pubmed.ncbi.nlm.nih.gov

The source of the Black Death in fourteenth-century central Eurasia - PubMed

The source of the Black Death in fourteenth-century central Eurasia

Maria A Spyrou et al. Nature. 2022 Jun.

Abstract

The origin of the medieval Black Death pandemic (AD 1346-1353) has been a topic of continuous investigation because of the pandemic's extensive demographic impact and long-lasting consequences1,2. Until now, the most debated archaeological evidence potentially associated with the pandemic's initiation derives from cemeteries located near Lake Issyk-Kul of modern-day Kyrgyzstan1,3-9. These sites are thought to have housed victims of a fourteenth-century epidemic as tombstone inscriptions directly dated to 1338-1339 state 'pestilence' as the cause of death for the buried individuals9. Here we report ancient DNA data from seven individuals exhumed from two of these cemeteries, Kara-Djigach and Burana. Our synthesis of archaeological, historical and ancient genomic data shows a clear involvement of the plague bacterium Yersinia pestis in this epidemic event. Two reconstructed ancient Y. pestis genomes represent a single strain and are identified as the most recent common ancestor of a major diversification commonly associated with the pandemic's emergence, here dated to the first half of the fourteenth century. Comparisons with present-day diversity from Y. pestis reservoirs in the extended Tian Shan region support a local emergence of the recovered ancient strain. Through multiple lines of evidence, our data support an early fourteenth-century source of the second plague pandemic in central Eurasia.

© 2022. The Author(s).

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1. Description of the investigated fourteenth-century Chüy Valley archaeological sites.

a, Locations of the Kara-Djigach and Burana archaeological sites in modern-day Kyrgyzstan. Regions encompassing Y. pestis foci at present are highlighted in blue (as in refs. ,). The map was created using QGIS v.3.22.1 (ref. ) and uses Natural Earth vector map data from

https://www.naturalearthdata.com/

. b, Area within the Kara-Djigach cemetery, referred to as ‘Chapel 1’ with the highest concentration of excavated burials dating between 1338 and 1339. Burial dates were determined on the basis of their associated tombstones (Supplementary Information 2). The site map has been redrawn based on the original created by N. Pantusov in 1885. Individuals from graves 6, 9, 20, 22 and 28 (the numbers in bold) were investigated using aDNA in this study. Burials shown with stripe patterns were associated with individuals BSK001, BSK003 and BSK007, which showed evidence of Y. pestis infections. c, Annual numbers of tombstones from Kara-Djigach (n = 456) and Burana (n = 11) (Supplementary Table 1). Dataset updated from ref. (see Supplementary Information 2 for details). d, Tombstone from the Kara-Djigach cemetery with legible pestilence-associated inscription. The inscription is translated as ‘In the Year 1649 [=

ad

 1338], and it was the Year of the Tiger, in Turkic Bars. This is the tomb of the believer Sanmaq. [He] died of pestilence [=mawtānā]’. For a tracing of the inscription, see Extended Data Fig. 1.

Fig. 2
Fig. 2. Comparisons between BSK001/003 and published Y. pestis genomic diversity.

a, Map of all historical Y. pestis genomes used in the present study (n = 48). The colours represent different genome ages on a scale between 1300 and 1800, as depicted in b. The colour scale is maintained across all panels of this figure. To aid visibility in overlapping symbols, a jitter option was implemented for plotting genomes on the map. The map was created with QGIS v.3.22.1 (ref. ) and uses Natural Earth vector map data from

https://www.naturalearthdata.com/

. b, Y. pestis maximum likelihood phylogeny based on 2,960 SNPs, visualized using GrapeTree. The depicted portion of the phylogeny contains the closest related lineages to BSK001/003. (For a fully labelled tree, see Extended Data Fig. 5). The colours of published historical strains are consistent with a. The scale denotes the number of substitutions per genomic site. c, Abundance of diagnostic SNP sharing in fourteenth-century Y. pestis genomes. The number of diagnostic SNPs (n) shared between all modern genomes on branch 1, and therefore defining this branch, were retrieved from a comparative SNP table of 203 modern Y. pestis genomes. SNP sharing was assessed by determining the allele status of each diagnostic position according to a threefold SNP calling threshold. The error bars denote the degree of missing data (n) in the respective ancient genome. Refer to Extended Data Fig. 6 and Supplementary Table 18 for an overview of diagnostic SNP sharing on different phylogenetic branches.

