A3: Genomics of local adaptation and hybridization in Daphnia in response to the trophic status of Lake Constance

Axel Meyer

State of the art

Drastic environmental changes such as the eutrophication of lakes can strongly affect the population genomics of some players in ecological networks. These genetic changes can be especially pronounced when native populations hybridize and back-cross with new species that invaded the lake e.g. during eutrophication phases. Introgression is likely to lead to genomic changes in the native species that will not be reversible even if the ecosystem restoration was successful (e.g. Chowdhury et al. 2014; Roy Chowdhury et al. 2015). The Daphnia egg bank offers a powerful tool to study past microevolutionary trends (Hairston et al. 1999; Brede et al. 2009; Frisch et al. 2014) including those due to hybridisation and introgression. Using the Daphnia egg bank microevolutionary responses and local adaptations (Savolainen et al. 2013; Tiffin and Ross-Ibarra 2014) to e.g., cyanobacterial toxins (Hairston et al. 1999), eutrophication (Frisch et al. 2014; Möst et a. 2015) and recovery from metal pollution (Turko et al. 2016) were examined.

In Lake Constance, native Daphnia longispina (former hyalina) were shown to hybridize and backcross with the invasive D. galeata (Jankowski and Straile 2004; Brede et al. 2009). Using the egg banks, we will compare current D. longispina and D. galeata populations in Upper Lake Constance with past D. longispina populations from before 1950s and D. galeata populations from the 1960s and 1970s. This will allow us to analyze how many and which genes in the current populations were introduced due to hybridization. The availability of the Daphnia genome (Colbourne et al. 2011) will greatly facilitate this task. Those results will be used to better understand the extent to which the hybridization and introgression events have affected the resilience and persistence of Daphnia populations in response to oligotrophication. This research will provide crucial data on fundamental evolutionary questions such as the possibility of “speciation-reversal” and “re-speciation” and the genetics of local adaptations and repeated evolution (reviewed in Elmer and Meyer 2011; Savolainen et al. 2013).

Utilizing next-generation DNA Sequencing (NGS) technologies, it is now feasible to conduct ‘population genomics’ studies to identify genomic loci that are targets of natural selection, and might be responsible for local adaptations (Luikart et al. 2003; Savolainen et al. 2013; Tiffin and Ross-Ibarra 2014). RAD-Seq techniques (Baird et al. 2008; Stinchcombe and Hoekstra 2008; Hohenlohe et al. 2010; Recknagel et al. 2013) and, if samples or clones of extinct populations are available, whole genome re-sequencing, allows the analysis of genome-wide patterns of polymorphisms and the analysis of genomic changes (such as structural variants, selection, selective sweeps etc.).

Preparatory work

Our lab has extensive experience in population genomics analyses. We applied and developed RAD-seq techniques and used whole genome re-sequencing (WGS) to identify genomic loci within a fish species complex that are presumably under divergent selection (e.g., Elmer et al. 2014; Kavembe et al. 2016).

Proposed project and role within the RTG

The doctoral researcher will compare the genomes of the recent species (D. longispina and D. galeata), characterized morphologically and by microsatellites, with two egg bank populations representing the pre-hybridization or recently invaded populations. The two egg bank populations will give us insight into the genomes of the “original” species. As D. galeata invaded the lake in the 1950s, we can use D. longispina eggs dated prior to the 1940s to sample a population not yet influenced by hybridization and introgression. D. galeata resting eggs will be sampled from the period of peak eutrophication, i.e. from the 1960s and 1970s. Previous studies have shown that during this period, only D. galeata eggs were deposited in the egg bank. However, we cannot exclude that already during this early part of the invasion some introgression occurred, possibly leading to an underestimation of the actual differences between the native D. longispina and the invasive D. galeata. These two “original” populations will be compared with the current populations of Daphnia sampled during three phases of seasonal succession (February, May, August). In February we will sample the overwintering populations, during May the populations that originated from the resting egg bank, and during August the populations that survived the summer period of food scarcity and increased predation. Sampled Daphnia will first be analyzed with microsatellites and individuals will be sorted into D. longispina, D. galeata and hybrids. Population genome scans will be performed for 50 individuals or resting eggs for each of the two “original” populations as well as for the current lake populations. Depending on the feasibility of using resting eggs for genomics (e.g. five populations of D. longispina 1900-1940 from which only resting eggs exist so far) or the possibility or necessity to produce clones we will adjust the exact method of data collection and sampling scheme accordingly. Ideally, we will aim for a higher temporal resolution (e.g. 1940s, 1950s, etc.) and spatial resolution (Untersee).

Using these methods (RAD-seq and/or WGS) we will identify “outlier” loci that are under strong natural selection. Through comparative genomic approaches we will identify candidate genes that might contribute to ecologically relevant phenotypic variation. The RAD-Seq approach provides a fast and cost-effective way of genomic characterization, as the reads could be mapped to the genome and the respective SNP's can be interpreted as either associated with either longispina or galeata. As NGS methods and NGS technology evolve very quickly we will adjust our protocols to include the most comprehensive and, at the same time cost-effective, methods for this project such as whole-genome sequencing. That will allow us to also quantify hybridization patterns on a genome scale and investigate structural variation in the “original” and hybridized genomes.