In keeping with this blog’s goal to be a meeting ground for interdisciplinary discussions, I’ll be hosting a series of dialogues between scholars in the humanities and sciences. If you would like to be involved in one of these dialogues, please use the contact form on the about page.
On behalf of today’s participants, I invite you to post comments on this dialogue or on twitter.
Monica H. Green (email@example.com, @MonicaMedHist ) is a historian of medieval medicine. An elected Fellow of the Medieval Academy of America, she teaches both global history and the global history of health. She was the editor in 2014 of Pandemic Disease in the Medieval World: Rethinking the Black Death, the inaugural issue of a new journal, The Medieval Globe.
Boris Schmid (@BorisVSchmid) is a theoretical biologist at the University of Oslo, Norway, and specializes in disease ecology and epidemiology. He recently described a link between climate fluctuations in medieval Central Asia and what looks like repeated introductions of plague into Europe’s harbors, a hypothesis that can be tested by the analysis of ancient DNA samples of Y. pestis. He works in a multidisciplinary team of theoreticians, archeologists, microbiologists and historians, led by Nils Chr. Stenseth.
“Tiny Changes with Huge Implications: Counting SNPs in Plague’s History”
I’d like to raise the question of SNPs and why understanding them is so important for those keeping an eye on the “plague aDNA debates.” SNPs – single nucleotide polymorphisms, one of the smallest levels of genetic change you can document on a genome — are critical to the evolutionary, and historical, story of Yersinia pestis (the causative agent of plague), because Y. pestis in fact has been proven to change so little over time. Nucleotide changes are a normal part of evolution. They just happen. But they don’t necessarily happen at the same rate in all organisms, or even necessarily at the same rate in the same species in all circumstances. On top of that, sudden proliferation of an organism — say, in the context of an epidemic in the case of pathogenic organisms — may increase the rapidity with which a random evolutionary change gets fixed in a population. (See Cui et al. 2013)
In the case of Y. pestis, it’s actually been shown to be “genetically monomorphic.” That’s another way of saying, it doesn’t change very much or very often. Within its core genome (the common set of genes shared by all Y. pestis bacteria), most any Y. pestis cell you find is going to look pretty much like every other one across much of its ±3.5 million nucleotides, differentiated only by a few dozen or a few hundred different SNPs (Achtman 2012). So when a ‘C’ (cytosine) in one position on the genome changes to a ‘G’ (guanine) or ’T’ (thymine) or ‘A’ (adenine), that can actually be a historically significant change. And when that new SNP suddenly shows up in a whole population, that’s a sign that something major has happened. That’s what Cui et al. 2013 talked about in the “Big Bang” they posited for the later medieval period: the old Branch 0 of Y. pestis (see fig. 1) suddenly split into four new branches, each characterized by a different SNP signature. And the sequencing in 2011 of the first three Y. pestis genomes from the London Black Death cemetery (Bos et al. 2011; I’ll want to talk about the 4th genome, London 6330, later), made clear that that split did indeed happen quite suddenly. The split happens (presumably somewhere in western China or central Eurasia) and then, “two SNPs later,” Y. pestis shows up in London. Given the distances involved in transmitting this single-celled organism — several thousands of kilometers — and the fact that there are no airplanes in the 14th century to account for rapid dissemination, that’s quite significant.
The mutation rates of bacteria is indeed a topic with quite some nuance, and one that is worth delving into in order to understand the slew of aDNA plague papers that are currently coming online.
The rate at which plague accumulates new point mutations has been calculated from the plague outbreaks in Madagascar (Morelli 2010) to be on average of 1 mutation for every 100 million nucleotides copied (Cui 2013). While that sounds like a very uncommon event, it really is not, given the length of its genome – a single bacterium that is replicating itself would after 4 generations (so roughly 7 hours within a human (Chauvaux 2007)) be a small colony of 16 bacteria, with a total of 105 million nucleotides of core genome copied, and therefore most likely already with 1 SNP among them. But here we are comparing two different things – the mutation rate used by Morelli and Cui is valid for the amount of variation you’d expect to accumulate in each surviving lineage of plague during an epidemic; many of the mutations that were generated are lost again simply because those bacteria were not the ones that continued the epidemic.
Bos (2011) notices with surprise that the plague at the time of the Black Death differed just 97 SNPs from CO92 (a sample collected from a human patient in 1992 in Colorado, and the plague strain that by convention is used as a baseline against which to compare other plague strains). So that is 1 accumulated mutation every 6.6 years, counting from 1348 to 1992. That compares pretty well to the mutation rate in a plague focus we in Oslo are currently studying, where we see an accumulation of mutations along each lineage of about 1 mutation every 8 years. Continuing with our back-of-the-envelope calculation, this “two SNPs later” since the sudden divergence of plague into 4 lineages puts the date of the “Big Bang” at just 10-20 years prior to the arrival of the Black Death in Europe in 1347. There are some caveats that come with this analysis, notably that the rate at which mutations are accumulated can be quite slower in some branches of plague (up to 6-fold, Cui 2013), but for now let us continue with a simplistic 1 new SNP being fixed into a plague lineage approximately every 7 years.
With that 1 SNP per 7 years rate in mind, let’s look at the stylized version of the evolution of Yersinia pestis, recently presented by Johannes Krause (fig. 2 below), with the recently published medieval plague sequences (and now the new publication from Spyrou et al. 2016), and see what it tells us in terms of the history of plague.