Fig. 3
Fig. 3. Molecular dating of Y. pestis branches 1–4.

a, Maximum clade credibility time-calibrated phylogenetic tree. The tree is based on 167 genomes (historical and modern) and was estimated using the coalescent skyline tree prior and a log-normal relaxed clock. Collapsed branches contain modern and ancient isolates dating after

AD

1400 (post-Black Death). The coloured arrows mark the nodes, for which equivalent posterior age distributions are shown in b. The estimated divergence dates (95% HPD intervals) of modern branches are shown on each corresponding node. b, Estimated posterior distributions based on the coalescent Bayesian skyline tree prior for the divergence of Y. pestis branches 1–4 (blue), for the estimated divergence of BSK001/003 (purple) and for the entire dataset used for this analysis (time to the most recent common ancestor of branches 1–4 and 0.ANT3, shown in grey). The dotted lines indicate mean posterior estimates and are annotated with the corresponding 95% HPD intervals.

Fig. 4
Fig. 4. Geographical isolation locations of modern 0.ANT lineages.

a, Maximum likelihood phylogenetic tree, based on 2,441 genome-wide variant positions. The tree was constructed to indicate the genetic relationships between available 0.ANT genomes depicted on the map and BSK001/003. Modern branches were collapsed to enhance tree clarity (see Extended Data Fig. 8 for a full tree). b, Map depicting the geographical isolation locations of 0.ANT strains (Supplementary Table 21), which belong to the closest ancestral branching lineages to the Kara-Djigach strain. The map includes both whole-genome data (further specified as 0.ANT lineages 1, 2, 3 and 5) and PCR-genotyped isolates that are broadly defined as 0.ANT, belonging to any of the 4 lineages. For strains in which exact geographical coordinates were unavailable, locations were approximated according to their associated plague reservoirs. To aid visibility in overlapping symbols, a jitter option was implemented for plotting objects on the map. The map was created with QGIS v.3.22.1 (ref. ) and uses Natural Earth vector map data from

https://www.naturalearthdata.com/

.

Extended Data Fig. 1
Extended Data Fig. 1. Available tombstone pictures from Kara-Djigach.

a–c, Available tombstone pictures from individuals investigated as part of this study. For a translation of the tombstone inscriptions, see individual descriptions within Supplementary Information 2. Tombstone dates are as follows: Grave 9 (1338-9 CE), Grave 19/20 (1338-9 CE), Grave 6 (year not inscribed). Picture credits to P.-G. Borbone. d+e, Tombstones identified in Kara-Djigach containing pestilence-stating inscriptions, dating to the years 1338 and 1339 CE. These tombstones do not correspond to individuals analysed within our aDNA dataset. The original tombstone on panel d (without traced inscription) is shown in Fig. 1. Complete  translations are available within Supplementary Information 2.

Extended Data Fig. 2
Extended Data Fig. 2. Ancient DNA damage substitution frequencies for all Y. pestis captured libraries.

C-to-T substitution frequencies characteristic of post-mortem deamination of ancient DNA are shown for the 5′ ends of sequenced reads aligned against the CO92 Y. pestis reference genome (NC_003143.1).

