Ancient DNA analysis has revealed an involvement of the bacterial pathogen Yersinia pestis in several historical pandemics, including the second plague pandemic (Europe, mid-14 th century Black Death until the mid-18 th century AD). Here we present reconstructed Y. pestis genomes from plague victims of the Black Death and two subsequent historical outbreaks spanning Europe and its vicinity, namely Barcelona, Spain (1300–1420 cal AD), Bolgar City, Russia (1362–1400 AD), and Ellwangen, Germany (1485–1627 cal AD). Our results provide support for (1) a single entry of Y. pestis in Europe during the Black Death, (2) a wave of plague that traveled toward Asia to later become the source population for contemporary worldwide epidemics, and (3) the presence of an historical European plague focus involved in post-Black Death outbreaks that is now likely extinct.

Here, we aim to address three outstanding questions regarding Y. pestis history. First, we investigate the possibility of disease entry via multiple pulses during the Black Death by comparing the genotype of a strain from the pandemic’s early phase to those circulating in other areas later in the pandemic. Material from Barcelona, Spain, one of the Mediterranean cities through which plague entered southern continental Europe, is compared to Black Death genomes from London. Second, we evaluate the likelihood of the proposed eastward migration of strains from Europe to Asia after the Black Death through the analysis of human remains from a 14century plague burial in the Volga region of Russia. Third, we take a further step toward understanding the relationship of post-Black Death outbreaks in Europe and evaluate the likelihood of a local reservoir. For this, we investigate a 16century plague outbreak in Southwestern Germany and compare it to both a London outbreak that occurred soon after the Black Death and to the Great Plague of Marseille, France in 1722. Following the success of previous genomic investigations of ancient bacterial disease (), we employ similar methods of DNA capture and high-throughput sequencing to retrieve the genomes of three historical Y. pestis strains. Our results suggest (1) limited Y. pestis diversity during the early phase of the Black Death, and likely a single entry into Europe; (2) a wave of plague that traveled eastward after the Black Death and later gave rise to the 19century pandemic; and (3) an involvement of the same plague lineage in two post-Black Death European epidemics that are 200 years apart.

After the Black Death, plague continued to strike Europe for another four centuries through subsequent outbreaks that ceased at the end of the 18century (). The reasons for its sudden disappearance in Europe are unknown. Sylvatic plague foci have a nearly worldwide presence today, but are absent in Europe (). The question of whether the recurrent European plague outbreaks of the 14to 18centuries were the result of multiple reintroductions of plague into Europe, or rather were attributed to now-extinct European plague foci, is still being explored. Previous studies that draw upon aDNA and climatic data favor the former hypothesis. Through a SNP-based PCR approach, purportedly distinct plague lineages were identified in different areas of Europe during the 14century and were thought to have entered via different pulses (). In addition, plague outbreaks documented in some of the main Mediterranean ports were found to coincide with extreme climate fluctuations in Central Asia, suggesting that recurrent maritime imports of plague from Asia might have been responsible for post-Black Death plague outbreaks (). By contrast, others have suggested a long-term persistence of plague in Europe (). Using a PCR SNP-typing approach of putative plague material from Southern and Northeastern Germany, identical Y. pestis SNP profiles were identified in strains circulating within Europe between the Black Death and 17century AD (), implying a single source population for the European plagues of that time period. A further genome-wide analysis of Y. pestis strains from the Great Plague of Marseille (1720–1722) has identified a previously uncharacterized lineage of Y. pestis that descends from a strain present during the Black Death (). While the lineage is considered to represent an historical plague focus potentially responsible for post-Black Death European outbreaks (), the use of material from a highly operational Mediterranean center that linked Western Europe with the East () makes identification of the disease source elusive.

The Black Death claimed up to 50% of the European population between 1347 and 1353 (). The disease is thought to have arisen from plague foci in East Asia and to have spread into Europe via trade routes (). Its origin, however, is still contentious due to a lack of convincing archeological or documentary evidence from the early 14century in East Asia (). Ancient Y. pestis genomes obtained from medieval victims have indicated the presence of a radiation event immediately preceding the Black Death that gave rise to most of the strain diversity circulating in the world today (). Based on the relationship of ancient European and modern genomes, it was recently suggested that a wave of plague might have traveled from Europe toward Asia after the Black Death, eventually settling in China and later giving rise to the third pandemic (). Genomes from its purported route are, however, missing in the discussions, and are needed to add legitimacy to the model.

