In nature, microbes exist in complex ecosystems. Bacteria “talk” by emitting extracellular signals and share resources in their environment. In synthetic biology, microbial consortia are increasingly being used to produce fuels and medicines.
Unfortunately, our mechanistic understanding of microbial consortia is still sorely limiting, especially in complex environments like the human gut. To address some outstanding questions on microbial consortia dynamics, Dr. Marika Ziesack led a project in which four bacterial strains, each taken from the mammalian gut, were engineered to overproduce one amino acid and rely on the other strains for three amino acids, thus creating a synthetic, “cross-feeding” community. Ziesack and colleagues found that this approach increased the evenness of the microbial population, but at the cost of fitness.
I sat down with Dr. Ziesack to learn more about this paper, discuss experimental challenges, and highlight future plans for this work.
This interview with Dr. Marika Ziesack on “Engineered Interspecies Amino Acid Cross-Feeding Increases Population Evenness in a Synthetic Bacterial Consortium”, published in mSystems, has been edited for clarity.
Niko McCarty: What can you tell me about yourself? Who are you, what’s your research background, and what drew you to this project initially?
Dr. Marika Ziesack: I’m Marika Ziesack, a postdoctoral fellow in Pamela Silver’s lab at Harvard, in the Department of Systems Biology. When I began this work, I was still a graduate student, also in the Silver lab. I’ve been doing synthetic biology for a long time, which is the reason why I came to the US initially. I’m originally from Germany, but I loved the science here in Boston. I was drawn to this project because I am very interested in microbial consortia, and think that this is an important area where synthetic biology should be headed. Microbes live in complex communities in nature; they’re together with other species, and it is in these natural ecosystems that they can fulfill some really interesting tasks.
Niko: Where did the idea for this project come from?
Marika: This idea was devised by two other postdocs in the lab, and it began as a DARPA (Defense Advanced Research Projects Agency) project. They had this idea to make a therapeutic, synthetic consortium that could sense and kill pathogens. Our lab proposed this idea to DARPA, with a fundamental component of our proposal being to engineer interspecies communication between a four-strain consortium. I joined the project, in part, because I’m interested in metabolism, and saw this as a great opportunity to do some metabolic engineering, but in the context of microbial communities. It was a perfect project for me.
Niko: What were your expectations when engineering this four-strain microbial consortium? Did anything happen that was unexpected?
Marika: We definitely set out with a certain vision and ran into challenges along the way. The idea, initially, was to have each species in the consortium share amino acid metabolites and, in that way, make them co-dependent upon one another. Then, we wanted to add this engineered consortium to an existing, mature community, such as that in the gut, and see if they could still establish themselves in a new environment. What we ended up finding instead was that each strain had very different nutrient requirements and overproduction values of the four amino acids. As a result, the engineered consortium was not balanced enough to actually outcompete a wild-type community; our engineered version grew about an order of magnitude slower.
Niko: In the paper, you mentioned that this engineered consortia is trading fitness for population evenness, right?
Marika: Yes, in this case, it does. If we had more time, I think the consortium could be made fitter and grow more quickly. The reason that it grows slowly right now is because the overproduction and consumption of each amino acid is not balanced. Some of the species overproduce quite well, while other species don’t produce very well; we call these latter strains “moochers”. They still contribute, but not as much as they should.
Niko: Why did you pick these four strains (E. coli, S. typhimurium, B. fragilis and B. theta) specifically? What is the benefit of using co-auxotrophies as opposed to something like quorum sensing to build this consortium?
Marika: The four species that we selected are all able to live in the gut, and specifically in the colon. We wanted to make sure that, in an actual environment, they would encounter each other and share metabolites. That was one aspect of using these strains. But we also needed species that were genetically tractable; in the gut, there are not many species that are easy to engineer. Now, some of the specific organisms that we used, such as E. coli and S. typhimurium, were selected because they can reside specifically in those ecological niches that pathogens also occupy. The Bacteroides species, on the other hand, are known to be very abundant in in the gut. Therefore, if we implement our original idea for the therapeutic consortium, the Bacteroides could be the species that actually kills the pathogens.
