Category Archives: archaeology

Molecular Confirmation of Yersinia pestis in 6th century Bavaria

Erasing any lingering doubts about the agent of the Plague of Justinian, a group of German biological anthropologists have shown conclusively that Yersinia pestis caused an epidemic in a 6th century Bavarian cemetery at Aschheim. Harbeck et al (2013) provide a convincing refutation of previous theories about the etiologic agent of the Plague of Justinian.   Returning to the same cemetery where plague was previously reported, two independent labs using the most modern standards to prevent contamination confirmed Yersinia pestis from multiple burials within the cemetery making this the best characterized Early Medieval plague cemetery.

The cemetery, called Aschheim, is in Bavaria outside of Munich. It contains the remains of 438 people with an unusually high number of multiple graves but no disordered mass graves. The 19 multiple burials contained two to five individuals arranged in lines. The cemetery was dated archaeologically to 500-700 AD with remains being carbon dated ranging from 530 to 680, all consistent with the 541 pandemic and its aftermath. Harbeck et al (2013) tested 19 individuals from 12 multiple graves. From these, there were eight positive samples, but only one produced enough aDNA to do some SNP genotyping. Added to the previous paper, this makes 11 positive individuals from this cemetery. Given the tenuous survival of aDNA, 11 positive individuals out of 21 tested in the two combined papers is a very good success rate. This is a cemetery that the F1 antigen test would be interesting since it could be used on the entire cemetery without great cost or labor. More sensitive than aDNA, the antigen test could tell us the percentage of plague deaths in the cemetery.

Individual A120 was screened with several SNPs that mapped it to an early region of the phylogenetic tree in the 0.ANT section. This makes the Plague of Justinian isolate ancestral to the Black Death isolates (yellow boxes below) from East Smithfield. This section whose only point of diversity is 0.ANT1 at node 4. Date predictions for the nodes of diversity in the tree fits with the Plague of Justinian falling in this region.  Modern isolates that  form this region of the phylogenetic tree all come from central Asia (around Tibet), suggesting that like the Black Death, the Plague of Justinian also originated in Asia. Overall, everything fits in well with expectations for the first pandemic.

(Harbeck et al, 2013. Fig. 1)
(Harbeck et al, 2013. Fig. 1)


Harbeck M, Seifert L, Hänsch S, Wagner DM, Birdsell D, et al. (2013) Yersinia pestis DNA from Skeletal Remains from the 6th Century AD Reveals Insights into Justinianic Plague. PLoS Pathog 9(5): e1003349. doi:10.1371/journal.ppat.1003349

Wiechmann I, & Grupe G (2005). Detection of Yersinia pestis DNA in two early medieval skeletal finds from Aschheim (Upper Bavaria, 6th century A.D.). American journal of physical anthropology, 126 (1), 48-55 PMID: 15386257

Metagenomics, Lyme Disease, and the Tyrolean Iceman’s Tattoos

When the genetic analysis of the 5,300 year old Tyrolean Iceman, better known as Ötzi, was published in February, most of the attention was naturally focused on his genomic DNA. His genomic DNA produced some interesting results: he had brown eyes, blood type O+, was probably lactose intolerant and from a southern European gene pool. He also had a collection of alleles that associate with atherosclerosis that correlate with calcifications found by CT scan in Ötzi’s arteries.

To round out a complete analysis of the single 100 mg specimen they took from Ötzi’s ileum, the largest bone of the pelvis, they did a metagenomic analysis to identify all of the non-human DNA sequences amplified. Pelvis is not really an ideal bone to take a specimen from given its proximity to the intestinal organisms that play a role in decomposing the body. Surprisingly, bacterial DNA was a very small 0.84% of the identified sequences. They oddly make no reference to the 18% of DNA reads identified as “other eukaryote”.  Of the bacterial species, 72% of the sequences were from the genus Clostridia, who are primarily spore-forming anaerobes found in the soil. The one pathogen of significance discovered was Borrelia burgdorferi, the agent of Lyme disease.

Iceman metagenome (Keller et al, 2012)
Dark field image of Borrelia burgdorferi. Photo Credit: CDC

The break down of the Iceman’s microbial phylum yielded an impressive array of bacterial diversity.  The Firmicutes include the anaerobic Clorstridium species that are found in the soil. The Proteobacteria include the enteric bacteria like Escherichia coli, many of which are facultative anaerobes. Both of these phylum would be included in decomposition of the body and as anaerobes could grow in the corpse. Borrelia burgdorferi, the agent of Lyme disease, belongs to the phylum Spirochaetes. They were able to sequence approximately 60% of the Borrelia burgdorferi genome. To find B. burgdorferi in the pelvis suggests that the infection was in a systemic phase.

There are two pieces of correlating data to support a Borrelia burgdorferi infection. The international team that did this work linked the infection with Ötzi’s atherosclerosis, an association previously shown between Lyme disease and several other systemic infections.

Tattoos on the Iceman cover or align with major joints and muscles. (South Tyrol Museum of Archaeology site)

Yet, a common symptom of systemic Lyme disease is joint and muscle pain. One of the earliest observations of Ötzi’s mummy is that he has a lot of tattoos specifically placed over joints and muscle groups in places where strain would be expected. These tattoos do not appear to be decorative or signs of inclusion in a community. Consensus appears to have formed early on that these tattoos were medicinal, probably for pain relief. Scans of the mummy do suggest some arthritis. With his lifestyle, an approximately 45-year-old man is expected to have some arthritis and pain.  Both atherosclerosis, and evidence of joint pain and some arthritis can be explained by other means, but when taken together with the B. burgdorferi DNA make a compelling case that Lyme disease contributed to his overall state of health.


