Corded Ware Culture (CWC) Genetics

Patrilocality and hunter-gatherer-related ancestry of populations in East-Central Europe during the Middle Bronze Age

Abstract
The demographic history of East-Central Europe after the Neolithic period remains poorly explored, despite this region being on the confluence of various ecological zones and cultural entities. Here, the descendants of societies associated with steppe pastoralists form Early Bronze Age were followed by Middle Bronze Age populations displaying unique characteristics. Particularly, the predominance of collective burials, the scale of which, was previously seen only in the Neolithic. The extent to which this re-emergence of older traditions is a result of genetic shift or social changes in the MBA is a subject of debate. Here by analysing 91 newly generated genomes from Bronze Age individuals from present Poland and Ukraine, we discovered that Middle Bronze Age populations were formed by an additional admixture event involving a population with relatively high proportions of genetic component associated with European hunter-gatherers and that their social structure was based on, primarily patrilocal, multigenerational kin-groups.

Introduction
Currently, it is well established that the European gene pool has been shaped by several major demographic events, including the postglacial spread of hunter-gatherers1; subsequent migrations of early farmers, which marked the beginning of the Neolithic in Europe2,3; and the later arrival of Pontic-Caspian steppe pastoralists4,5,6. However, there is still extensive debate surrounding the scale and exact nature of these demographic events and how they affected the genetic makeup of different regions across Europe7.

East-Central Europe, in particular, is a region often on the frontier of these events, resulting in a mosaic of genetically distinct populations associated with a variety of cultural entities. By the turn of the Bronze Age, this region was dominated by populations associated with the Corded Ware Culture (CWC) and Bell Beaker Culture (BBC) and characterised by high levels of steppe ancestry4,5,6,8. Descendants of steppe pastoralists are thought to have replaced and admixed with the late Neolithic populations, which were characterised by large amounts of Anatolian farmer ancestry with additional low to medium levels of hunter-gatherer ancestry9,10. However, the Funnel Beaker Culture (FBC) and Globular Amphora Culture (GAC), major entities predating the arrival of steppe pastoralists associated with the later Neolithic in the region, are thought to have been long lasting, with some of their local variants continuing well into the Early Bronze Age (up to 2000 BCE, in the case of the GAC)11. As no ancient DNA (aDNA) data are available from individuals associated with later populations, their genetic makeup can only be inferred from their 3rd millennium-BC counterparts. The northern and eastern parts of East-Central Europe followed slightly different trajectories, with populations associated with the sub-Neolithic forest zone12,13, characterised by high levels of hunter-gatherer ancestry (and, to a large extent, lifestyle) throughout the Neolithic. In these regions, Anatolian farmer ancestry was introduced along with steppe ancestry during the onset of the Bronze Age14,15. Similar patterns have been observed in other regions surrounding the Baltic Sea, such as southern Scandinavia and Gotland, where hunter-gatherer individuals (both in terms of lifestyle and genetic composition) of the Pitted Ware Culture (PWC) coexisted in close proximity (but without significant gene flow) with the FBC in the 4th millennium BCE and the Battle Axe Culture (BAC), the Scandinavian variant of the CWC, in the first half of the 3rd millennium BCE16,17.

The cultural landscape of Early Bronze Age (EBA) in East-Central Europe (2400–1800 BCE) is widely believed to be a direct continuation of processes that started during the onset of the epoch. For example, the cultural entities present in the region, such as those associated with the Mierzanowice, Iwno and Strzyżów archaeological cultures (MC, IC and SC, respectively)18,19,20, are largely seen as continuations of groups associated with the CWC and BBC21,22,23. In addition, steppe24,25 or the northern forest zone26 cultures have been suggested to have influenced, to some extent, the SC.

The Middle Bronze Age (MBA) in the region (1800–1200 BCE) was in turn dominated by the Trzciniec Cultural Circle (TCC). This cultural phenomenon extended from the Oder River drainage basin to the Desna and Seym River basins (ca. 1200 km) and from the Baltic seashore to the Prut basin (ca. 750 km), exhibiting several territorial variants27. This study focuses on MBA individuals associated with two of these variants (Fig. 1A): the Trzciniec Culture (TC), which occupied the lands belonging to modern-day Poland and central-western Ukraine, and the Komarów Culture (KC), found in modern-day southwestern Ukraine and neighbouring parts of Romania and Moldova27. These MBA cultures retained many cultural aspects of their EBA counterparts, such as styles of pottery and bronze artefacts as well as funeral practices including under-barrow graves and cremation27. This cultural similarity have been shown to coincide with genetic continuity between CWC and both EBA and MBA populations as seen in mitochondrial genome data28. However, some elements of the TCC are unique to the EBA or MBA, particularly the predominance and scale of collective burials. Collective burials of multiple individuals were also prevalent among Middle and Late Neolithic populations in Central and East-Central Europe associated with local variants of the FBC or GAC29,30. Recent studies have shown that these Neolithic collective burials often contain remains of multiple individuals who belonged to, in most cases, patrilocal kin groups31,32,33,34. The biological relatedness among individuals found in these MBA collective burials associated with the TCC, remains unexplored. The presence or absence of kinship among these populations would, however, greatly increase our understanding of the social organization and structure of these societies. The re-emergence of collective burials is the subject of another debate concerning whether this predominant burial custom in the TCC was a result of genetic shifts or social changes within the Bronze Age populations30.

