Building a Kill Switch in SBOL

Completing my first biocomputing project

Maggie Chua
7 min readApr 11, 2021
Photo by Michael Schiffer on Unsplash

A while back, I wrote an article titled, The Future of Biology: Biocomputing. For those of you who are new, biocomputing is the idea of using biological parts (i.e. DNA, RNA, proteins) to create biological computers. When I say computer, we’re not talking about the physical device that you are using to read this, but the fact that cells can receive inputs, process them, and create outputs the way that computers can.

As a result, one of the applications that this can be used for is kill switches. Kill switches are biological safety mechanisms that cause a bacteria to die when exposed to certain environmental signals. Environmental signals include light, temperature, biomolecules, etc.

There are two potential routes you can take when building a kill switch: 1) add an environmental signal that switches the bacteria from the survival state to the death state or 2) reprogram the bacteria to rely on an environmental signal to survive, so when you remove it, it will die.

For instance, an iGEM team reprogrammed GM crops grown in vertical farms to contain a kill switch that causes them to die when exposed to sunlight. In the second case, several different groups reprogramed bacteria to die when the temperature reaches below a certain temperature. By consistently keeping a survival temperature, the bacteria will continue to receive the signal and stay in the survival state.

Project Details & Design

In this project, I based it on a research paper where they constructed a kill switch that relied on the addition of specific biomolecules to change states. Below is a diagram of the design:

I remade the diagram because screenshotting would make it blurry, but all credits go to

In the diagram, there are two repressor proteins, which are TetR and LacI. Since they mutually repress each other, when one is expressed, the other is repressed: If TetR is expressed, it will repress LacI, and vice versa.

Since the gene downstream from LacI encodes for the toxin, expression of LacI would lead to toxin production and death. On the other hand, the gene downstream from TetR expresses mcherry, which is a type of fluorescent protein. Fluorescent proteins are useful for research purposes as when viewed under a fluorescent microscope, it emits a vibrant color signal.

For this example, the environmental signals are biomolecules ATc and IPTG. ATc represses TetR while IPTG represses LacI. If we are in the survival state where TetR is expressed, by adding ATc, we prevent TetR from repressing LacI. As a result, since LacI is expressed, toxin production will begin, and the bacteria will die.

  • TetR expressed — inhibits LacI — LacI can’t be expressed — no toxin production
  • add ATc — inhibits TetR — LacI can be expressed — toxin production

In the opposite case, if we have LacI being expressed, it will produce the toxin and induce death. Even if IPTG is added, it won’t be enough to save the bacteria and switch into the survival state. However, if LacI was instead attached to another fluorescent protein (i.e. Green Fluorescent Protein or GFP), the circuit would be able to switch between both states. Instead of survival or death, it would be red (TetR) or green (LacI) signals.

Platform: SBOL

To design it, I used a program called Synthetic Biology Open Language (SBOL), which is a software program that is used for designing genetic circuits. Genetic circuits are creating using biological parts so that cells can respond in a logical manner.

A perfect example of this is…a kill switch! But you probably guessed that already.

image of SBOL logo from

Originally I wanted to use other platforms, but they weren’t user-friendly and I couldn’t find any tutorials online. So if you are a beginner to this (like me :) I highly recommend that use SBOL to start designing your circuits.

If you are interested in learning more, my notion project page includes some resources that I used while I was doing research.


Being the good research student I am, I decided to add this section because I think it’s important to also acknowledge the limitations that certain projects/technologies face. These limitations are specific to this kill switch design, but may apply to other kill switches and by extension, genetic circuits.

Transcriptional Leakage

In most kill switch designs, when the bacteria is in the survival state, they have to repress the other protein to prevent toxin production. Even when there is strong repression, the toxin can still be produced, but in lower amounts. Because of this, the toxin can still accumulate and the bacteria may die. Another thing is that strong repression means a higher expression of your other element. With stronger expression, that means increased toxicity and decreased survival.

There is a kill switch design called an antitoxin/toxin kill switch which fixes this issue. However, in cases where transcriptional leakage isn’t accounted for, it can become a problem for circuit function and stability.

Dependent on human surveillance

In the future, if we want to release GM bacteria to serve as a solution for applications such as bioremediation, it is necessary for the bacteria to have its own regulation system where in the case of mutation, it self-destructs. In this specific design, someone has to add the biomolecules into the system to switch states and induce death. However, if we want to release bacteria into an open system, the kill switches need to function independently and autonomously without causing harm.

No Evolutionary Advantage

When circuits are programmed into bacteria, they can be a metabolic burden, meaning they take energy and resources that are usually used to ensure the bacteria’s survival. The more complex the circuit, the higher the metabolic burden. By taking these resources, the circuit is lowering the bacteria’s chances for survival and increasing genetic instability. Since all organisms optimize for survival, it can cause the bacteria to evolve the circuit out.

What I learned from this process

Something that I’ve realized is that after I complete my projects, I usually don’t debrief on them. Of course, I think about things that I would do differently and new insights that I had, but I never capture them. So, I’m planning to add a reflection part to every project I write about in the future.

  • Break things down as much as possible — Ever heard of breaking down your task into smaller parts, so it’s easier to make progress on them? Well, that was something I thought I did when I started working on this, but looking back, I realized that the tasks I put down still weren’t specific enough. For example, my initial task list included: +build in SBOL +write article +film, edit, upload video. However, I could have split things down even further to make myself feel like I was making more progress. That way I could build my momentum.
  • Create balanced thoughts — I kept stressing about the entire project, especially about how I would fail at it, get my information wrong, or a bunch of other things. Instead of countering negative thoughts with positives ones, I think it’s easier to create balanced thoughts. Basically, you balance the negative and the positive sides while remaining pragmatic about the situation.
  • Keep Producing Bad Work — Without context, that sounds like horrible advice, but what I mean is that when you start something you usually suck at it. If you have no experience or knowledge whatsoever, you stumble around blindly with no direction. What you end up creating will often fail to reach your own high standards (esp. if you are perfectionistic). But I think it’s important to acknowledge that this is the best your current self can do and that you need to keep producing bad work to create good work.

Closing Thoughts

Ultimately, the use of kill switches will be beneficial in the future for biological containment purposes and regulating microbes if we use them for applications outside of the lab. For example, creating bacteria therapies to treat gut diseases, creating bacteria that will break down plastics, or creating bacteria to replace current industrial processes.

Most importantly, investing in biocontainment research is becoming increasingly relevant as our ability to manipulate biology is increasing. Creating safety mechanisms will not only create a backup plan for if things go haywire, but also ensure that we are regulating these tools to create the most positive good.

Hi! My name is Maggie and I am an ambitious 16-year-old looking to impact the world through emerging biotech. At the moment, I’m looking to explore topics such as biocomputing, philosophy, ethics, and climate change.

If you got to the end, thank you for reading my article! Feel free to connect with me on Linkedin or sign up for my personal newsletter if you would like to receive monthly updates on my biocomputing journey!



Maggie Chua

A 17-year-old who knows less about life than she thought she thought. I write about anything that captures my interest.