Yersinia pestis evolved from the closely related zoonotic enterobacterium Y. pseudotuberculosis () to become one of the most virulent pathogens known to humans. Its recent identification in ancient human material from Altai, Siberia suggests it caused human infections as early as 5,000 years ago, though its ability for flea-borne transmission leading to bubonic disease might have been absent in these older, divergent lineages (). To our knowledge, bubonic plague, and presumably also the pneumonic and septicemic forms, have been the likely culprit of three major pandemics, namely the Plague of Justinian (Eastern Roman Empire, 6and 8centuries AD), the second-wave plague pandemic (Europe, mid-14century Black Death until the mid-18century AD), and the third plague pandemic that started during the late 19century in China. Differences in mortality rate and epidemiology of the three pandemics initiated controversy over whether they shared a common etiologic agent (). In recent years, however, ancient DNA (aDNA) has confirmed a Y. pestis involvement in both historical pandemics ().

We find no detectable differences between our Black Death strain from Barcelona and three previously genotyped strains from London 1348–1350 (). The Bolgar City strain, however, contains additional differences in four positions compared to Black Death isolates: two of these are shared with London individual 6330 (positions p3 and p4, Figure 2 B and Table 1 ), one is shared with all modern Branch 1 strains (p6), and one is unique to this individual (p7, Figure 2 B). Additionally, the Ellwangen strain groups in a sub-branch of Branch 1, together with five strains previously typed from the Great Plague of Marseille (L’Observance), 1720–1722 ( Figure 2 B) (). Our analysis reveals 20 positions shared with the strains from L’Observance and three unique SNPs ( Table S4 ). That the Ellwangen strain is ancestral to the Observance strains comes as no surprise given the older age of the samples ( Figure 2 B). This “Ellwangen-Observance” lineage originates from Black Death strains currently represented by the isolates from London and Barcelona. Like the strain from Marseille, that from Ellwangen does not share additional derived positions with other ancient or modern strains ( Figure 2 B), as no modern descendants have as yet been identified in this sub-branch.

All three reconstructed historical genomes group in Branch 1, and all possess diagnostic SNP positions here referred to as “p1” and “p2” ( Table 1 ), which were previously identified in historical Y. pestis genomes from the Black Death () ( Figure 2 B, Table 1 ). The positioning of the strains reported here in the phylogeny confirms their authenticity as ancient. To date, all Y. pestis genomes isolated from the historic 2plague pandemic group in Branch 1.

(B) A magnified version of Branch 1 is shown to enhance its resolution. The branch of lineage 1.ANT was manually reduced to adjust tree scaling. A detailed description of p1-p7 SNPs is given in Table 1 (see also Table S2 Table S4 and Figure S2 ).

(A) Maximum Parsimony phylogenetic tree of 141 modern and 10 historical Y. pestis strains. 3,351 SNP positions were considered for the phylogeny. The reconstructed tree shows the topology of the new isolates from Barcelona, Bolgar City, and Ellwangen relative to previously sequenced modern and ancient Y. pestis strains. Asterisks ( ∗ ) indicate bootstrap values of 100. Collapsed branches are represented by triangles, to enhance tree clarity. Strains belonging to Branch 1 are represented in red, Branch 2 in yellow, Branch 3 in blue, Branch 4 in orange, and Branch 0 in black. Ancient Branch 1 strains are indicated by their archaeological or radiocarbon date and by a (+). Because of the great number of derived SNP positions of the 0.PE3 lineage, its branch was reduced to adjust scaling of the tree. Geographic region abbreviations correspond to: CHN (China), USA (United States of America), MDG (Madagascar), IND (India), IRN (Iran), MNM (Myanmar), RUS (Russia), GB (Great Britain), DE (Germany), FRA (France), SP (Spain), MNG (Mongolia), NPL (Nepal), FSU (Former Soviet Union), AGO (Angola), CGO (Congo), and UGA (Uganda).

