A sample of ancient fossilised poo from the 14th century contains genes for antibiotic resistance – a surprising find as the drugs were not developed until hundreds of years later.
A team of researchers from Aix Marseille Université analysed a 600-year-old human faeces from Belgium and found it contained viruses with genes for antibiotic resistance.
The sample was found following an urban renewal project in the city of Namur, where latrines dating back to the 14th century were discovered underneath a square.
A range of phages – viruses that infect bacteria rather than organisms like plants and animals – were found in the ancient stool sample.
Most viral sequences found in the sample were related to viruses currently known to infect bacteria found in poo, including harmless bacteria that lives in the human gut.
Published in the journal Applied and Environmental Microbiology, the communities of phage found in the faecal sample were different from those found in modern human faeces, but the function they carry out is very similar.
This suggests the viral community is intrinsic to the human gastrointestinal tract and has remained largely unchanged for centuries, corresponding author Christelle Desnues said, adding there is considerable evidence to show bacteria is very important to human health.
Desnues said genes found in the poo sample include antibiotic resistance genes and genes resistant to toxic compounds. Both toxins and antibiotics are common to nature, so may have emerged to protect the gut from them.
"Our evidence demonstrates that bacteriophages represent an ancient reservoir of resistance genes and that this dates at least as far back as the Middle Ages," Desnues said.
"We were interested in viruses because these are 100 times more abundant than human cells in our bodies, but their diversity is still largely unexplored. In the present study, we thus focused on the viral fraction of the coprolite by using, for the first time, a combination of electron microscopy, high-throughput sequencing and suicide PCR (quantitative real-time) approaches."