Keller, A., Graefen, A., Ball, M., Matzas, M., Boisguerin, V., Maixner, F., Leidinger, P., Backes, C., Khairat, R., Forster, M., Stade, B., Franke, A., Mayer, J., Spangler, J., McLaughlin, S., Shah, M., Lee, C., Harkins, T., Sartori, A., Moreno-Estrada, A., Henn, B., Sikora, M., Semino, O., Chiaroni, J., Rootsi, S., Myres, N., Cabrera, V., Underhill, P., Bustamante, C., Vigl, E., Samadelli, M., Cipollini, G., Haas, J., Katus, H., O’Connor, B., Carlson, M., Meder, B., Blin, N., Meese, E., Pusch, C., & Zink, A. (2012). New insights into the Tyrolean Iceman’s origin and phenotype as inferred by whole-genome sequencing Nature Communications, 3 DOI: 10.1038/ncomms1701

South Tyrol Museum of Archaeology permanently houses and studies the mummy.

Retrospective Diagnosis in the 21st Century

Black Death at Tourinai, 1349

The way we make and think about retrospective diagnosis is changing. Over the last decade, laboratory results have become the preferred (maybe even mandatory) method of making a retrospective diagnosis [1]. To extrapolate a few positive laboratory results to cover an entire epidemic, it must correlate with reported signs and symptoms and ideally epidemiology. There are pitfalls at every step of the process.

Today, laboratory results focus on identifying species specific biomolecules, aDNA and non-DNA molecules such as specific capsule lipids. I previously posted more detail on the non-DNA biomolecules used in paleomicrobiology. Each of these methods has its critics and proponents. Laboratory results should always be logical with the osteological analysis of the skeletal remains and with at least the majority of the written record. We have to remember that just because an infection was detected, it doesn’t necessarily mean that they died from it. If a skull has a sword wounds, it is unlikely that they died from tuberculosis. There is a significant difference between dying with a disease and dying from it.

For these science reports to be helpful to historians and anthropologists, they need to provide better context on the skeletal remains including how the remains were dated and associated with a particular epidemic. To give one example, a recent letter to Emerging Infectious Diseases reports Yersinia pestis in unerupted teeth from tombs under Sant’ Andrea church in Barletta, Italy [4]. They suggest this identifies the 1656-1657 epidemic as the plague. Although the tombs were said to have been used “since the 14th century” and have been hypothesized to have been used for the 1656-1657 epidemic, they do not give any indication how these five youths were selected for study. They report that their negative control from the same tomb complex dated to before the epidemic, so presumably this could be easily corrected by giving the rationale an archaeologist used to select these youths.

Assessing signs and symptoms and the epidemiology of past disease requires correctly using ancient and medieval documents. Medical treatises are not necessarily as helpful as they may seem. Medieval authors often copied classical texts that were centuries old and their new observations were heavily influenced by contemporary philosophical paradigms [2]. Instead, a variety of documents that can be broadly classified under government records (tax and manor rolls, wills, court records, etc); biographies and hagiographies; chronicles, annals and medieval histories can all be very useful in part because the disease is often not the main focus of the account [2].

Using such an array of ancient and medieval works has plenty of pitfalls even for historians, much less those from other disciplines. Mitchell has outlined a number of pitfalls to retrospective diagnosis from medieval texts that I will try to summarize with a few notes of my own [2].

  1. Not enough information in the source for an accurate diagnosis.
  2. An apparent eye-witness account is actually copied from an older source.
  3. Translation is not representative enough of the original language for diagnosis.
  4. Inadequate knowledge (of the researcher) of disease signs and symptoms and how they are represented in this cultural context.
  5. Cherry picking symptoms to fit a theory.
  6. Failure to understand that a source is not representative of the body of contemporary sources.
  7. Not realizing that more than one disease is present. Likewise, assuming that two or more diseases are present when they are not.
  8. Assumption that the antique or medieval disease is one that still exists today.
  9. Assumption that the characteristics of the disease have not changed from then until now, including transmission methods. Likewise, assuming that characteristics of the microbe must have significantly changed over time.
  10. Failure to understand how the environment and living conditions change the epidemiological characteristics of the epidemic.

Epidemiology is the most difficult parameter to assess. For ancient and medieval epidemics, I find it very difficult to consider epidemiology part of the retrospective diagnostic process. There are simply too many variables in the environment and human cultures of the past. Even diseases like malaria and influenza that we think we know so well still throw us surprises today [3]. Not even seasonality is set in stone as the 2009 H1N1 influenza pandemic recently showed us. After now centuries of study, we still can be hard pressed to explain the rise and fall of malaria. By it’s very definition, a pandemic is an extraordinary situation compounded by historical distance for study of historic outbreaks.

Even removing it from the diagnostic process, epidemiology is still one of the most important fields of study for past epidemics. If we can answer the important epidemiological questions on past epidemics then I think we can say that we understand at least the science of the epidemic. For many well-known pathogens like influenza and plague, we have had many outbreaks but very few pandemics to learn from to prepare for future threats.

[1] Little, L. (2011). Plague Historians in Lab Coats Past & Present, 213 (1), 267-290 DOI: 10.1093/pastj/gtr014

[2] Mitchell, P. (2011). Retrospective diagnosis and the use of historical texts for investigating disease in the past International Journal of Paleopathology, 1 (2), 81-88 DOI: 10.1016/j.ijpp.2011.04.002

[3] Raoult D (2011). Molecular, epidemiological, and clinical complexities of predicting patterns of infectious diseases. Frontiers in microbiology, 2 PMID: 21687417

[4] Scasciamacchia S, Serrecchia L, Giangrossi L, Garofolo G, Balestrucci A, Sammartino G, et al. Plague epidemic in the Kingdom of Naples, 1656–1658. Emerging Infectious Disease. 2012 Jan .