Engineering Stable Microbes
By Riley Stockard.
Synthetic biologists are bad at making stably engineered cells. Enforcing microbial cooperation can help, if only we can study and recreate it.
As much as synthetic biologists like to compare their work to electrical engineering — we throw around terms like chassis and circuit — the current state of our field is nowhere near matching nature’s capabilities. Building a genetic circuit is not the same as programming a computer. Biology is messy and unpredictable. Engineered cells, often, are fragile, even within the carefully controlled environment of a test tube. A single mutation can ruin an entire batch.
Much like the Wright Brothers’ early airplane designs mimicked the wings of birds, synthetic biology works best when it borrows ideas from nature. And on our crowded planet, multicellularity is the norm, the standard, the dominant idea.
Bacteria communicate with neighbors by emitting and sensing small molecules (this is called quorum sensing) to coordinate group behaviors, like building biofilms. Microbes also swap genes via horizontal gene transfer, or HGT, thus sharing their blueprints for useful traits, like antibiotic resistance. Communities and shared DNA make microbes more resilient to changing environments, much like global supply chains make nations more resilient to droughts and disaster.
Why, then, do we grow genetically-engineered microbes in monoculture? Over time, subsets of these engineered cells lose their added functions, and take over the entire batch. To accelerate our transition to a circular bioeconomy, we must make more resilient, stable cells. And to make such cells, we must first emulate and borrow ideas from the group interdependence and resilience found in nature. This is what I’m working on as a New Science fellow.
I’ve seen the chasm that separates electrical and biological engineering — in an engineering course, I once built an infrared heartbeat sensor in just a few days. It was easy to detect and fix errors, and it felt like magic to watch my heartbeat travel across a screen. I left that course feeling as if I could build anything…until I set foot in a biology lab.
In my four years of undergraduate research, and two more at Ginkgo Bioworks in Boston, I was often disappointed to see my engineering thwarted by biology: toxicity, slow growth, and mutations corrupted my genetic circuits and freed cells from their mandated duties.
I’m not alone in dealing with these problems. Scientists constantly push the limits of circuit complexity, but those designs quickly break. In a large-scale monoculture, engineered E. coli abandon their task of producing mevalonic acid in just 70 generations, due to high metabolic burden and mutation. And when a metabolic pathway for oxygenated taxane production was divided across S. cerevisiae and E. coli in an effort to reduce the burden placed on any one strain, there were gains in productivity, but dire challenges in balancing the ratio of cells to avoid the dominance of a single species.
In nature, massive populations of cells work together, in harmony, without ‘breaking.’ All of the cells contribute to the community’s goals, and this cooperation makes the group more productive. There are multicellular cyanobacteria, for instance, that divide into two states: non-growing, nitrogen-fixing cells altruistically share nitrogen with quickly growing, carbon-fixing cells. Chemical pathways for nitrogen and carbon fixation normally interfere with one another, but this division helps the cells maximize their growth. Mathematical modeling also found that this reproductive division of labor is resistant to invasion by cheaters.
This is just one example of how nature uses multicellularity not just to maximize collective growth, but to protect community stability against a tragedy of the commons. While most synthetic biologists focus on the former, my New Science project speculates on the latter.
Specifically, I aim to understand what happens to community behavior when cells genetically modify one another. My goal is to create cooperative behaviors by using HGT to transfer parts of a genetic circuit between neighboring cells. I’ll use my method to probe how cell-cell relationships dynamically change as those cells swap DNA. Use cases for my project remain hypothetical, but I hope that, in studying HGT, synthetic biologists will think more creatively about stability and resilience in engineered cells.
Microbes enforce cooperation in many ways. Division of labor — such as the oxygenated taxane example — is not the same as HGT. The former enforces cooperation because each community member benefits from shared resources while, in HGT, cells take the “risk" of accepting random DNA from their neighbors because that DNA might encode a gene that enhances their overall fitness. The transferred gene will continue to spread if the fitness gains are large. These swapped genes are often used to create public goods, such as extracellular enzymes.
Last week, researchers from Duke University also showed that microbial communities can stabilize the total number of a gene’s copies by circulating it with HGT; a form of “dynamic redundancy” that ensures critical functions of the microbiome run smoothly even when the populations of each species fluctuates. I was excited to see this insight about microbial cooperation — the way many different species teach each other important genetic instructions for the health of the community. I’m also curious to know how “cheater” strains fit into this narrative.
For instance, genes encoded and spread via HGT often encode public goods, but we know that cheater strains arise specifically to take advantage of those goods! Constant HGT is theorized to enforce cooperation through reuptake of producer genes into cheaters, though, thus turning them back into ‘productive’ members of the community. Experiments have shown that HGT often preferentially spreads cooperative behaviors over cheater ones. Still, much more research is needed.
My project, which aims to use HGT as a tool to dynamically reprogram surrounding cells to partake in cooperative behaviors, will be difficult because, unlike standard methods to build synthetic communities — or an electrical circuit — utilizing HGT to assemble circuit parts into different strains means that the community dynamically changes over space and time.
I plan to build a well-studied, synthetic quorum sensing circuit using HGT. A donor cell will contain half a quorum sensing circuit in the genome and pass the other half into the recipient population with HGT, activating a new donor behavior (e.g. a “kill switch”) once a target number of cells use the quorum sensing system. This project is ambitious for a single summer, but even showing that the donor strain can distribute the correct genetic part successfully would be a promising step forward.
My project aims to provide a blueprint as to how we may further engineer HGT or provide a “toy model” for testing ecological theories. For instance, it would be interesting to know if we can cause reuptake of a genetic “backup copy” to extend the lifetime of a synthetic circuit. For now, the project would lay down a blueprint for a circuit where HGT is used to dynamically modify the behavior of surrounding cells to cooperate with each other, much as nature does all the time.
Though there is a lot of work ahead, melding theories from ecology and evolution with synthetic biology may increase the panel of strategies to solve grand challenges in sustainability, biodiversity, disease, and climate change. Scaling biology and fermentation is difficult right now — some companies wait more than a year to rent a bioreactor, and engineered microbes can quickly lose their abilities when grown at scale. In understanding and engineering group cooperation, perhaps we can accelerate the planet’s adoption of a circular bioeconomy, and better use microbes to produce our medicines, materials, and biofuels.
Edited by Niko McCarty
Thanks to Wendell Lim, Devan Shah, Sasha Targ, and Adam Arkin for reading drafts of this essay.
Cite this essay:
Stockard, R. “Engineering Stable Microbes." newscience.org. 2022 September. https://doi.org/10.56416/091wuj