that the stock is comprised of a single panmictic population. However, more recent studies have challenged the genetic homogeneity of Jasus edwardsii across the Tasman sea and described high levels of self-recruitment in a Stewart Island subpopulation. A disregard for the underlying genetic structure in the management of a fishery can lead to excessive removal of individuals from populations contributing to the overall genetic diversity of the stock and thus reduce the species adaptability. The ability to adapt to new environments is particularly important in the context of global climate change and can significantly affect the long-term sustainability of the stock. Thus, the goal of this study was to identify specific patterns of genetic diversity of Jasus edwardsii population and provide an interpretation and assessment of the impact on the NZ fishery. The first objective was to optimize and validate molecular and bioinformatic protocols of Single Nucleotide Polymorphism (SNP) discovery for the red rock lobster Jasus edwardsii. The double digest restriction-site associated DNA (ddRADseq) protocol was optimized for the relatively large red rock lobster genome, which also has a high paralog content. The impact of bioinformatic processing on the population genetic inferences was then assessed by testing three different SNP discovery pipelines with the Rad-loci pipeline producing the most optimal marker discovery rate with a low level of missing data and a low SNP error rate. An analysis of technical replicates confirmed the reproducibility of both the molecular and bioinformatic protocols and also the validated the data generation process suitable for population genetic analyses. The second objective of my thesis was to investigate the genetic structure and population connectivity of adult red rock lobsters. The SNPs discovered were characterised as selectively neutral or under divergent selection (outlier) and both types of markers were analysed using Bayesian model-based clustering (STRUCTURE), non model-based multivariate analysis (Discriminant Analysis of Principal Components (DAPC)) and F-statistics. A lack of population differentiation using neutral genetic markers indicated a high level of gene flow and connectivity between populations. In contrast, there was evidence for selective pressure as a result of the analysis of outlier markers. Three main regions were identified: North-East NZ, North-West NZ and South NZ sub-populations, as part of a larger NZ metapopulation (FST ranged from 0.025 to 0.049, P < 0.001). The results of this study suggested that high levels of gene flow and connectivity are counteracted to some extent by the local selection that promotes the survival and reproduction of locally adapted genotypes. However, the strength of this selective pressure still permits low levels of survival and reproduction of non-optimal genotypes causing allele frequency homogenisation of the new generation of lobsters. The third objective was to investigate the levels of connectivity and adaptive divergence of the red rock lobster pueruli/juvenile lobsters for comparison with pattern of divergence of adult lobster in order to investigate the mechanisms of population structure formation. A suite of Bayesian clustering, non-model multivariate analysis and F-statistics were employed in the assessment of neutral and outlier markers developed for pueruli/juveniles. Similar to adult lobsters, pueruli/juveniles were characterised by a low level of divergence of the neutral markers indicating effective larvae dispersal. Outlier markers detected population differentiation patterns likely to originate from a phenotype – environment mismatch resulting in post-settlement mortality of non-adapted genotypes. The similarity between patterns of genetic divergence of adult lobsters and late juvenile/early juveniles indicates that post-settlement mortality, driven by local environmental conditions, has most likely occurred on earlier developmental stages of Jasus edwardsii, which were not possible to sample in my study. The final objective was to explore environment–genotype associations of Jasus edwardsii. Biological Environment Stepwise (BEST) analyses, redundancy analyses (RDA) and generalized linear modelling (GLM) consistently indicated a correlation between the annual amplitude of sea surface temperature (SST) and adaptive population divergence. In addition, an influence of spatial distribution on the patterns of adaptive population differentiation was also detected via RDA. From these results I propose a mechanism underlying the patterns of population differentiation discovered in Chapters 3 and 4: a latitudinal gradient of SST appears to be the selective force promoting the adaptive divergence of the lobster populations with local patterns of connectivity distorting the gradient and thus forming three distinct temperature adapted genotypes (North-West, North-East, and South). An environmental association analysis offered 43 candidate loci, which after alignment of transcriptome-mapped reference catalog sequences to annotated protein databases identified a candidate gene for thermal adaptation - UDP-glycosyltransferase (UGT). UGT is a detoxification enzyme involved in the metabolization of a variety of endogenous and environmental compounds and its activity and gene expression patterns have been linked to temperature. This study provides evidence for the local adaptations of the NZ population of Jasus edwardsii to SST, which together with the efficient mechanism of larval dispersal creates a system likely resilient to changes in temperature. This feature is important in the light of climate change-induced range shifts and supports the long-term sustainability of the red rock lobster fishery. The three genetically distinct regions identified coincide with existing boundaries of the management units and therefore do not require an adjustment of the current management regime