Ecological genomics of Australian glass shrimp-Paratya australiensis: population structure and local adaptation to altitude in south-east Queensland

Population genetics provides a framework for understanding genetic drift, mutation, migration and selection in natural populations. In the past, population geneticists concentrated on a handful of markers due to limitations of technology. However, with the advancement of Next Generation Sequencing Techniques, the field of population genomics has emerged, which facilitates understanding of evolutionary processes at a genomic level. Any population genetic measure which used to be just a point estimate can now be estimated across the genome. In addition to calculating genome wide averages for genetic differentiation among populations, we can also identify specific regions of the genome under natural selection. It has been of great interest to study natural selection in populations adapted to different environmental conditions and with the development of the technology this has become more feasible. Due to global climate change, species need to adapt to the changing environment and this is only possible if there is sufficient adaptive genetic variation at the molecular level. Adaptive divergence among populations has been studied across different altitudes, in different host forms, along latitudinal clines and between different coastal thermal regimes. This thesis is focused on the species Paratya australiensis which consists of 9 highly divergent mtDNA lineages. Lineage 4 and Lineage 6 of this species have been observed to have preference for upstream and downstream respectively and have been suggested to be adapted to different temperature regimes. In order to evaluate this, I have used a genomic approach in this thesis to investigate population structuring and evidence of adaptation between upstream and downstream populations of P. australiensis in the Mary River catchment in south east Queensland, Australia. My thesis has focused on a single lineage, lineage 4. Three streams (Broken Bridge, Booloumba and Obi Obi Creek) within the Mary River catchment were selected for the study, each consisting of upstream and downstream sites (6 populations). There were three main parts to this research: (1) testing if the population structure followed the Stream Hierarchy Model (SHM) and to compare to past conclusions based on conventional genetic markers (2) to identify putative genomic regions under selection between up and downstream populations and (3) to examine the distribution of lineage 4 and lineage 6 and to determine if they co-occur and, if so, whether they interbreed. In the past, it was observed that there was restricted gene flow between upstream and confluence populations of P. australiensis. Also, it was observed that P. australiensis in the Conondale Range followed the Stream Hierarchy Model (SHM), particularly in the Brisbane River catchment. However, those studies employed allozyme and mtDNA markers. In Chapter 3, I used Single Nucleotide Polymorphism (SNP) markers produced by the technique of Restriction Site Associated DNA Sequencing (RAD-Seq). I found a high level of genetic divergence between up and downstream populations of Booloumba Creek, which is similar to the past findings and indicates that this high level of divergence has been maintained between populations of this stream (Booloumba) over the years. Populations in the other two streams (Broken Bridge and Obi Obi Creek) did not show such high levels of divergence. Also, population structure did not fit with the Stream Hierarchy Model (SHM) as there was more genetic divergence between sites within streams than between streams in the catchment. Such results were possibly due to large divergence between Booloumba Creek populations. So, this study was more robust utilizing a large genotype data set compared to previously used markers. Furthermore, the new set of markers was developed for the first time and was successfully utilized to evaluate the present condition of population structure and genetic diversity of P. australiensis in the Conondale Range. The high divergence in Booloumba Creek in my study suggests that long term isolation and restricted geneflow in Booloumba could lead to local adaptation. Based on previous findings on the preference for up and downstream locations in lineage 4 and 6 respectively, I tested for evidence of local adaptation in these streams and looked for parallel patterns of adaptation across all three streams in Chapter 4. There were 44 outliers identified across the 6 populations, indicating putative genomic regions under selection. Furthermore, there was fixation of alternate alleles for these outliers between up and downstream populations in one stream (Booloumba Creek). Some outlier loci were common across two streams indicating parallel pattern of adaptation to some degree. In addition to the neutral divergence, adaptive divergence was also observed among the populations based on the outliers. As the studied populations were selected based on altitude, it is suggested that the putative evidence of local adaptation could be due to environmental differences between altitudes, most probably temperature differences. So, there are putative genomic regions under selection in P. australiensis, which suggests that with climate change and changing environmental conditions this species may have enough adaptive genetic variation to enable some level of adaptation to future climate change conditions. Due to a past translocation event in the Brisbane River catchment, artificial sympatry was observed between lineage 4 and 6 of P. australiensis. Also, asymmetrical hybridization between the two lineages and introgression of lineage 4 genes into lineage 6 was observed in the translocation area. However, natural sympatry between these 2 lineages had not been reported, despite extensive sampling. It would be interesting to observe if the 2 lineages show hybridization or not if they co-occur naturally. In this thesis, I have identified a case of sympatry between L4 and L6 in the Mary River catchment which is described in Chapter 5. It was thought that lineage 6 was absent from the Mary River catchment. However, I found evidence of lineage 6 in one of the sites together with lineage 4. Based on phylogenetic analysis using COI gene (mtDNA), there was evidence of sympatric lineages in only one site (Broken Bridge High site) in the Mary River catchment. Haplotypes identified in the Mary River catchment were different from those in the Brisbane River catchment. Analysis of nuclear genes in the two lineages (L4 and L6) showed that there were no lineage 6 alleles based on the allozyme data, but microsatellite loci were all under Hardy Weinberg Equilibrium (HWE) expectations, indicating random mating between the lineages. Here I observed that the 2 lineages are interbreeding but we need further investigation to understand the nature of hybridization and if there is introgression or not. Overall the thesis has revealed that P. australienis still shows restricted gene flow and population divergence in some parts of the Conondale Range. Due to long term divergence, some populations have become locally adapted to some extent. Sympatric lineages of P. australiensis have been identified in the past, although sympatry between lineage 4 and 6 was not identified previously. So, in this thesis, I present evidence of co-occurrence of lineage 4 and 6 for the first time in the Conondale Region. Also, the thesis provides information regarding interbreeding status of these 2 lineages. The thesis presents development and utilization of genomic data for the very first time in this species and can benefit other studies dealing with genomics of crustaceans. In addition, this thesis provides an in depth knowledge on the technology and analytical techniques for genomic data.