Advances in our understanding of genetics and ability to manipulate gene expression have expanded the field of synthetic biology quite dramatically over the years. As a result, an increasing amount of people are looking at synthetic biology as a means of achieving a variety of objectives. For example, the field provides a basis for research on gene expression, an efficient way to biosynthesize materials of desirable qualities, potential solutions to environmental issues, and a multitude of other applications. Essentially, a world of possibilities can be found right at our micropipette tips in synthetic biology.

As the field of synthetic biology expands, there exists considerable public concerns over the possibility of genetically modified organisms (GMOs) escaping from the laboratory setting and into the outside environment. If the GMO has a selective advantage over the species naturally present in the environment, it can potentially wreak havoc upon the ecosystem it escapes into. In addition to this, many people are afraid of the possibility that a GMO in the environment may have negative consequences for public health. In order to counter these fears and concerns and increase public support for synthetic biology, our project’s goal is to develop a “kill switch” device that activates cell apoptosis when a cell escapes the laboratory setting.

The switch will be activated through exposure to UV radiation and the cause of death will be the protein expression of ccdB - which is a naturally occurring toxin that interferes with vital processes such as DNA replication and RNA transcription. In other words, the device will utilize inducible promoters that respond to exposure to UV radiation as well as the ccdB coding gene. To control the expression of ccdB, a riboregulators that consists of a cis-repressive sequence and a trans-activating sequence will be utilize.

The gene that will be cis-repressed in our kill switch is the ccdB gene. The ccdB gene is part of a naturally occurring toxin-antitoxin system in which the ccdB gene produces a toxin and a ccdA gene produces an antitoxin. ccdB acts as a toxin by interfering with processes that involves DNA helicase, such as DNA replication and RNA transcription. During these processes, DNA helicase unwinds DNA and creates positive supercoils on DNA. When left alone, the supercoils slow down helicase activity or destroy the DNA itself. Another enzyme called DNA gyrase helps to relieve the tension caused by the positive supercoiling by breaking apart the double strand, adding a negative supercoiling, and rejoining the strands of DNA. When gyrase and helicase are together, gyrase acts ahead of helicase so that the positive supercoils can be neutralized before problems arise. ^[2]

ccdB is able to disrupt helicase activity by binding onto the GyrA subunit of gyrase and messing up its protein structure. This binding causes the gyrase to stay clamped onto the DNA after it cleaves it. As a result, it can no longer add negative supercoils onto the positively supercoiled DNA. Thus, the cell can no longer perform the vital processes of DNA replication and RNA transcription. ccdA acts as an antitoxin by binding onto ccbD, allowing GyrA to be freed up and restoring the gyrase’s protein structure. However, if the ccdA gene is not present, then the ccdB will act on GyrA unhindered, ultimately causing cell death ^[3]. As a result, our choice of poison is none other than the ccdB gene.

In order to control the expression of taRNA, the gene coding for it will be placed downstream of an inducible promoter. Since the condition for cell apoptosis in the death switch is exposure to UV radiation, the inducible promoter would have to be one that is related to this condition. As a result, the promoters of choice are the UV and recA promoter. The UV promoter, as the name implies, is a promoter that is induced by the presence of UV radiation, making it a very straight forward choice for the kill switch.

The recA promoter, on the other hand, does not respond directly to UV radiation. Instead, it responses to the presence of single stranded DNA (ssDNA). ssDNA is in abundance when DNA gets damaged. Since UV radiation, especially short UV radiation, causes damage to DNA, the recA promoter is also a suitable promoter for the expression of taRNA.

The taRNA coding sequence will be placed under the control of the UV and recA promoter as stated above. It is to be noted that the taRNA sequence available to us had two double terminators attached to the end of it already, allowing for more accurate transcription of taRNA. The constructs will look as follows:

The trans-activating construct under two different promoters. There is a double terminator at the end of the taRNA sequence because the part was only available in that form.

