A molecular and population genetics study of the Bluff Oyster (Ostrea chilensis)

Abstract

<p>New Zealand has a long history of isolation and has evolved a unique biota. Spanning from a sub-tropical climate in the North to a sub-Antarctic climate in the far south New Zealand provides an interesting opportunity to study the processes that lead to evolution. This thesis attempts to study the evolution of Ostrea chilensis at a population genetics and molecular level. Chapter Two: Microsatellite DNA loci represent an ideal marker for population genetics studies due to high levels of length polymorphism between individuals. Genomic sequencing technologies offer the potential to quickly identify thousands of loci, from which PCR primers can be developed and screened for polymorphisms. I aimed to develop PCR primers to amplify length polymorphic microsatellite loci and to use the genomic data set to elucidate patterns and processes of microsatellite evolution. DNA was extracted from a single Ostrea chilensis individual and used for a 1/8 plate sequencing run on a Roche 454. The subsequent quality checked DNA database was annotated for microsatellite loci. 6,208 dinucleotide, 7,326 trinucleotide repeats, 2,414 tetranucleotide, 33 hexanucleotide and 356 pentanucleotides were annotated on the partial genome. Four microsatellite loci were successfully amplified and genotyped. The loci have a low number of alleles compared to other bivalve studies and two have significant departures from HWE (Fst = 0.126 and -0.348). There were a number of highly significant BLAST hits (< 1xE⁻²⁰) with repetitive Oyster DNA sequences obtained from GenBank. Due to difficulties the microsatellite loci were abandoned as markers for later population genetic analysis. This work, however, provides the ground work for further developments of PCR primers for polymorphic microsatellite DNA and provides some observations of molecular evolution of repetitive DNA, which will lead to a greater understanding of these sequences. Chapter Three: This chapter forms the first population genetic study of Ostrea chilensis using New Zealand and Chilean populations. The life history traits of O. chilensis are thought to reduce the dispersal of the species. Using randomly amplified polymorphic DNA (RAPD) I aimed to test the population genetic structure with the null hypothesis that there is panmixia (i.e. high levels of gene flow and no barriers between populations). I then aimed to see if there was an isolation-by-distance (IBD) profile. Barriers to gene flow at around 420S have been identified in a number of studies around New Zealand. I aimed to see if those barriers were present in this study. Significant spatial genetic differentiation was found among populations (Fst= 0.194, p<0.00001). Over all spatial scales a significant IBD was not found, until ‘outlier’ (those with two standard deviations from the mean Fst) were removed, then a slight IBE was found (rxy=0.324, p=0.030, r2 =0.1052). In an AWclust analysis two main clusters were revealed, but they did not correspond to above and below the 42⁰S. It is possible that the brooding life style of Ostrea chilensis has resulted in reduced gene flow between populations. Much of the genetic structure was not congruent with geographic location; this apparent chaotic patchiness could be influenced by human mediated movements and/or environmental variables. Chapter Four: The genetic structure found in the previous chapter was analysed in the context of near shore environmental and geo-spatial variables. My aim was to elucidate the environmental variables that best explain the apparent genetic structure of O. chilensis. Using genetic data from the previous chapter a Generalised Linear Model (GLM) was used to test the effect of ten environmental variables and two geospatial variables on average genetic distance (Fst and PHIst). A complimentary BEST analysis was used to test the effect of the same variables on the individual alleles for each population. Using the BEST analysis the model that best explained the apparent genetic structure (Rs = 0.263) included the environmental variables Sediment (SED) and mean sea surface temperature (SSTGrad). The GLM showed a more complicated model including most of the variables tested. Previous studies have shown that bivalve dispersal patterns are associated with sediment and spawning times can be influenced by temperature. This line of enquiry is important as it could lead to the identification of candidate genes for selection. Chapter Five: Genomic datasets contain a wealth of data that can be used to further understand genome structure and arrangements. My aim was to discover mitochondrial gene fragments in the genomic data set, use them to assemble the Ostrea chilensis mitogenome, and then analyse the genome in a phylogenetic framework with other oyster mitogenomes. A custom BLAST database was created using the 454 genomic DNA data set; this was then BLASTED against all available oyster mitogenomes on GenBank. The DNA sequences with good statistical support (<1 x E⁻²⁰) where mapped against the mitogenome of Ostrea edulis. The O. edulis mitogenome was then used to annotate the resulting mitogenome. I was able to recover 10, 086 bp of mitochondrial DNA – this represents around 65% of the full genome. The subsequent topology of the Bayesian phylogenetic tree was similar to that found in previous studies. PCR primers have been designed to sequence the gaps in the mitogenome. This will allow full annotation. Full genome annotation will aid further research into genome evolution. Chapter Six: In the mitogenome, evolution is thought to mainly result in synonymous substitutions, due to functional constraints. I aimed to describe the selection pressures acting on each protein coding gene of the oyster mitogenome by comparing rates of non-synonymous (dN) and synonymous (dS) substitutions. I then aimed to discover if there was a codon bias in the oyster mitogenomes. Genetic distance was assessed using p-distance. The ratio of dN/dS was calculated using the method of Nei and Gojobori (1986). The overall ratio of dN/dS was <1 for all protein coding genes, this would suggest that the genes are under purifying selection (non-synonymous mutations are selected against). A consistent codon bias was found across all protein coding genes, this could indicate translational selection. This multidisciplinary approach aimed to explain the patterns and processes of evolution in O. chilensis. This thesis research developed molecular tools, and provided information that will aid fisheries and aquaculture management.</p>