Extended Data Fig. 3
Extended Data Fig. 3. Evaluation of BSK007 after whole-genome Y. pestis capture.

a, Post-capture coverage distribution of BSK007 across the Y. pestis CO92 chromosome. Mean coverage was estimated across the genome in 4,000 bp windows. The dotted gray line indicates the mean coverage across the entire genome (0.125-fold). b, Krona plots showing the taxonomic classification of BSK007 reads mapping against all Y. pestis CO92 elements (chromosome NC_003143.1, pMT1 NC_003134.1, pPCP1 NC_003132.1 and pCD1 NC_003131.1). Numbers in brackets next to element designations correspond to the number of assigned reads in MALT. The colours of Krona sectors represent different taxonomic levels and their completeness is proportional to the relative abundance of summarised reads at each corresponding taxonomic node. The shown percentages indicate the species-level (outermost circle) proportion of reads aligned to taxa other than Y. pestis and Y. pseudotuberculosis.

Extended Data Fig. 4
Extended Data Fig. 4. Read length and ancient DNA damage distribution of contaminant SNP regions in the combined BSK001/003 dataset prior to metagenomic filtering.

Overlayed length distributions of reads mapping against the CO92 Y. pestis reference genome, calculated for the entire dataset (gray) as well as for 117 regions surrounding putatively contaminant SNPs (blue). Regions were extracted within a 150 bp window surrounding each putatively contaminant SNP. Dotted lines represent average fragment lengths for the entire dataset and for the 117 putatively contaminant SNP regions in gray and blue, respectively. Reads comprising contaminant SNP regions show a distinct length distribution compared to the one observed across the entire BSK001/003 genome, with a marked shift towards longer read lengths. The 76 bp fragment length peak represents the uppermost possible read length of single-end sequenced reads, which comprised the majority of data within the present dataset. Ancient DNA damage patterns were compared between the entire dataset (upper panel) and the putatively contaminant SNP regions (lower panel), showing a near 3-fold reduction in the latter as estimated for the terminal 5′ base.

Extended Data Fig. 5
Extended Data Fig. 5. Phylogenetic comparisons between BSK001/003 against ancient and modern Y. pestis diversity.

Full length maximum likelihood phylogenetic tree using 1,000 bootstrap iterations for estimating node support and visualised using FigTree v1.4.4. The tree is constructed with 203 modern and 48 historical Y. pestis genomes, and is based on 2,960 SNPs (96% partial deletion). Scale denotes the number of substitutions per genomic site.

Extended Data Fig. 6
Extended Data Fig. 6. Evaluation of phylogenetically diagnostic SNPs across 14th century Y. pestis genomes.

The estimated variant calls were retrieved from a SNP table comprising 203 modern and 48 historical Y. pestis genomes (full dataset contains 3,533 SNPs). Error bars indicate uncertainty due to the presence of missing data (Ns) within the variant calls of the respective genomes.

Extended Data Fig. 7
Extended Data Fig. 7. Comparison of phylogenetic diversity measures computed for the complete modern Y. pestis phylogeny against the Branch 1-4 subclade.

The left panels show maximum likelihood substitution trees considering only extant Y. pestis genomes, annotated based on the compared clades and their corresponding Faith’s phylogenetic diversity index (FPD). Phylogenetic branches considered for the Branch 1-4 FPD computation are shown in green (full dataset) and red (subsampled dataset). The right panels show violin plots indicating the distribution and mean of pairwise phylogenetic distances based on maximum likelihood trees (MPD). Median estimates and 95% percentile intervals were derived from the resampled bootstrap trees (1,000 bootstrap iterations). Points within violin plots indicate the mean estimated phylogenetic distance for all datasets.

Extended Data Fig. 8
Extended Data Fig. 8. Phylogenetic relationships of 0.ANT lineages.

Maximum likelihood phylogenetic tree based on 2,441 genome-wide variant positions (all SNPs). The tree was constructed to indicate the genetic relationships between all previously published extant 0.ANT genomes and BSK001/003. Scale denotes the number of substitutions per genomic site. Node support was determined by 1,000 bootstrap iterations.

Comment in

  • Ancient DNA traces origin of Black Death.

