Mitochondrial d-loop variation and DNA preservation in wild and domestic equids (Equus sp.) in Switzerland from the Palaeolithic to the Iron Age

The general aim of the thesis was a chronological genetic investigation of archaeological horse remains from Switzerland including mitochondrial d-loop variation, coat colour and sex identification. Three main subjects were in the focus.
Firstly, mtDNA preservation of archaeologic horse remains in the context of different burial conditions. A systematic synthesis of the influence of different burial conditions on DNA amplification success concerning teeth and bones from open dry and wetland sites and caves from the Pleistocene until Roman times has been made. It led to the conclusion that Pleistocene material from cave and abri (rockshelter) sites is generally genetically well preserved and it was even possible to obtain genetic information from Neolithic waterlogged bones. The depositional environment is the most influential factor affecting DNA preservation. The age of the specimens also plays a crucial role, although it transpired that samples from very favourable conditions, like deep caves, can be better preserved than younger material. Under similar conditions, older samples are less well preserved and accumulated more post mortem damage derived lesion.
Within this project, for the first time all Palaeolithic and Neolithic sites with more than one remain of a certain species, the horse, and a preliminary selection of Iron Age and Roman time samples have been screened for DNA preservation in Switzerland. The outcome of this test is very promising and applicable to other species and further investigations of demographic developments and phenotypic characteristics. The continuous augmentation of data from different depositional contexts and periods is warranted.
Secondly, mtDNA d-loop variation of Pleistocene and Early Holocene wild horse populations was investigated and put in context with palaeoclimatological, palaeoenvironmental and archaeological data. For Switzerland, a discontinuous population history within the last 50 k years was described. The demographic development, an expansion after the LGM, was in disagreement to the development in other parts of Eurasia, particularly north eastern Asia, where abundance peaked during the LGM and decreased from then on. The yet low sample sizes from the transition time Late Pleistocene/Early Holocene allows only tentative speculating on the local dispersal/replacement/extinction pattern of wild horses. Beside the methodological challenges due to the discontinuous and unbalanced representation of equid sequences this analysis provided the first comprehensive investigation of wild horse remains from one restricted region. This approach has offered the opportunity to focus on aspects of horse population development that might be overlooked in the global picture by demonstrating sensitive reaction patterns to changing environmental conditions. To draw a picture of Eurasian horse demographic development, it is necessary to follow a comprehensive geographically and chronologically dense sampling approach, comparable to the Swiss example.
Thirdly, we examined mtDNA d-loop variation and coat colour of Iron Age domestic horses. A possible genetic differentiation of morphologically different animals was enquired via matrilinear diversity, and the investigation of coat colouration served to detect phenotypical noticeable individuals and to relate their incidence to the archaeological context. Female lineages were highly diverse, yet coat colourations were limited to the basic colours bay, black and chestnut. The few larger horses in our dataset were not genetically distinct from the smaller majority. Breeding interests were apparently focussed on good performance and low maintenance horses, and to avoid inbreeding. This emphasises the skill and expertise Celtic stock keepers executed on this highly valuable and prestigious animal.
The lineage diversity present in Iron Age Switzerland does not differ from (roughly) contemporaneous variation in the rest of Europe except for Iberia and is particularly similar to eastern Europe. This finding supports the hypothesis, based on archaeological evidence of e.g. eastern European type bridle fragments, of an eastern origin of Swiss domestic horses. A contribution of local wild mares to the domestic gene pool can be ranked low, not least because they were most probably on the edge of extinction by the 3rd millennium BC.
The contextualisation of Swiss wild and ancient domestic horses with the Eurasian variation revealed that, at our current state of knowledge, domestics had higher nucleotide diversity than wild horses. This observation contradicts the general ideas of domestication (bottleneck) and of the velocity of mutation rates (within 5 ka since domestication started). Predictably this ratio will be adjusted as more pre-domestic specimens from Eurasia will be sequenced. Recent technical progress and leaps in the understanding of wild horse genetics (Orlando et al., 2013, Schubert et al., 2014, Imsland et al., 2016) as well as regional approaches to chronological genetic investigations as presented here, are heralding fascinating new insights into the evolution and history of a species as much appreciated as the horse.