Now it may be asked why we did not simply place the inducible promoter right in front of the ccdB gene, thereby avoiding the entire crRNA and taRNA scenario and giving us a “simple” device. Our hypothesis is that the cis-repressive and trans-activating system will be better since riboregulators allow for more control of expression. In order to test this hypothesis, we will be comparing the efficiency of the two system using ccdB and the GFP translational unit as the genes of interest. The “simple devices” we will be examining will be as follows:

The constructs above that will be compared with those in the cis-repressive and trans-activating system. The four constructs are each composed of the target gene (either ccdB or GFP translational unit) downstream of an inducible promoter. Effectively, these constructs are “simpler” versions of their riboregulator counterparts.

The beauty about death switches is that there is more than one way to design one. As a result, we have decided to design another death switch still using ccdB as the kill gene. However, the condition for cell apoptosis will be the lack of arabinose in the surrounding. The death switch we came up with is a bit more complicated than the cis-repressive and trans-activating system as it involves two repressor for a less leaky system (meaning ccdB is not expressed when it is not supposed to) that is based off the naturally occurring arabinose operon. To fully understand this device, let us examine the way the device is designed.

A kill switch based off of the naturally occurring arabinose operon. The parts used are all available as BioBricks in the iGEM registry. It is to be noted that the GFP portion of the device serve no use in this design. It was originally intended to show a successful ligation of a promoter upstream of it, but the pAraC promoter initiates the expression of araC in the reverse direction so GFP is not expressed.

In the device, a promoter for araC (pAraC) initiates the expression of the protein araC. On its own, araC is a repressor for pBAD promoter, preventing the expression of Cubitus interruptus (CI). CI is a transcription factor that would repress the pCI promoter. However, without CI, pCI functions uninhibited and codes for ccdB, thereby killing off the cell. At least, this will be what occurs in a non-laboratory setting. In the laboratory setting, scientists can add L(+) arabinose to the media used to grow the bacteria that contains the kill switch described. L(+) arabinose is a commonly occurring form of the monosaccharide arabinose that is harmless and quite easy to obtain. With the presence of L(+) arabinose (a concentration of 0.001% to 0.02% is more than sufficient), L(+) arabinose will bind onto araC and change its conformation ^[4]. As a result, the pBAD promoter is no longer being repressed, so CI is now expressed. With the expression of CI, the pCI promoter is subsequently repressed so that ccdB is no longer expressed. Thus, the arabinose death switch will be extremely effective as long as scientists add L(+) arabinose to media in the laboratory setting.

The Riboregulated Kill Switch

Arabinose Kill Switch

We searched the registry for UV-responsive promoters, found the following parts, and requested it from iGEM Headquarters:

Part BBa_I765001 – UV promoter

Part BBa_J22106 – recA promoter

It is to be noted that our experience with the UV promoter from iGEM have been a negative one. It seems as though we were unable to isolate the part out of the plasmid it was in. In fact, we even sent it out for sequencing, but it came back with no results. The fact that it was not part of the distribution kit (meaning it did not pass all of iGEM’s guidelines) makes us dubious about the trustworthiness of the part. Nonetheless, because of our experience with the UV promoter, the constructs we designed containing it have been tabled.

The ccdB gene had been BioBricked but was unavailable, so it had to be synthesized. We made a double-stranded piece of DNA (a ‘string’) that encodes a version of the ccdB gene with an upstream RBS that was flanked by the BioBrick prefix and suffix for cloning downstream of the UV-sensitive promoters. We also had the pTet – cis-repressive sequence – ccdB sequence synthesized and flanked by the BioBrick prefix and suffix delivered in a commercial ampicillin resistant plasmid. The taRNA sequence was in the registry with a downstream double terminator and was available in the iGEM distribution kit (Part BBa_J01008).

We achieved ligation of the recA promoter to the trans-activating sequence – double terminator part via 3A Assembly. Unfortunately, all attempts to date to put the ccdB sequence into bacteria, even when downstream of the cis-repressive sequence, have failed, even in a cell line sold by Life Technologies as “ccdB resistant”. As a result, the second part of the riboregulated kill switch system has not yet been tested with ccdB. To test the system, we are in the process of ligating the GFP translational unit downstream of the cis-repressive sequence.