Fig. 1: The geographical and temporal context and genetic affinities of the analysed Bronze Age individuals.


A Maps showing the locations of samples published in this study and the geographical range of their associated cultural entities; the size of the marker corresponds to the number of samples from each site. The map was created using QGIS 2.12.249 and basemap from NOAA National Geophysical Data Center. 2009: ETOPO1 1 Arc-Minute Global Relief Model. NOAA National Centers for Environmental Information. Accessed 2013. B The age of the newly generated genomes (calculated as an average of 2σ BCE dates) corresponding to the temporal range of the archaeological cultures they are associated with. C The results of unsupervised admixture analysis (K = 7) on the selected populations. D PCA plot of ancient individuals projected onto contemporary individuals from West Eurasia from the Human Origins reference panel (not shown). The symbols in both the PCA and admixture analysis correspond to individuals associated with the following cultures: IC - Iwno Culture, KC – Komarów Culture, MC – Mierzanowice Culture, SC – Strzyżów Culture, TC – Trzciniec Culture (from this study and reference populations), AFN – Afanasievo Culture, ALP – Alföld Linear Pottery Culture, AN – Anatolian Neolithic, AND – Andronovo Culture, BABL – Bronze Age Baltic, BAC – Battle Axe Culture, BACz Bronze Age Czechia, BAHU – Bronze Age Hungary, BAP – Bronze Age Poland, BASC – Bronze Age Scandinavia, BBC – Bell Beaker Culture (Poland, Czechia and Germany, respectively), BKG – Brześć Kujawski Group, CAT – Catacomb Culture, CWC – Corded Ware Culture, EBG – Early Bronze Age Germany, EHG – Eastern Hunter Gatherers, ENSt – Eneolithic Steppe, FBC – Funnel Beaker Culture, GAC – Globular Amphora Culture, HGBL – Hunter Gatherer Baltic, LBK – Linear Pottery Culture, LNBG – Late Neolithic/Bronze Age Germany, MNG – Middle Neolithic Germany, NBL – Neolithic Baltic, NEU – Neolithic Ukraine, POL – Poltavka Culture, PWC – Pitted Ware Culture, SHG – Scandinavian Hunter Gatherers, SNT – Sintashta Culture, SRB – Srubnaya Culture, UNC – Únětice Culture, WHG – Western Hunter Gatherers, YAM – Yamnaya Culture.

This study, explores genetic affinities between individuals from various EBA and MBA cultures and their genetic relations to populations of preceding cultural complexes as well as possible kinship structures within MBA societies. To achieve this we conducted population genetic analyses using 91 newly generated genomes from Bronze Age individuals associated with EBA and MBA cultures from modern-day southern and south-eastern Poland and western Ukraine.

Increase in hunter-gatherer ancestry in East-Central Europe in the Middle Bronze Age
The MBA individuals analysed here were dated to a range between 1750 and 1200 BCE and were associated with the TCC, representing both the TC and KC. The majority of these individuals clustered together in PCA space and shared similar admixture proportions (Fig. 1C and D). This apparent genetic relation is further highlighted by f3 and D statistics, which indicate that when analysed separately, KC and TC individuals do not, in majority of cases, display any statistically significant closer genetic affinity to either of the two populations (Supplementary Data 7 and Supplementary Data 12, Supplementary Figs. 3C and 4B). These results are in accordance with the archaeological interpretation that questions the separation of the TC and KC, arguing in favour of treating them as regional variants of the same phenomenon27,36.

Interestingly compared to EBA populations, the MBA individuals were closer in the PCA space to various hunter-gatherer populations from Europe (Fig. 1D), something that previously was not detected in analyses of mitochondrial genome data alone28. Moreover, admixture analysis indicated elevated amounts of genetic components maximised in hunter-gatherers (Fig. 1C). This suggests an additional admixture event at the beginning of the MBA involving a population with relatively high proportions of this genetic component. However, there were notable deviations to this trend, with three individuals associated with TC from Pielgrzymowice site and poz643, a relatively early KC male from Beremiany, clustering closer to EBA populations in PCA space and displaying the lowest levels of shared genetic drift with both TC and hunter-gatherer populations (Supplementary Data 7). When using qpAdm to test for possible two-way admixture models that resulted in the formation of MBA populations, several models were determined to be plausible (Supplementary Data 13) with the highest p value (p = 0.21) obtained for pair consisting of IC and Neolithic Baltic hunter-gatherers (NBL). Similarly high pvalues were found for other pairs including IC and other hunter-gatherer populations: Western Hunter Gatherers (WHG), PWC hunter-gatherers from Gotland, hunter-gatherer buried in BKG context (BKGout) and hunter-gatherer populations predating the NBL (HGBL) (p = 0.207, 0.204, 0.164, 0.160 respectively). These results are in accordance with the archaeological hypothesis that the IC greatly contributed to the emergence of the TC27. However, this finding should be interpreted with caution, as only one individual associated with the IC had sufficient coverage for inclusion in qpAdm. High fit values were also obtained for pair of populations, containing individuals from modern day Estonia associated with the CWC, as the EBA predecessor and PWC, as the source of hunter-gatherer ancestry (p = 0.11).

