Abstract
Yersinia pestis, the causative agent of plague, crippled population across Eurasia in three well-described historical pandemics: The Justinianic plague (541-780), the Second Pandemic (1346-1840) and the Third Pandemic (1894-). The Second Pandemic was the deadliest, with the Black Death (1346-1351) killing up to 60% of Europe's population, far exceeding the population mortality of the Third pandemic, despite few genomic differences between their plague strains. To identify unique genetic variation that could help explain the high death toll of the Second Plague pandemic, we used variation-graph based pan-genomics on a large dataset of 110 new and 12 published shotgun sequenced ancient plague genomes to comprehensively characterise the pangenome diversity of the Second Pandemic plague. Our pangenomic approach revealed the presence of a cryptic 34kb plasmid, pTP33, in almost all strains from the Second Pandemic, but undetectable in all other ancient strains. We find that pTP33 is a phage-plasmid-like derived element with homologues in Enterobacterales bacteria with established insect- or invertebrate-associated lifestyle phases. Our results suggest that pTP33 widened the range and infectivity of vector species available for plague transmission, consistent with reported effects on biofilm formation and bacterial retention. We hypothesise that pTP33 broadened the vector range of Y. pestis beyond the classical rat/flea cycle, facilitating the establishment of new rodent-ectoparasite plague reservoirs, which, critically, increased the risk of flea- or lice-mediated plague transmission between humans. Taken together, our findings offer a biological explanation for differences in transmission dynamics and population mortality between the Second and Third plague pandemics.