The alternative system using arabinose has fared slightly better. Although we were not able to put the ccdB sequence into this system, we were able to demonstrate that the control system works by ligating a CI-repressible promoter upstream of the GFP translational unit in part BBa_R0051, which also contains the CI gene under the control of pBAD. This resulted in white colored cells if the bacteria were plated on arabinose-containing LB agar with the backbone-appropriate antibiotic and green colonies if plated on media with no arabinose. So we showed that at least the control part of our systems works. We have attempted to ligate a part that contains the promoter and coding sequence for the repressor araC in the opposite direction (part BBa_K808000) with a part containing an araC repressible promoter that codes for CI (part BBa_K611014) thereby creating the system described above, but have been unsuccessful to date.

Like any good scientists, we had to practice proper laboratory safety procedures throughout the entirety of the project. The following are general safety precautions that have been taken during our project and description of the danger involved in the project.

• 1. Would any of your project ideas raise safety issues in terms of: researcher safety, public safety, or environmental safety? There are three recommended steps in addressing this question.
  • 1. To start, please list organisms you are using and organisms from which your parts are derived, indicating the risk group or biosafety level for each.

    We used two commercially-available debilitated laboratory strains of Escherichia coli: NEB 20-beta C3019H and Life Technologies One Shot® ccdB Survival™ 2 T1R. Both these are not harmful to humans and appropriate for use in BSL1 facilities. All the parts we used or made are from E. coli or bacteriophages or widely-used fluorescent proteins and are not toxic to humans.

  • 2. Then consider risks to team members, publics and environment if the project goes according to plan. Please describe risks posed by lab equipment and chemicals as well as biological parts and organisms.

    Since the goal of the project was to create an even safer construct in E. coli though the use of a gene damaging to the bacteria that would turn on if it escaped, the project would actually increase public safety. All students were taught good microbiological practices including wearing gloves, avoiding contamination, and wiping down benches with alcohol after use.

    No lab equipment outside of that typically found in a molecular biology lab (microfuge, PCR machine, incubator-shaker, UV light box, gel electrophoresis equipment, autoclave, pitpettors, etc.) was used. Chemical hazards were minimal and included the ethidium bromide used to stain gels, and this reagent was handled with appropriate personal protective clothing. In addition, exposure to UV light was also minimized with personal protective clothing and the use of protections such as goggles and face shields.

  • 3. How are you addressing these issues in project design and lab work? Have you received biosafety training and other laboratory safety training? If so, please briefly describe the training.

    All students participated in laboratory safety training before being allowed to work in the lab area. The training included general lab safety (location of fire extinguishers, eyewash stations etc. and proper use of lab equipment), chemical safety (introduction to potential hazards of commonly used chemicals in the lab such as flammability of alcohols, carcinogenicity of ethidium bromide, the location and use of material safety data sheets, appropriate protective clothing and eyewear), and biological safety in a program designed by MIT’s biosafety officer Dr. Claudia Mickelson and administered by Genspace director Dr. Ellen Jorgensen.

• 2. Do any of the new BioBrick parts (or devices) that you made this year raise safety issues?

  Since the ccdB gene is toxic to most E. coli strains, there is the possibility that if this gene were to be transferred to a wild strain that was resistant to it, it may cause death of beneficial bacteria in the micro-biome of a human or of an environment. However, laboratory strains are deficient in the mechanisms needed to share DNA and are not expected to survive long in the wild (especially if our system works) so we judge the risks to be minimal. Additionally, all parts and organisms were BSL1.

• 3. Is there a local biosafety group, committee, or review board at your institution?

  There is no formal committee at Stuyvesant but the science teachers review all student projects and the school complies with all NYS Board of Education safety standards and guidelines.

  Genspace, where much of the lab work was performed, has a safety advisory board that reviews projects. This project, since it was well within BSL1 guidelines as set forth by the NIH, was acceptable and appropriate for the facilities. All student work is supervised by adults with advanced degrees in the fields of molecular and microbiology.

• 4. Do you have any other ideas how to deal with safety issues that could be useful for future iGEM competitions? How could parts, devices and systems be made even safer through biosafety engineering?

  We feel that our project addresses this question nicely.