Severe infections emerge from colonising bacteria by adaptive evolution

In a study published today, Modernising Medical Microbiology researchers present an investigation into the relationship between bacteria found from people with severe infections cause by the bacteria Staphylococcus aureus.

Staphylococcus aureus, like many other the bacteria which cause human disease, are commensals, meaning they are carried as part of the human microbiome. They are found harmlessly carried (or ‘colonising’) much more often than they are found in disease. An unanswered and important question is whether changes in the bacterial genome – the code for how bacterial cells grow and behave – affect whether carried bacteria will cause disease.

To help answer this question, we investigated the genomes of Staphylococcus aureus bacteria found in 105 patients with both infection and nose colonisation. We took multiple bacteria from the nose and infection sites (including infection of the bloodstream, bone, joints and soft tissues), and compared the genetic sequence of bacteria found within a single person. By doing so we have been able to describe, in a level of detail not previously achieved, the relationship between bacteria in colonising and infection populations.

Our study found that these bacteria are closely related. 90% of people with Staphylococcus aureus nose colonisation and infection have related bacteria at both sites. We are able to demonstrate that in the majority of cases (72% of all infections) the infection population emerged from the colonising population. However while the nose and infection populations are closely related, in most (78%) at least one genetic difference separates them.

The genetic differences (or “mutations”) that separate colonising and infecting bacteria are a promising group to study for insights into how changes in the bacterial genetic code might increase the chance of disease. We investigated for genes or sets of genes that were had mutations occurring in multiple individuals which we predicted would change the protein produced by that gene. We found that AgrA, a “regulator” which determines which products bacteria express, was affected by mutations much more often that would be expected just by chance. We also found that genes under control of AgrA and another regulator (Rsp) were more likely to be affected by mutations. These mutations were especially frequently found in genes that code for proteins which form part of the bacteria’s outermost surface, and play a role in how the bacteria binds to human cells and avoids the human immune system. This pattern of clustered mutations is evidence for adaptation: changes in the bacterial code that either makes bacteria more able to cause an infection, or occurs in response to infection.

We compared these changes to the mutations observed in Staphylococcus aureus carriage without disease, and did not find the same patterns in carriage. Nor were the results explained by investigating which genes were most likely to have mutations over many years. We conclude that these findings of adaptation are specific to infection. These results open up new possibilities for how we might identify infection, as well as future targets for new ways to treat infection. They also highlight the fascinating and complicated balance between humans the bacteria that form our microbiome.

Original paper: Bernadette C Young, Chieh-Hsi Wu, N Claire Gordon, Kevin Cole, James R Price, Elian Liu, Anna E Sheppard, Sanuki Perera, Jane Charlesworth, Tanya Golubchik, Zamin Iqbal, Rory Bowden, Ruth C Massey, John Paul, Derrick W Crook, Timothy E Peto, A Sarah Walker, Martin J Llewelyn, David H Wyllie, Daniel J Wilson. Severe infections emerge from commensal bacteria by adaptive evolution. eLife 2017;6:e30637
https://elifesciences.org/articles/30637