Our ancient genomes were then added to a Y. pestis phylogeny constructed from previously published genomes including 130 modern genomes (), 7 historical genomes (), and 11 newly available modern Y. pestis strains from the Former Soviet Union () ( Table S3 ). Our maximum parsimony tree revealed that the modern Former Soviet Union genomes group with what was previously thought to be diversity restricted in China, specifically lineage 0.ANT3 (). They also add further diversity to the 2.MED1 lineage and, importantly, to the 0.PE2 lineage, which is the second deepest branch in the Y. pestis phylogeny ( Figure 2 A, Figure S2 , and Table S3 ). This reveals a more extensive Y. pestis diversity outside of China than was previously thought.

Whole-genome array capture was performed using the chromosome of Y. pseudotuberculosis () and the Y. pestis plasmids pMT1 and pCD1 as template for probe design ( Supplemental Experimental Procedures ). Array captures produced average genomic coverage of 10.3-fold for Barcelona 3031, 19.3-fold for Bolgar City 2370, and 4.9-fold for Ellwangen 549_O ( Table S1 and Table S2 ). Owing to its low coverage, data presented for sample 549_O are from a pool of two independent libraries produced from two teeth of the same individual ( Table S1 and Table S2 ).

A total of 223 DNA extracts from teeth of 178 individuals were evaluated for the presence of Y. pestis DNA through a species-specific quantitative PCR (qPCR) assay targeting the plasminogen activator (pla) gene located on the PCP1 plasmid () ( Supplemental Experimental Procedures ). Results indicated 53 potentially positive DNA extracts stemming from 32 individuals. All extraction and PCR blanks were free of amplification products. Amplification products were not sequenced, as samples from potentially positive individuals were directly turned into double-stranded next-generation sequencing libraries and were used for whole-genome array capture. After capture, three individuals had sufficient Y. pestis DNA for genome-level analysis. These were tooth specimens 3031 from Barcelona, 2370 from Bolgar City, and 549_O from Ellwangen ( Figure 1 Table S1 and Supplemental Experimental Procedures ).

Samples were collected from a mass grave in Barcelona, Spain, a single grave in Bolgar City in Russia, and a mass grave in Ellwangen, Germany ( Figure 1 and Supplemental Experimental Procedures ). Aside from the Bolgar City site that was dated to the second half of the 14century using coin artifacts known to have been minted after 1362 ( Supplemental Experimental Procedures and Figure S1 ), archaeological dates were not available. To estimate or confirm the historical period during which each of the outbreaks occurred, radiocarbon dates from bone fragments and tooth roots were obtained. The dates yielded were 1300–1420 cal AD for Barcelona, 1298–1388 cal AD for Bolgar City, and 1486–1627 cal AD for Ellwangen ( Figure 1 and Table S1 ).

Discussion

Figure 3 Plague Introduction and Dispersal Show full caption th century, eastward travel out of Europe after the Black Death, and global dissemination from China during the third plague pandemic (see also Map describing our favored dissemination pattern of Y. pestis during the second and third plague pandemics. All strains included in our dataset are depicted as points on the map. Branch 1 strains are in red and include both second pandemic (triangles) and modern (circles) isolates. Branch 2 strains are in yellow, Branch 3 strains are in blue, a single Branch 4 strain is in orange, and Branch 0 strains are in black. Positioning of modern strain distribution on the map corresponds to geographic location, but for the purpose of our study an accurate coordinate system was not necessary. Red arrows indicate Branch 1 cycling through Europe during the 14century, eastward travel out of Europe after the Black Death, and global dissemination from China during the third plague pandemic (see also Table S3 ). Our genomes from Barcelona, Bolgar City, and Ellwangen group on the same phylogenetic branch (Branch 1), adding further legitimacy to the notion that the Black Death and subsequent plague outbreaks in Europe, as well as the worldwide third pandemic, were caused by the same Y. pestis lineage ( Figure 2 Figure S2 , and Figure 3 ). Further analysis of ancient and modern strains of this branch could reveal important clues to explain why this particular lineage was involved in both the second and third pandemic.

