Population and landscape genetics in the water flea Daphnia

Summary The population genetic structure of natural populations in landscapes is the result of a complex interaction between ecological and evolutionary processes such as gene flow, genetic drift and selection, which in turn are shaped by landscape characteristics (environmental gradients and geogra...

Full description

Bibliographic Details
Main Author: Haileselasie, Tsegazeabe Hadush
Other Authors: De Meester, Luc; U0008482;, Mergeay, Joachim; U0037361;
Format: Doctoral or Postdoctoral Thesis
Language:English
Published: 2016
Subjects:
Online Access:https://lirias.kuleuven.be/handle/123456789/556843
https://lirias.kuleuven.be/bitstream/123456789/556843/1//Thesis-Print-Final-THH-JM-LDM-02.pdf
Description
Summary:Summary The population genetic structure of natural populations in landscapes is the result of a complex interaction between ecological and evolutionary processes such as gene flow, genetic drift and selection, which in turn are shaped by landscape characteristics (environmental gradients and geography), historical processes, or their combined effects. Understanding the relative role of these processes is fundamental to improve our ecological insight and to design strategies for biodiversity conservation. Landscape genetics is a young discipline that enables us to understand the processes behind patterns of genetic structure as observed in nature by relating spatial genetic variation to local environmental conditions and geographic location. The main goal of this dissertation was to analyse the relative role of environmental variables and spatial variables in structuring genetic variation in Daphnia populations. I used multivariate approaches that are commonly applied in community ecology to quantify genetic diversity within and genetic differentiation among populations and disentangle the relative contribution of local environmental conditions and spatial drivers to explain variation in genetic composition in natural landscapes. I applied this approach to four different settings: clonal composition in obligately asexual species of the Daphnia pulex species complex in Greenland, coexistence and genetic variation in an obligately asexual and a cyclically parthenogenetic D. pulex species in northern Finland, genetic variation in the cyclic parthenogen D. magna in coastal rock pools along the Baltic, and genetic variation in the cyclic parthenogen D. sinensis in young artificial reservoirs in Ethiopia. In all these settings, ecological and geographic distances among populations were related to genetic distances in multivariate redundancy analyses (RDA) to explore whether genetic variation in the landscape was more structured along environmental gradients or along spatial gradients. In Chapter I, I show that the newly created habitats created by recent deglaciation in Greenland get colonized by multiple genotypes within decades of their formation. Furthermore, an increase in clonal richness with age of habitat resulted in a clear gradient of clonal richness in the direction of retreat of the Greenland ice sheet. In Chapter II, I build further on these results and disentangle the processes that structured the asexual clonal assemblages in two age clusters using multivariate partitioning analysis techniques. The results show that a significant proportion of variation in clonal composition was attributed to environmental variation among habitats, whereas pure spatial models did not account for a significant portion of the variation in clonal composition among habitats. From this, I concluded that the cause of the increase in clonal richness with age as observed in Chapter I is due to an increase in habitat environmental heterogeneity with age, leading to an increase in niche diversity in older habitats. The results of my analyses show that local environmental variables act as an important filter determining colonization of newly created habitats and reveal that dispersal is high enough to sustain a pattern of clonal sorting so that clonal composition reflects local environmental conditions. In Chapter III, I have estimated the number of founders of Daphnia sinensis populations inhabiting recently created reservoirs in Ethiopia based on nuclear and mitochondrial markers. The results showed that the genetic structure of the recently established D. sinesis populations are strongly influenced by founder effects and reflect a small number of founders (typically lower than five). I did not detect any evidence for the importance of environmental and spatial variables in determining landscape genetic structure. In Chapter IV, I quantified the degree to which two regionally coexisting and ecologically very similar species coexist in single pond or pond clusters in the tundra of Northern Finland. I quantified the importance of environmental variation and space in determining landscape genetic structure of two species: the cyclically parthenogenetic species Daphnia pulicaria and the obligately asexual species I refer to as panarctic Daphnia pulex. I relate both community structure and the population genetic structure of the two species (clonal composition in the asexual species and allele frequencies in the cyclically parthenogenetic species) to environmental and spatial drivers. I observe a significant effect of environmental variation but no effect of space in determining both species composition as well as clonal composition in the asexual taxon, whereas there is no environmental nor a spatial signal in the landscape genetic structure of the cyclically parthenogenetic species. The niche overlap between the cyclically and obligately parthenogenetic species is high. My results indicate that there is no dispersal limitation in this system, and that the lack of coexistence of the two species at the level of single ponds or pond clusters likely reflects strong priority effects. In Chapter V, I have report strong genetic differentiation among local populations (FST = 0.382) and low genetic diversity within populations in Daphnia magna inhabiting rock pools along the Swedish Baltic coast. There was a significant correlation between genetic and geographic distances at regional level, which largely reflects differences among populations located on different isolated islands, but no correlation between salinity levels and genetic differentiation. Overall the different analyses done on different species and settings amongst others revealed that 1) there is a significant correlation between genetic distance and environmental distance but not geographic distance in the obligate parthenogens from Greenland and Finland, 2) a significant correlation between genetic distance and geographic distance but not with ecological distance in the populations of the cyclically parthenogenetic species D. magna populations from Sweden, and 3) no correlation between genetic distance and both environmental and geographic distance in the cyclically parthenogenetic species D. sinensis population from Ethiopia. My results strongly suggest that isolation-by-environment is a key process structuring genetic variation at the landscape level in obligately parthenogens, whereas the sexual populations in the Baltic coast rock pools show a weak sign of isolation-by-dispersal limitation. I also obtained strong indications for founder effects, suggesting that the landscape genetic structure of Daphnia populations is also influenced by patterns of colonization. The results I obtained for the Daphnia populations in the Ethiopian reservoirs suggest that the regional population structure is strongly impacted by founder effects that are the result of a low number of colonists and the absence of strong subsequent gene flow. More in general, the results of the landscape genetic analyses in this thesis reveal highly repeatable patterns of genetic structure of Daphnia populations even though our analyses involved strongly diverging habitat types, landscapes and geographical settings. One recurrent pattern was that there is little evidence for strong dispersal limitation but also little evidence for strong gene flow. The observed patterns of landscape genetic structure were strongly influenced by the reproductive mode of the water flea species studied. status: published