Now, why amino acids? That has two reasons, one practical and one biological. First of all, there is precedence in nature that amino acids can be cross-fed between members of microbial communities. Metagenomics studies have found that a lot of the gut species, especially in the colon, are auxotrophs which indicates that they cross-feed in this manner already. The other reason, which is more on the practical side, is that amino acid overproduction has been studied for a long time for industrial purposes. There was a lot of knowledge in the literature, which we used to engineer these species more quickly.
Niko: What are your future plans for this microbial consortium? Do you still aim to convert them into a therapy for killing pathogens in the colon?
Marika: Maybe in the future it could be adapted into a therapy, but it is not ready in its current form. I think it is more likely that this consortium will be used as a model system for studying microbial interactions and behaviors.
Niko: What was the biggest challenge that you encountered?
Marika: During this project, we spent a lot of time to optimize our in vitro growth conditions. Each of the four strains are different in their growth rates and requirements, and developing the right kind of media and growth protocol to have them grow equivalently was quite a challenge. I think that our development and troubleshooting of the growth media could make this synthetic consortium a good model system; other researchers can apply what we learned. Overcoming this obstacle also made me appreciate the complexity of the gut, because it was much easier to establish this consortium in the mouse gut than to grow them in vitro.
Niko: What was in the media that enabled growth?
Marika: A lot of things, because each strain has its own requirements, some of which counteract the others. The media is M9-based, but the Bacteroides have a couple of requirements; they need heme and cysteine, but cysteine actually interferes with the production of certain amino acids in E. coli, so we had to add those amino acids to the media as well. There were also a couple of vitamins that we needed to add, along with buffers to optimize pH.
Niko: Now that this project has wrapped up, what do you plan to work on next?
Marika: Well, I am still very excited about microbial consortia, especially in the context of environmental applications. I’m now working on a project related to the Bionic Leaf, which used hydrogen-oxidizing microbes, combined with a water-splitting catalyst, to produce fuels. Now, we are working to diversify the products that these microbes can produce. To do that, I’m building a co-culture between organisms that can be engineered to produce chemicals, like E. coli and yeast, and the hydrogen-metabolizing bacterium, C. necator, which we have engineered to produce the sugars that feed the heterotrophs. This is a simpler co-culture – it has just one positive interaction — but it will enable us to engineer consortia that produce lots of different chemicals from sunlight.
Niko: What can you tell me about the other authors on this paper? Did anybody offer unique perspectives that made the project possible?
Marika: Yes, definitely. A few things come to mind right away. On this paper, we had computational collaborators in on the paper, Georg Gerber and Travis Gibson, that developed a network interference algorithm, called MDSINE. We applied this algorithm to our four species consortium, which enabled us to actually identify the kind of interactions within the consortium without performing very tedious experiments. This was an incredibly useful tool for us.
Another collaborator on this project, John Oliver, also developed a quantitative PCR (qPCR) technique, with species-specific primers, that enabled us to quantify species abundance. Without this tool, it would have taken us a crazy amount of time to select for individual strains.
Niko: Do you have any advice for synthetic biologists interested in working with microbial communities?
Marika: Well, I hope that the benefits of microbial communities came across during this interview! But yeah, I think that microbial consortia is really and important area of research where synthetic biology, and microbiology, should be headed. This is how bacteria live naturally, so we need to understand how they live in complex ecosystems if we actually want to understand bacteria. Also, natural consortia can fulfill amazing, complex tasks that individual species could never accomplish. As we learn more about microbial communities, and how to control them, they will become powerful tools for synthetic biology and microbiology.
Dr. Marika Ziesack is a postdoctoral fellow in Pamela Silver’s laboratory at Harvard University, in the Department of Systems Biology. She previously obtained a Ph.D. in Biological and Biomedical Science from the same laboratory. Ziesack’s research has previously appeared in Science, Nature Communications, and many other journals.