Evolution of the wild tomato species Solanum chilense

Demography and adaptation are important factors determining the
evolution of plant species. Many plant species are substructured
into populations or demes connected by migration (metapopulations).
The spatial distribution of populations and migration patterns
depend on the means of dispersal. Since plants are sessile
organisms, they also have to cope with both biotic and abiotic
stresses. Therefore adaptations to local environmental conditions
are essential to ensure survival and duration of the species. Wild
tomato species (Solanum section Lycopersicon) are native to western
South America. They occur in diverse and often extreme habitats
including rain forests, coastal regions, high altitude habitats in
the Andean Mountains and also hyperarid deserts in the Atacama
Desert. Therefore, wild tomatoes are a good model system to study
plant evolution and genomic bases for plant adaptation. This study
focuses on the wild tomato species Solanum chilense, which exhibits
a metapopulation structure with populations distributed from
southern Peru to northern Chile. In its native range, S. chilense
is confronted with different abiotic stresses including drought,
cold and salinity. I sequenced 30 unlinked nuclear genes from 23
populations using next generation sequencing. 16 genes are involved
in the abiotic stress response and serve as candidates for
selection and adaptation. The remaining 14 genes are used as
references to study the genomic average and species past
demography. In the first part of this study, I investigated the
demographic history of the wild tomato species Solanum chilense.
Genetic data analyses revealed a north-south cline. This cline
includes 1) a decrease of genetic variation from north to south, 2)
an increase in the strength of population expansion along the
cline, and 3) an increase in genetic differentiation from other
wild tomato species towards the south of the range. Results further
revealed that the populations form four groups: a central group and
three peripheral groups. Altogether the results suggest that S.
chilense originated in the northern part of its current
distribution and migrated to the south, via two routes, along the
coast and higher up in the Andes. During this north-south
colonization, at least three bottlenecks occurred. In the second
part of this study, I investigated natural selection and local
adaptation in S. chilense. Signatures of selection and local
adaptation were detected in the abiotic stress-related genes, for
example signatures of positive selection in high altitude
populations were found possibly indicating adaptation to low
temperatures. Interestingly, signatures of balancing selection were
detected as well in high altitude populations reflecting probable
adaptation to different types of abiotic stresses. The coastal
populations showed a distinct pattern. Several genes involved in
the salt stress response exhibited signatures of local adaptation.
Performing a salt stress experiment, I revealed that low altitude
populations cope better with such stress than populations from
intermediate or high altitudes. The coastal populations also showed
an accumulation of nonsynonymous and possibly deleterious genetic
variation, which can be explained by extreme bottlenecks and
potential occurrence of selfing in some populations. Signatures of
selection and local adaptation in S. chilense were mainly detected
in populations from the peripheral groups and not in the central
region, in agreement with the hypothesis that local adaptation is
associated with the colonization of new territories. In summary,
this study showed that demography plays an important role in the
evolutionary history of S. chilense and that local adaptation for
key abiotic stresses occurs more frequently in the marginal ranges
of the species distribution.