Detecting adaptive divergence in a widespread aquatic plant using genetic markers under selection and genome scans
Final Report Abstract
Seagrasses are ecosystem engineering species of outstanding importance for soft sediment coastal habitats. An ongoing decline of seagrass meadows in the last decades has caused rising interest in the ecology, genetics and evolution of these species. Eelgrass (Zostera marina L.) is distributed in the northern hemisphere in widely diverging habitats which makes it suitable for the investigation of habitat dependent selection and adaptation. We conducted multilocus genome scans, a bottom-up approach to detect molecular markers under selection. In principle, genome scans apply statistical tests to detect outlier loci that fall outside a neutral distribution and are thus likely to be influenced by selection. 29 gene-linked microsatellite markers were successfully developed from an expressed sequence tags (EST) database and fast and cost-effective genotyping procedures were established. Initial screening revealed that EST microsatellites proved to be almost as variable as anonymous markers. Information about the genes these markers are linked to make them a valuable tool for genome scanning as the putative functions of genes located close to outlier loci are known. In a first genome scan, we tested 25 microsatellites, 14 of which were EST-derived, for habitat dependent selection. We made use of a three times replicated habitat contrast that served to affirm the outliers detected by two different neutrality tests. As there is little congruence among the detected outliers when different tests are compared, I used a combination of several tests, a suitable approach for an explorative genome scan. Results showed that EST microsatellites were not more often detected as outliers than anonymous markers in our study. We found signs of divergent selection in three loci that were repeatedly detected in all three habitat contrasts. The most interesting candidate is linked to a putative nodulin gene building water channels in cellular membranes suggesting a functional link with habitat differences. In order to assess the validity of genome scans, we re-tested our three population pairs by adding single nucleotide polymorphisms (SNP), a novel marker type for Z marina, adding a third neutrality test and by nearly doubling the number of markers assessed. Results of the second scan were reassuringly consistent with the first scan as the three markers detected earlier were confirmed by the extended approach. Nevertheless, neutrality tests showed strikingly different results and some signals found in the earlier scan became weaker in the re-test, which warns us of taking genome scan results as granted without careful evaluation of the methods used. In the re-test, three other loci showed consistent signs of selection across habitat contrasts. Two newly detected loci are linked to an acid phosphatase gene (related to hyper-osmotic stress) and a seed maturation protein, respectively, both suggesting effects of habitat dependent selection in the Wadden Sea environment. The candidates detected by the neutrality tests could serve as starting points As an additional goal, we aimed to investigate how fitness and multilocus heterozygosity (MLH) are related in Z. marina clones. Using 37 microsatellites, we assessed the correlation of fitness measured as clone size and MLH in two Z marina populations from the Baltic Sea. We showed for the first time that the intermediate heterozygosity principle, stating that offspring from parents with an intermediate level of relatedness will be have the highest fitness, applies in a clonal plant. In ecosystems that are structured and maintained by only one species, like seagrass beds, heterozygosity represents an important component of genetic variation that has rarely been assessed before. Overall, the results of this work suggest that multilocus genome scans can be a useful tool to detect habitat dependent selection in exploratory studies. It has to be kept in mind, though, that inherent caveats of genome scans must be taken seriously and repeated population pairs represent the only way to handle this problem. Multilocus heterozygosity has rendered valuable insights into possible effects of inbreeding and outbreeding in natural populations, which is even more meaningful in ecosystems built by a single clonal species.
Publications
- (2007) Genome scans detect consistent divergent selection among subtidal vs. intertidal populations of the marine angiosperm Zostera marina. Molecular Ecology 16:5156-5157
Oetjen K, Reusch TBH
- (2007) Identification and characterization of 14 polymorphic EST-derived microsatellites in eelgrass (Zostera marina). Molecular Ecology Resources 7: 777-780
Oetjen K, Reusch TBH
- (2008) Characterization of single nucleotide polymorphism markers for eelgrass (Zostera marina). Molecular Ecology Resources 8:1429- 1436
Ferber S, Reusch TBH, Stam WT, Olsen JL
- (2010) New evidence for habitat-specific selection in Wadden Sea Zostera marina populations revealed by genome scanning using SNP and microsatellite markers. Marine Biology 157:81-89
Oetjen K, Ferber S, Dankert I, Reusch TBH