Thanks, Boris, for laying out these calculations. So, we’ve established that, even though Y. pestis is genetically monomorphic—that it doesn’t have sex and doesn’t even often switch genetic material laterally—that it does have a tendency to develop random mutations over time. Par for the course for basic biology. I’d like to return now to a point I already made, which is more historical: “And when that new SNP suddenly shows up in a whole population, that’s a sign that something major has happened.” The big question we’re looking at with the Black Death strain and the “founding” of Branch 1 of the Y. pestis phylogenetic tree is: how did one—just one—Y. pestis cell get its SNP profile to be the founder of the whole lineage that caused millions of deaths in the first wave of the Black Death? This seems to be a classic “founder effect” scenario, the Mitochondrial Eve of this new phase of plague history. (And yes, I know that Y. pestis doesn’t have mitochondria.) Granted, we only have fully sequenced aDNA from two Black Death sites: London and Barcelona. But Spyrou et al. report that the Barcelona sample is identical in every respect with the three London Black Death genomes that were sequenced in 2011.
Let us assume then, for the moment, that our historical sources are correct that the famous outbreak of plague in Caffa in 1346 (the bodies being hurled over the walls) caused a single chain of events that carried our one strain of Y. pestis all the way to Barcelona and London within a two-year period. Presumably, if we had samples from all the other places plague struck in those brutal first years as it spread from the Black Sea to the Mediterranean—Alexandria, Aleppo, Almeria, etc., etc.—they would all have the same SNP signature that Haensch et al. 2010 reported from sample SNP typing for Hereford (UK) and Saint-Laurent-de-la-Cabrerisse (France), and that Seifert et al. 2016 reported, also from sample SNP typing, for Manching-Pichl (Bavaria). None of those sites are as precisely dated as the first three genomes sequenced from the East Smithfield cemetery in London (which is dated from documentary records to late 1348 to 1350). But for the sake of argument let’s assume that they are, in fact, all first outbreak burials. So all those human deaths are caused by an identical organism, moving into new vulnerable hosts at the same time.
But now we have to ask: what happened after the Black Death? What was Y. pestis’s next stage of evolution? The samples we have thus far, whether whole genome sequences or just SNP profiles, are all taken from human bodies. But humans are not natural carriers of plague. Every aDNA sample found thus far is probably a dead end, literally: the end of the line for the unique SNP profile (if it had one) of the Y. pestis that infected that individual. It died with them. If we’re going to look for what sustained plague, what allowed it to focalize and continue replicating for centuries thereafter, then we need to look at what plague strains survived in rodent populations.
The question of where plague’s reservoir was in medieval Eurasia is indeed the question that, as a biologist, I was eager to work on, Monica. Validation of any theory of how plague has moved across Eurasia in the past would have to come from the analysis of the SNPs that plague gathered along the way. However, the interpretation of those medieval SNPs turns out to less clear-cut than most of us had anticipated, as we will discuss in the next blog post.
Achtman, M. (2012). Insights from genomic comparisons of genetically monomorphic bacterial pathogens. Philosophical Transactions of the Royal Society B: Biological Sciences, 367(1590), 860–867. http://doi.org/10.1098/rstb.2011.0303
Bos, K. I., Schuenemann, V. J., Golding, G. B., Burbano, H. A., Waglechner, N., Coombes, B. K., et al. (2011). A draft genome of Yersinia pestis from victims of the Black Death. Nature, 478(7370), 506–510. http://doi.org/10.1038/nature10549
Chauvaux, Sylvie, Marie-Laure Rosso, Lionel Frangeul, Céline Lacroix, Laurent Labarre, Angèle Schiavo, Michaël Marceau, et al. 2007. “Transcriptome Analysis of Yersinia Pestis in Human Plasma: An Approach for Discovering Bacterial Genes Involved in Septicaemic Plague.” Microbiology 153 (Pt 9): 3112–3124
Cui, Y., Yu, C., Yan, Y., Li, D., Li, Y., Jombart, T., et al. (2012). Historical variations in mutation rate in an epidemic pathogen, Yersinia pestis. Proceedings of the National Academy of Sciences, 110(2), 577–582. http://doi.org/10.1073/pnas.1205750110/-/DCSupplemental/sd01.xls
Haensch, S., Bianucci, R., Signoli, M., Rajerison, M., Schultz, M., Kacki, S., et al. (2010). Distinct Clones of Yersinia pestis Caused the Black Death. PLoS Pathogens, 6(10), e1001134. http://doi.org/10.1371/journal.ppat.1001134.t001
Krause, Johannes (4-12-2016) Oral Presentation #S577: Ancient pathogen genomics: what we learn from historic pandemics. European Congress of Clinical Microbiology and Infectious Diseases http://eccmidlive.org/#resources/ancient-pathogen-genomics-what-we-learn-from-historic-pandemics
Seifert, L., Wiechmann, I., Harbeck, M., Thomas, A., Grupe, G., Projahn, M., et al. (2016). Genotyping Yersinia pestis in Historical Plague: Evidence for Long-Term Persistence of Y. pestis in Europe from the 14th to the 17th Century. PLoS ONE, 11(1), e0145194–8. http://doi.org/10.1371/journal.pone.0145194
Spyrou, M. A., Tukhbatova, R. I., Feldman, M., Drath, J., Kacki, S., de Heredia, J. B., et al. (2016). Historical Y. pestis Genomes Reveal the European Black Death as the Source of Ancient and Modern Plague Pandemics. Cell Host and Microbe, 19(6), 874–881. http://doi.org/10.1016/j.chom.2016.05.012