Adapting to extreme environments: rapid evolution of heavy metal tolerance in Silene

Abstract

We live in a rapidly changing world, with species under threat from human activities across the world. Understanding the factors that facilitate adaptation and whether the process is predictable is vital to estimate the impact of ongoing anthropogenic activity on species. In this thesis, I aimed to elucidate the factors facilitating rapid adaptation to extreme, polluted environments. Specifically, I focussed on the relative role of phenotypic plasticity in speeding up adaptation as well as the contribution of introgression as a source of adaptive alleles. Additionally, I aimed to understand the extent to which repeated adaptation to the same environment is accompanied by the reuse of similar phenotypes and genotypes. I studied the genomics and transcriptomics of repeated heavy metal adaptation across two closely related Silene species, Silene uniflora and Silene vulgaris, to meet these aims. Both species have adapted to old mine slag heaps contaminated with heavy metals such as zinc, lead and copper.
From genome-wide selection scans, I found a highly polygenic basis for rapid adaptation across both species. Following the exposure of ancestral coastal and descendent mine-adapted populations to ancestral and descendent cues (zinc and salt), I showed that ancestral gene expression plasticity facilitates rapid adaptation. Particularly, I demonstrated that plasticity in response to the past cue of salt in the ancestral coastal habitat facilitates subsequent adaptation to zinc pollution. This represents a new route via which plasticity contributes to adaptation. I investigated the contribution of introgression from S. vulgaris into S. uniflora to novel adaptation in S. uniflora, using genome-wide introgression and selection scans. Adaptive introgression was limited in scope to a few loci across small locations in the genome, potentially representing old events. Introgression may therefore represent a limited source of adaptive genetic variation during rapid, polygenic adaptation, and instead novel mutation and/or standing genetic variation may be more important. Using further modelling to detect these sources of genetic variation may shed light on the processes that are most important in speeding up adaptation.
To assess the predictability of adaptation, I found common outliers from FST genome scans across repeated mine versus non-mine divergence events within S. uniflora and between S. uniflora and S. vulgaris. I estimated the degree of parallelism at different biological levels, from shared selection at the same genetic variants, such as individual polymorphisms and small genomic windows, to genes under selection, to the phenotypes involved in adaptation (estimated gene functions). The degree of parallelism between phenotypes under selection was greater than for genes and genetic variants, with genetic variants showing the lowest degree of parallelism. This pattern held both within and between species. Therefore, adaptation is more predictable for phenotypes than for genotypes, even between different species. This pattern is likely caused by genetic redundancy underlying phenotypes. Overall, the parallelism degree was lower between species than within species, which suggests that future adaptation is easier to predict using data from the same species rather than from different one. Further investigation of the extent of parallelism within S. vulgaris as a species and the sources of genetic variation contributing to this could further our understanding of the factors affecting the predictability of adaptation.
Overall, adaptation can be highly rapid to novel environments and appears to be predictable to an extent. Therefore, multiple species may be able to adapt as anthropogenic interference continues to affect the globe. However, furthering our understanding of the factors affecting adaptation speed and future work to assist in predicting adaptation based on genomic information is needed before we can fully anticipate how successfully species will adapt to future environmental change.

Date of Award	13 Apr 2026
Original language	English
Awarding Institution	Bangor University
Supervisor	Alexander Papadopulos (Supervisor), Aaron Comeault (Supervisor), Simon Creer (Supervisor) & Michael Fay (Supervisor)