From seals to humans and beyond: Tuberculosis transmission in ancient South America

Ancient DNA shows that what likely began as a zoonotic infection of tuberculosis from marine mammals to coastal peoples spread far into the inland regions of South America, pointing to more complex transmission pathways.

Like Comment
Read the paper

Written by Tanvi Honap and Åshild Vågene

Read our paper: Geographically dispersed zoonotic tuberculosis in pre-contact South American human populations

Tuberculosis (TB) was one of the diseases thought to have been introduced to the Americas by European colonizers beginning in the late 15th century. This view is supported by the genetic diversity of TB strains seen across the Americas today, which largely reflects the diversity seen in Europe. But this narrative does not explain the geographically widespread and abundant skeletal evidence for TB across both South and North America from the pre-European contact era. Our current ancient DNA study of TB in pre-contact South America builds upon decades of work by archaeologists, bioarchaeologists, and paleopathologists who have studied skeletal TB in these regions.

What does ancient DNA tell us about TB variants circulating in the pre-contact Americas?

The first genetic evidence of TB in pre-contact South America was found in 1994, when Wilmar Salo and colleagues used a PCR-based approach to detect TB DNA in the lung lesion of a 1,000-year old mummified individual from coastal Peru. But PCR-based studies, though useful for detecting TB DNA, provided limited resolution as to the types of TB variants circulating in the pre-contact era. In the past decade, advances in ancient DNA and next-generation sequencing technology have enabled the recovery of complete pathogen genomes from ancient human bones. To apply the power of ancient DNA to skeletal TB cases from the pre-contact Americas, an interdisciplinary collaboration was developed by bioarchaeologist Jane Buikstra and ancient DNA researchers Anne Stone and Johannes Krause.

In 2014, the team led by Kirsten Bos and colleagues recovered the first complete ancient TB genomes from three individuals who lived in the coastal areas of the Osmore River valley of Peru about 1,000 years ago. After comparing these ancient TB genomes to those of modern TB strains from all around the world, they made a surprising discovery - these ancient Peruvians were infected by a variant of TB that is today associated with marine mammals, such as seals and sea-lions. Since it infects pinnipeds, this TB variant is known as Mycobacterium pinnipedii. Archaeologists had long known that the coastal Peruvian peoples regularly hunted seals for their meat and fur. Thus, it seemed that the process of handling and consumption of infected seal meat had transmitted TB to these peoples.

Seal meme

This was indeed an exciting discovery of an ancient zoonotic TB transmission event caused by M. pinnipedii. Nevertheless, several questions remained, such as:

Which variants caused TB in inland populations that did not have direct contact with marine mammals?

This brings us to our current study! Our aim was to determine which TB variants were responsible for pre-contact era skeletal TB cases in the non-coastal parts of South America. The paleopathologists on the current team included Buikstra, as well as Felipe Cárdenas-Arroyo, Laura Paloma Leguizamón, and Judith Arnett. Together, they identified nine skeletal TB cases from multiple inland archaeological sites located in the highland regions of the Andes in Colombia. They also included a case from the inland M6-Estuquiña site from the Osmore River valley of Peru. They collected bone samples, either vertebrae or ribs, from these individuals for ancient DNA analysis. 

Our work began with retrieving ancient DNA from the bone samples in cleanrooms specially designed for working with ancient biomolecules at Arizona State University, U.S.A. and the Max Planck Institute for the Science of Human History, Germany. Working in the cleanroom involved wearing full-body Tyvek suits, gloves, face shields and masks (long before COVID made the latter a part of our everyday wardrobe) not to protect ourselves from infection but to protect the precious ancient DNA samples from contamination with our own “modern” DNA.

Ancient DNA lab-work involves long periods of sitting around, waiting for tubes to
finish spinning, while dressed from head-to-toe in protective gear.
Photo by Sarah Johnson

Next, we employed a target enrichment aka “fishing” strategy in which we used “bait” molecules matching the TB genome sequence to increase the amount of ancient TB DNA in our samples prior to sequencing. After a copious amount of sequencing and data analysis, followed by re-sequencing and re-analysis, we recovered three new ancient TB genomes! One was from the non-coastal Peruvian individual from M6-Estuquiña and the others from two individuals from the Colombian sites of Las Delicias and Candelaria La Nueva located near modern-day Bogotá. All of these new three ancient TB genomes resemble M. pinnipedii - the same TB variant found in the ancient coastal Peruvian individuals and in modern-day seals and sea lions.


Phylogenetic tree. The branches highlighted in orange are M. pinnipedii genomes recovered from ancient humans. The ones in bold orange are from this study. This figure appears as Figure 2 in our paper.

A combination of archaeological evidence and stable isotope data suggest that these inland individuals from Peru and Colombia did not exploit or consume marine vertebrates, and therefore could not have contracted TB directly from infected seals. So how exactly did M. pinnipedii make it all the way to inland highland Colombia?

We don’t know for sure, but we can make an educated guess. TB strains show broad host tropism, meaning, they are rather promiscuous and can “jump” from one mammalian species to another with relative ease. Colombia has a wide variety of terrestrial mammals, so M. pinnipedii could have been brought inland via the animal life. Or in a more likely scenario, it could have been brought inland via human-to-human transmission facilitated by trade routes. Or perhaps a combination of both! 


Potential scenarios of TB transmission in pre-contact era South America

More research is necessary in order to shed light on these transmission pathways. But before thinking about future steps, it is worth reflecting on how we got here.

How was the existence of pre-contact TB in the Americas discovered in the first place?

TB generally affects the lungs, however, in a minority of cases it can spread to the skeleton, eroding the lower spine and causing it to collapse. Since the 19th century, medical doctors and anthropologists had identified TB-like lesions in ancient human bones from the Americas. In 1973, paleopathologist Marvin Allison used an interdisciplinary approach, which combined observations of disease-altered bones with radiography and other contemporary biomedical approaches. He identified acid-fast bacteria, likely to be the TB pathogen, from the organs of a mummified Andean child from Peru dating to approximately 700 CE. This provided early biomedical evidence of pre-contact TB from coastal Peru. Recognizing the rich potential of Andean mummies and other human remains for advancing knowledge of ancient TB, Buikstra subsequently initiated field research to excavate and identify TB in a series of ancient peoples from coastal sites along the Osmore River valley of southern Peru. This led to the biomolecular studies described previously and a subsequent focus on skeletal identification of TB in non-coastal locations, such as the site of M6-Estuquiña in the Osmore River valley of Peru and the highland regions of Colombia, as well as in North America where TB cases start appearing much later around 900 CE. 

Future plans for ancient TB research

We hope that future ancient DNA research focusing on the recovery of TB genomes from animal remains from the Americas will help answer some questions about the transmission of these ancient TB strains. Perhaps we will find that animals domesticated by ancient South American peoples, such as guinea pigs or llamas, acted as reservoirs for the disease by maintaining the pathogen and passing it on to humans. But diagnosing TB in animal skeletal remains is difficult and genetically screening animal skeletal collections for ancient TB DNA requires a heavy investment of time, money, and effort.

Additionally, the recovery of M. pinnipedii genomes from pre-contact era North American human populations will provide a more complete picture of its geographical dispersal. On the other hand, we may find that an entirely different TB variant was present in this region! This is an exciting time for ancient TB research and we look forward to finding new pieces that help solve this important evolutionary puzzle.

Tanvi Honap

Research Assistant Professor, University of Oklahoma