BEACON Researchers at Work: Spatial patterning in microbial communities

This week’s blog post is by Fred Hutchinson Cancer Research Center postdoc Babak Momeni.

Synthetic communities may help us understand the biology of natural microbial communities.

Microbial communities in nature are abundant, with amazing diversity and huge impact on life around us. Our understanding of microbial communities, despite their importance, is limited. We know little about their structure: what their constituents are, and why they are composed the way they are. As for the functions of the communities, their activities and influence on their surroundings, some of the outcomes, for example nitrogen fixation in soil communities, are known, but we know little about the underlying mechanisms leading to such outcomes.

As part of human’s drive to “tame” the nature, almost an obsession, we would like to understand microbial communities better. How do we control microbial communities? Can we use them as miniature plants to do our dirty job for us in waste management? How do we stop them from hurting us when they are not welcome, for example in our lungs and gums? How can we make sure we stay on good terms with the massive microbial communities we are carrying around? Of course, in studying microbial communities, like many other topics in biology, the complexity is daunting. Considering the network of so many populations interconnected through multitude of diverse interactions, how do we even start to decipher the order underneath the complexity?

We thought (and “we” means my lab-mates and I from the Shou Lab at the Fred Hutchinson Cancer Research Center) it would be a good starting point to study systems we know better. That means working with mathematical models, abstract yet fully controllable, or model systems, not fully understood, but biologically relevant and at least studied extensively.  Being from an engineering background, I followed the habit of making stuff from components: what if we made communities from scratch? The idea is simple: to reduce complexity, we tried to take some of the complexity out of the picture. We built communities with only a few populations and known interactions, with the hope that understanding these simplified communities will be an entry point into the world of more complex communities.

The particular problem I am studying is about spatial patterning within microbial communities. Quite often some order has been observed in spatial distribution of populations in communities of microbes. The order may come from gradients in environmental factors, for example oxygen concentration, from specific division of labor, for example surface adhesion of some of the community members, or from interactions among populations. We focused on the latter and looked into how different community interactions may affect the spatial patterning of populations.

The interactions we chose to study first were competition for shared resources and cooperation – a mutually beneficial interaction between the two partners. Competition is ubiquitous in all microbial communities, and cooperation is thought to be one of the main drivers of complexity in life. In an attempt to avoid unnecessary complexities, we chose to look at these interactions at the ecological level, regardless of what their molecular mechanisms were.

We first used individual-based simulations, giving “cyber-cells” a set of rules to live by, and followed the development of communities under either competition or cooperation interactions. The advantage of these simulations was that it gave us an initial intuition about how interactions affected community patterning. In addition these simulations offers a precise picture of developments within the community, a complete life history including all the details of the environment for all individuals.  However, the precision and controllability of these simulations comes at a cost: the abstract model we have full control on may be out of touch with reality.

To ensure that conclusions from the mathematical model were realistic, we constructed communities of engineered yeast populations.  In competitive communities, two different strains tagged with different fluorescent proteins were competing for all the resources in the environment. To implement cooperation, we used strains of yeast each requiring a metabolite that the partner overproduced. Neither of these cooperative populations could grow on its own without supplementing their required metabolite, but when together, they complemented each other and formed a community that could grow. These engineered communities were a step toward more realistic communities, yet they enabled us to engineer desired interactions in the communities to see the effect of specific interactions.

One point we learned was that interaction-driven patterning may quickly, within a few generations, influence communities. One immediate implication of this observation is in studying evolution of microbes. We have learned a lot by tracking the fate of populations in experimental evolution in well-mixed tubes. Is evolution trajectory different in a spatially structured environment, with microbes actively influencing their environment through spatial arrangement of populations? If it is, how?

The next step would be to investigate how our observations in simplified systems hold when we move to more complex communities. How does adding more populations and/or other interaction types affect spatial patterns of communities? Can we extend our observations to natural communities? In addition, we would like to understand how the spatial  patterning driven by interactions affect the function and evolution of communities. The goal is to uncover basics of how communities develop and function from simplified communities and eventually integrate the gathered knowledge to understand more complex communities.

For more information about Babak’s work, you can contact him at bmomeni at fhcrc dot org.

This entry was posted in Uncategorized. Bookmark the permalink.

Comments are closed.