The eternal enemy of bioengineers: mutation and evolution

Like Comment
Read the paper you have experiences in facing problems as follows? You have successfully built genetic constructs and confirmed their sequences by Sanger sequencing. Your ‘first’ test showed the genetic circuit behavior you wanted to see. After several months or years of troubleshooting, fine-tuning, and failed cloning attempts, you were ready to celebrate by submitting a paper. Well, your advisor asked you to repeat the same experiments on different days and for a longer period to see the reproducibility and the stability of your genetic circuits, respectively. You were confident because your hard work should pay off. Unfortunately, your second experiment data was not good enough, especially when you tested the constructs for longer cultures. In another scenario, you cloned and expressed Cas9 or dCas9 to engineer your cells of interest, but you had difficulty obtaining sequence-confirmed Cas9 or dCas9 genes and functional CRISPR systems. What is going on?


In many such cases, mutation is one of the major reasons for your troubles. Even computers sometimes do not function well, although we expect them to be perfect. In contrast, living cells tend to adapt to fluctuating environments often by evolution. Thus, mutation or evolution is an intrinsic property of living systems, although it is an enemy to engineers who want control and programmability. My team was not an exception, and it caused more problems to us than any other bioengineers because our project goal is to ‘kill’ our engineered cells once they accomplish their goals.


In many spy movies, betrayed secret agents become bad guys. Microbes want to survive, and if they are forced to die, they find ways to avoid the ‘killer’ mostly by mutations. My student Austin (the first author of this paper) has suffered from this intrinsic problem for more than five years. I conceived the kill-switch idea a decade ago and presented it in multiple conferences. However, due to many issues, including mutation, my initial lab members failed to demonstrate any working kill-switches. Unfortunately, we have been scooped by leading synthetic biology groups.


We claim that our kill-switches are the best ones ever built. This amazing achievement is thanks to Austin’s tireless efforts and perseverance even after multiple failures due to mutations. We have overcome this challenge as follows. First, we used functional redundancy. Today, I did wake up at 3:30 am as usual, but in a hotel room. We had a power outrage after midnight, which triggered the backup generator to provide electricity. Similarly, we installed multiple backup kill-switches to cope with random mutations. Second, we reduced the SOS response-mediated mutation by deleting the error-prone DNA polymerases and recombinase. Third, to prevent the antibiotic resistance spread, we used a plasmid addiction system which allows bacteria to maintain their plasmid without antibiotic selection. Last, to mimic the real-world environment where native microbiota coexists, we tested our kill-switch-containing microbe with the control strain lacking the switch. Using all these features, we have achieved the complete elimination of engineered microbes in mouse models.


Biocontainment is a critical issue for many practical applications that involve the potential release of genetically engineered microbes (GEMs) into the environment. We demonstrate GEMs can be equipped with robust kill-switches that are stable for at least four weeks, generalizable to be used in many microbes, and modular to be connected to diverse microbial sensors. We are currently developing multiple kill-switches with various biosensors in a variety of microbes for bioremediation, biomineralization, sustainable bioproduction, and climate mitigation using photosynthetic microbes. Notably, GEM deployment into the environment should be approved by the regulatory agency and the public. To this end, my team is tirelessly working every day because the benefit or opportunity of GEMs is too huge to miss.

Genetically stable CRISPR-based kill switches for engineered microbes

Nature Communications volume 13, Article number: 672 (2022

Tae Seok Moon

SynBYSS Chair & EBRC Council Member

Associate Professor

Department of Energy, Environmental and Chemical Engineering

Washington University in St. Louis

One Brookings Dr. Box 1180

Brauer Hall Room 3004

Tae Seok Moon

Professor, Washington University in St. Louis