The process of admixture was likely more complex, as indicated by the three-way admixture model with the addition of a Neolithic population. These models yielded even better fits than the two-way models. Multiple plausible scenarios were found, all displaying high fit values (p > 0.9) including CWCes in addition to various hunter-gatherer and Neolithic populations (Supplementary Data 14). The more discriminatory rotating outgroup approach to qpAdm37 used for both two- and three- way models helped to narrow down the number of plausible scenarios. After excluding the models containing sources consisting of one individual, only three-way scenarios including CWCes and either GAC or FBC as Neolithic and NBL or HGBL as hunter-gatherer population were found to be plausible (Supplementary Data 15 and Supplementary Data 16). Based on geographical and temporal proximity as well as results of D statistics directly comparing the potential ancestry sources for each MBA individual (Supplementary Data 12, Supplementary Fig. 3) we find CWCes, NBL and GAC to be the best proxies for populations involved in the admixture process. However high fit values and close genetic affinity to the individual with hunter-gatherer ancestry buried in BKG context, indicate that further studies might help to better define the populations involved in the process.

The most likely hypothesis is that these admixed MBA populations originated in the confluence of the sub-Neolithic forest zone, associated with populations with dominant WHG ancestry14,15 as well as post-CWC groups characterised by a large proportion of steppe ancestry. The sub-Neolithic forest zone” is a broad term that includes various archaeological cultures from north-eastern Europe, characterised by long-lasting preservation of a predominantly hunter-gatherer lifestyle, and the incorporation of cultural elements of Neolithic and Bronze Age origin38. These populations remained genetically distinct from the Neolithic and post-Neolithic populations, although they maintained some level of long lasting cultural and economic exchange14 It is possible that this lead to some degree of gene flow between those populations, similar to the one observed in the case of PWC in Gotland6,16 followed by subsequent contacts with EBA descendants of steppe pastoralists. Moreover, the TCC and sub-Neolithic forest zone exhibited similar cultural traits, mostly in the form of pottery and technologies12,13,39,40,41. These similarities have often been interpreted as signs of primarily cultural exchange. Our results, showing an increase in WHG ancestry during the MBA, indicate that at least some level of admixture occurred during these interactions.

Notably, the two EBA individuals (poz794 and poz758) associated with the SC, that displayed closer genetic affinity to the MBA populations both came from the south-eastern part of modern-day Poland (Supplementary Text); of these two, the individual with a direct date (poz794) predated the MBA samples analysed here. This observation suggests that the contact zone described above is not the only place where admixture took place and/or that the process was more geographically diffused. Either is plausible, given the range and duration of exchange networks seen in the EBA40. Individual poz794 might even signal the beginning of the observed gene flow, which would date it to approximately 1800 BCE.

The SC is usually seen as a continuation of CWC traditions with additional elements from steppe cultures such as the Catacomb Culture24,25. However, the genetic shift toward an increase in WHG ancestry cannot be explained by additional migration from the steppe, a notion that we previously proposed based solely on mitochondrial data28. This idea is not supported by any indication of increase of steppe ancestry as calculated by three-way qpAdm modelling including WHG, AN and YAM as best proxies of major ancestries in European gene pool (Supplementary Data 17, Supplementary Fig. 4A). Moreover, two-way qpAdm models that explored the scenarios resulting in the emergence of the MBA populations, including EBA individuals and hunter-gatherer populations, yielded higher probabilities than models including additional steppe populations such as Andronovo, Afanasievo, Sintashta, Poltavka, Karasuk, or Srubnaya (Supplementary Data 13).

The process of admixture, which began around 1800 BCE, appears to have been a continuous rather than a result of a single migratory event, as evidenced by the presence of individuals with very high or very low proportions of hunter-gatherer ancestry throughout the whole temporal range of MBA samples analysed here (Fig. 2C, Supplementary Data 17). However, gene flow was likely more extensive in the beginning, as both shared genetic drift (as identified by f3 statistics) and admixture proportions (calculated with qpAdm) show that the proportion of hunter-gatherer ancestry decreased slightly over time (Fig. 2A and Fig. 2C). The results of this event must have been long lasting, as Late Bronze Age individuals from modern-day Latvia and Lithuania14 retain the same genetic composition as our MBA individual despite living nearly half a century later and were found, based on f3 statistics, to have the closest genetic affinity to the MBA individuals presented here out of all Bronze Age populations (Supplementary Data 7 and Supplementary Fig. 3C).