    Callaway E. Callaway E. Nature. 2022 Jun;606(7915):635-636. doi: 10.1038/d41586-022-01673-4. Nature. 2022. PMID: 35705867 No abstract available.

  • Black Death unravelled.

    Gonzalez Lopez A. Gonzalez Lopez A. Nat Rev Microbiol. 2022 Sep;20(9):509. doi: 10.1038/s41579-022-00769-y. Nat Rev Microbiol. 2022. PMID: 35778565 No abstract available.

Similar articles

  • Historical Y. pestis Genomes Reveal the European Black Death as the Source of Ancient and Modern Plague Pandemics.

    Spyrou MA, Tukhbatova RI, Feldman M, Drath J, Kacki S, Beltrán de Heredia J, Arnold S, Sitdikov AG, Castex D, Wahl J, Gazimzyanov IR, Nurgaliev DK, Herbig A, Bos KI, Krause J. Spyrou MA, et al. Cell Host Microbe. 2016 Jun 8;19(6):874-81. doi: 10.1016/j.chom.2016.05.012. Cell Host Microbe. 2016. PMID: 27281573

  • The Stone Age Plague and Its Persistence in Eurasia.

    Andrades Valtueña A, Mittnik A, Key FM, Haak W, Allmäe R, Belinskij A, Daubaras M, Feldman M, Jankauskas R, Janković I, Massy K, Novak M, Pfrengle S, Reinhold S, Šlaus M, Spyrou MA, Szécsényi-Nagy A, Tõrv M, Hansen S, Bos KI, Stockhammer PW, Herbig A, Krause J. Andrades Valtueña A, et al. Curr Biol. 2017 Dec 4;27(23):3683-3691.e8. doi: 10.1016/j.cub.2017.10.025. Epub 2017 Nov 22. Curr Biol. 2017. PMID: 29174893

  • Integrative approach using Yersinia pestis genomes to revisit the historical landscape of plague during the Medieval Period.

    Namouchi A, Guellil M, Kersten O, Hänsch S, Ottoni C, Schmid BV, Pacciani E, Quaglia L, Vermunt M, Bauer EL, Derrick M, Jensen AØ, Kacki S, Cohn SK Jr, Stenseth NC, Bramanti B. Namouchi A, et al. Proc Natl Acad Sci U S A. 2018 Dec 11;115(50):E11790-E11797. doi: 10.1073/pnas.1812865115. Epub 2018 Nov 26. Proc Natl Acad Sci U S A. 2018. PMID: 30478041 Free PMC article.

  • Molecular history of plague.

    Drancourt M, Raoult D. Drancourt M, et al. Clin Microbiol Infect. 2016 Nov;22(11):911-915. doi: 10.1016/j.cmi.2016.08.031. Epub 2016 Sep 8. Clin Microbiol Infect. 2016. PMID: 27615720 Review.

  • Yersinia pestis: the Natural History of Plague.

    Barbieri R, Signoli M, Chevé D, Costedoat C, Tzortzis S, Aboudharam G, Raoult D, Drancourt M. Barbieri R, et al. Clin Microbiol Rev. 2020 Dec 9;34(1):e00044-19. doi: 10.1128/CMR.00044-19. Print 2020 Dec 16. Clin Microbiol Rev. 2020. PMID: 33298527 Free PMC article. Review.

Cited by

References

    1. Benedictow, O. J. The Black Death, 1346–1353: the Complete History (Boydell & Brewer, 2004).
    1. Clark, G. A Farewell to Alms: A Brief Economic History of the World Vol. 25 (Princeton Univ. Press, 2008).
    1. Pollitzer, R. Plague (World Health Organization, 1954). - PubMed
    1. Norris J. East or west? The geographic origin of the Black Death. Bull. Hist. Med. 1977;51:1–24. - PubMed
    1. Dols, M. W. The Black Death in the Middle East (Princeton Univ. Press, 1979).

Publication types

MeSH terms

Substances