Gottfried, 1983 Gottfried R.S. The Black Death: Natural and Human Disaster in Medieval Europe. Benedictow, 2004 Benedictow O.J. The Black Death, 1346-1353: the complete history. Our analysis reveals that the strain from Barcelona is identical to a previously sequenced Black Death Y. pestis strain from London (1348–1350). Barcelona was one of the main entry points for the Black Death into Europe, with historical reports suggesting the disease first entered there in the spring of 1348 (). In London, the earliest reports of the illness are from autumn 1348 (). This indicates a contemporary presence of the same strain in both southern and northern Europe, supporting the notion of a single wave entry, with low genetic diversity in the pathogen. Historical sources indicate that plague first came into view in 1347, with outbreaks in the southern islands of Crete, Sicily, and Sardinia, followed by entry into mainland Europe via the heavily trafficked ports of Genoa and Marseille. Samples from these locations and those further afield from its purported source population in East Asia may provide us with relevant details regarding the microevolution of a highly virulent pathogen at the beginning of a mass pandemic.

Haensch et al., 2010 Haensch S.

Bianucci R.

Signoli M.

Rajerison M.

Schultz M.

Kacki S.

Vermunt M.

Weston D.A.

Hurst D.

Achtman M.

et al. Distinct clones of Yersinia pestis caused the black death. Schmid et al., 2015 Schmid B.V.

Büntgen U.

Easterday W.R.

Ginzler C.

Walløe L.

Bramanti B.

Stenseth N.C. Climate-driven introduction of the Black Death and successive plague reintroductions into Europe. Cui et al., 2013 Cui Y.

Yu C.

Yan Y.

Li D.

Li Y.

Jombart T.

Weinert L.A.

Wang Z.

Guo Z.

Xu L.

et al. Historical variations in mutation rate in an epidemic pathogen, Yersinia pestis. Bos et al., 2011 Bos K.I.

Schuenemann V.J.

Golding G.B.

Burbano H.A.

Waglechner N.

Coombes B.K.

McPhee J.B.

DeWitte S.N.

Meyer M.

Schmedes S.

et al. A draft genome of Yersinia pestis from victims of the Black Death. As the geographical origins of the “p1” and “p2” SNPs are unknown ( Figure 2 B), the possibility of Branch 1 lineages arising from pre-existing diversity in Asia and independently dispersing into Europe must be considered (). This model is supported by climatic evidence, where regular westward pulses of plague from an Asian focus throughout the second pandemic are thought possible (). We find this model for the second pandemic difficult to reconcile with our current data. Although it has been previously shown that Y. pestis has an extremely variable substitution rate (), our Russian strain has only two additional derived substitutions (p6, p7, Figure 2 ) compared to London Y. pestis genome 6330 (), dated to 1350–1400. This close genetic similarity suggests that our Russian strain represents a new outbreak subsequent to that which occurred in London after the Black Death. The alternative “Asian origin” model would require a minimum of four separate lineages exiting together from the same focus to account for the level of diversity observed in Europe during the Black Death and its aftermath, i.e., (1) London/Barcelona, (2) London 6330, (3) Bolgar City, and (4) Sub-Saharan Africa. We regard the likelihood of such similar strains leaving Asia in a short time frame to be low, but acknowledge it would be possible if (1) their ancestral focus was in a location particularly conducive to westward travel, or (2) there exists a biological reason for their greater ease in rapid long-distance travel. While the above scenarios could equally explain the sole involvement of Branch 1 in contemporary plague outbreaks outside of China, we regard a single exit followed by an eastward travel as a more parsimonious explanation for the current data. Under this scenario, historical strains carrying the previously described “p3” SNP ( Figure 2 B) subsequently traveled east to later become established in China, whereas those giving rise to the Ellwangen-Observance lineage did not ( Figure 3 ). Once in the Former Soviet Union, plague likely became established in rodent populations in an area accessible to western Russia and evolved locally, as evidenced by the single unique derived position in our strain from Bolgar City ( Figure 2 B and Figure 3 ). Given that all modern Branch 1 lineages descend from a close hypothetical relative of our Russian strain, these European forms may well have given rise to the third plague pandemic in China and beyond.