It’s easy to forget how impressive the human skin really is. It’s an impenetrable barrier covering most of our body and without it, we’d be susceptible to a whole array of infections. Equally as impressive as the skin are the microbial inhabitants. They survive on a hostile environment that can swing between dry and oily with very scarce nutrients, not to mention the risk that they will be abrasively brushed away at any moment. Understanding the interaction between microorganisms and the skin, has been a pursuit of microbiologists for centuries.
When we study the skin microbiota, we either take snapshots from humans, or try to delve into the details using experimental models such as mice. Looking at humans gives us the “real-life” picture of what’s going on, but we are limited in the types of experiments we can do. Experimental models are much more flexible but when we do experiments in animal models, there is always the niggling question, is it the same in humans?
A new study from a research team led by Keira Melican at the Karolinska Institute in Stockholm describes a new hybrid model that hopes to answer this very question. In a process that took years of refinement, Melican, along with postdoc Anette Schulz optimized a humanised skin model, where human skin was grafted onto mice, to study how human skin responds to bacterial colonisation with the flexibility of an animal model.
“We’re looking at colonisation. Most studies look at subcutaneous infection of Staphylococcus aureus which requires the introduction of large numbers of bacteria under the skin. We have refined our humanised model to the point where we can look at subtle changes in relatively low concentrations of bacteria on the surface of the skin,” explained Melican.
This is important. In order to understand how commensal organisms like S. aureus can cause disease, we have to understand how they colonise and behave under normal conditions. “The idea is to understand pre-infection,” described Melican. “If we can understand pre-infection, then we can understand how to prevent infection.”
This model was adapted from a model Melican had previously used during her post-doc to study blood vessel infection with the human specific bacteria Neisseria meningitidis. Setting up the model in her own lab to study skin surface, however took a lot of complex optimisation. For a start, human skin is difficult to come by. It turns out that the best source of healthy human skin is… plastic surgery. “This puts you at the mercy of the plastic surgeon and their surgery timetable, you have to plan all of your experiments around that,” explained Melican. If the skin came in the middle of the night, this is when you did the experiments. If you desperately needed one replicate to finish up, well you just had to wait.
On top of sourcing healthy skin, there were experimental issues to contend with. Before the technique was perfected the team found that wounds were easily introduced into the graft. “This produced blood and gave the bacterial cells access to iron resulting in higher abundances on the skin,” explained Anette Schulz. However once this problem was overcome, the model began to feel like it was working consistently. “You could see that there was no inflammation around the skin graft which meant the model was stable and we were much less likely to get interference in our results,” said Schulz.
Despite progress in the lab, external signs were not always so positive. With all the will and persistence in the world, funders seemed reluctant to back the project. “In the end, we relied on small amounts of funding from a diverse group of small foundations to keep the experiments going,” explained Melican. This seems to be a situation that an increasing number of young PIs find themselves in. Despite the difficulties, the model started to yield more and more results which was validation of the hard work and painstaking optimisation. As Schulz described, “despite variability in the skin and the labour intensive optimisation, it was great that we started to see that the results were very reproducible.”
So what happens when you model S. aureus colonisation on the humanised skin model? “Human skin responds completely differently to mouse skin,” explained Melican. “A key difference was the role of corneocytes which we typically don’t see in any other model, from in vitro to mouse.” The human corneocyte layer is substantially thicker than that found in mice and it is a strong barrier, preventing bacterial cells from penetrating. In mouse skin, the bacterial cells are often able to penetrate the skin and can easily be found in subcutaneous layers.
The signalling pattern is also completely different. When analysing the human skin, they found a different pattern of cytokines being produced compared to what had been seen in mouse skin before. These signals resulted in the recruitment of neutrophils leading to the bacteria being killed, resulting in a transient colonisation in the humanised skin model. These results match perfectly what has been seen on human skin across hundreds of studies in the literature. S. aureus, tends to be a transient visitor and this is the first model that has been able to highlight an immunological answer as to why.
Critics can argue that the experiments are still carried out in mice. However, these results show human immune responses to colonisation that produces data that has been observed on human skin by many other studies. Although the work was done in mice, it was done on human skin and it’s much easier to see the relevance.
This breakthrough could potentially set a new benchmark for the study of microbe skin interactions. The model is established now, it just needs to be used. Thanks to the patience and persistence of this study, we can now appreciate the intricate biology of the human skin. When I did my undergrad, we were often told that the skin is a desert, essentially inert and microbes go there to dry out and either die, or wait until they end up somewhere nicer. Thankfully now we know that is not the case and it seems that we are just scratching the surface of these fascinating microbe-host interactions.
“Neutrophil recruitment to noninvasive MRSA at the stratum corneum of human skin mediates transient colonization”. Anette Schulz, Long Jiang, Lisanne de Vor, Marcus Ehrström, Fredrik Wermeling, Liv Eidsmo and Keira Melican. Cell Reports, online 29 October 2019, doi: 10.1016/j.celrep.2019.09.055.