Population genetic history of the British roe deer (Capreolus capreolus) and its implications for diversity and fitness

Abstract

The first part of this study examined post glacial recolonisation by UK roe. Previous studies established three main roe deer lineages exist across Europe: a western (Iberian Peninsula), an eastern (Balkan region) and a central lineage (which spans across central Europe). It was unknown which group British roe deer populations belonged. Using a 419 bp region of the mt-DNA d-loop (HVR1) amplified from ancient and modern UK samples a direct comparison was made with previously published European data. Results showed that UK populations belong to the central lineage, indicating a post glacial re-colonisation that is likely to have occurred via an eastern route. The estimation of a substitution rate, which was applied to coalescent based methods, detected a signal for divergence of UK roe from continental roe at 5,600 YBP (HPD 3,500 - 11, 200 YBP), not long after the proposed date for the land bridge split (7,500 YBP).
Since post glacial re-colonisation, roe were known to have undergone severe fluctuations in population size. Perhaps the most significant fluctuation began during the medieval period, when roe suffered severe declines (bottlenecking) due to over hunting and deforestation. These declines were so severe that, by the 16th century, roe were believed to have been extirpated (locally extinct) from all southern areas of UK and considered scarce in northern areas. However, by the 19th century roe began to recover. Recovery in the south may have resulted solely from re-introductions (involving both native and non-native stocks) whilst, in the north, recovery resulted from natural re-colonisation from remnant native stocks. The second part of this study investigated the impacts of this more recent history.
This was first investigated using a 750 bp of the mt-DNA d loop region (HVR), 16 microsatellite loci and 18 skull traits from modern roe from across the UK to examine structure and diversity. Results based on both DNA and morphology revealed strong differentiation. Northern roe appeared least impacted by recent events; maintaining patterns of isolation by distance (IBD) and high genetic diversity (compared to southern populations). In contrast, southern roe appeared more strongly impacted by recent events; in particular, IBD was non-significant (although this may have been due to a sample size effect) and genetic diversity was lower (compared to northern populations). The roe re-introduction records indicated that the south western population was native in origin (Perthshire). Genetic data showed that this population was, however, highly differentiated from its proposed source; which could reflect the powerful impact of genetic drift resulting from small founder populations. Alternatively, it may be that the ancestry of the south western population is more complex than previously assumed. For the other southern population (Norfolk), re-introduction records indicate a non-native (German) origin. In line with this, both genetic and morphological data implied that these roe were highly distinct.
The impacts of bottlenecks (including medieval declines and founder events) on roe populations were also examined. Bottleneck analyses examined ‘signatures’ in modern populations based on 16 microsatellites. The strongest evidence of bottlenecking was detected in the Norfolk population, consistent with the small founder group size introduced into this location relatively recently. For the other populations bottleneck signatures tended to be weak and non-significant. Direct comparisons of ancient (pre-bottleneck) and modern (post –bottleneck) populations were made based on 419 bp of mt-DNA d –loop (HVR1). Results showed considerable losses in genetic diversity between time frames consistent with medieval declines. Northern populations were also found to harbour the highest number of ‘native’ haplotypes and southern populations the lowest. The southern population of Norfolk exhibited only one ‘novel’ haplotype confirming its non-native origin.
The impacts of bottlenecks on populations are of concern because they have been shown to reduce population fitness and increase the risk of extinction. Therefore, fitness of roe was examined using fluctuating asymmetry (FA) of 10 skull traits as an indicator of developmental stability. Correlations of FA and genetic diversity indices were examined at the level of individuals within populations, across all populations and among populations. All correlations existed in expected directions; however, correlations tended to be weak and non-significant. Furthermore, among population level FA did not vary significantly across populations providing no indication as to whether fitness has been impacted by past population history.