Before entering the laboratories, it was mandatory for us to take the Safety Tour organized by the Laboratory Coordinator at our Host Institution; Center for Synthetic Biology at University of Copenhagen. All team members, incl the ones that might not be working in the lab later on, took this safety tour to ensure safety and education for the team as a whole. Especially proper waste disposal of GMO material and personal lab safety was underlined.

Due to our choice of bacterial strain (see more below), we were given instructions as to how we handle our strain with extra precautions in the laboratory, including:

Creating a buffer zone in the lab, that only members with the right protection may enter while handling the bacteria
Wearing mouth, nose and eye protection
Wearing lab coats at all times

These precautions were decided upon in collaboration with our internal Safety Coordinator and Head of Studies Kirsten Jørgensen from the University of Copenhagen.

Bacteria: Evaluating the safety risks

Our bacteria proved to be a challenge when we tried to ensure we followed proper safety guidelines.

Bacteria

The injectisome system needed for our project could be obtained from several different bacteria (1).

We chose to work with a chassis encoding the type three secretion system (T3SS) from Enteropathogenic Escherichia coli (EPEC). EPEC can cause diarrhea in humans when it adheres to the intestinal tissue and successfully injects effector proteins through its 'needle' - the T3SS.

We chose this particular strain since it fulfilled several of our safety requirements:

Non-pathogenic chassis E. coli K-12
The strain had effector protein, promoters and transcriptional regulators removed. There was therefore no risk that the bacteria we would be working with would be able to inject its normal disease causing proteins through the T3SS.
The assembly of the T3SS was inducible

By choosing this particular non-pathogenic bacterial strain we tried to minimize the risk that working with it would bring.

The natural attachment site is intestinal cells, which is why mucus-lining protection has been implemented in our safety precautions.

Transport restrictions

Our first choice was from the Salmonella enterica subspecies enterica serovar Typhi, that turned out to be on the Australia Groups list (2) and therefore regulated.

The Australia Group rules apply not only to whole organisms but also, as stated on their website, to "Genetic Elements and Genetically-modified Organisms" - that includes "any gene or genes specific to any listed bacterium... [which] could endow or enhance pathogenicity."(2)

A chassis containing the Salmonella T3SS would therefore qualify for regulation.

After discussion with Piers Millett from the iGEM Safety Counsel and some research on the Danish safety regulations, we were informed that acquiring and transporting the bacteria might require us to obtain extra permissions.

We chose to substitute the Salmonella T3SS part with the EPEC version, to side step problems about permissions, since this part isn't on the Australia Group's List of human and animal pathogens and toxins for export control and is therefore not regulated in the same way. This meant that we were able to obtain a bacterial strain with a working injectisome (T3SS) for our experiments.

Since the two injectisome systems could be considered functionally - if not genetically - equivalent, this made us question why only the Salmonella T3SS is regulated. This question prompted Piers Millett to provide the Australia Group with an overview of how our project connects to the challenge of functional equivalence in genetic elements between listed and unlisted bacterial pathogens. We have been informed that this topic may be discussed at the next formal implementation meeting of the group.

Working in a level 1 or 2 lab

To ensure we worked in a laboratory that followed the proper safety regulations, we needed to determine whether a level 1 or level 2 lab was necessary.

As stated on the iGEM Safety Hub:

"Virulence factors refer to the properties (i.e. gene products) that enable a microorganism to establish itself on or within a host of a particular species and enhance its potential to cause disease. Virulence factors include bacterial toxins, cell surface proteins that mediate bacterial attachment, cell surface carbohydrates and proteins that protect a bacterium, and hydrolytic enzymes that may contribute to the pathogenicity of the bacterium." (3)

This means that the Injectisome and the connected effector proteins can be considered virulence factors and can cause disease in humans or mediate bacterial attachment to mammalian cells.

We chose to work with a bacteria with the natural disease-causing effector proteins removed. Therefore we were allowed to work in a level 1 classified laboratory. Still, this required us to apply for permission from Danish Working Environment Authority (Arbejdstilsynet).

This decision was taken after a discussion between ourselves, our supervisors and our internal Safety Coordinator at University of Copenhagen, Dept. of Plant and Environmental Sciences, Kirsten Jørgensen. The decision was later confirmed by another Safety Coordinator connected to the University of Copenhagen and sent to the iGEM Safety Council to confirm that we were working in the appropriate laboratory both by direct contact with iGEM representative Piers Millett and by submitting a safety check-in form.

Mammalian vs plant cells in experiments

To test our concept we wished to show secretion of protein by the injectisome and through a membrane. To avoid working with mammalian cells, we decided to only use artificially constructed lipid liposomes and onion cells to prove our concept.

Experimenting safely with chicken eggs

Since we wanted to experiment using the egg yolk membrane from unfertilized chicken eggs, we made sure to get the needed permissions. In order to get the permissions which included only working with eggs declared Salmonella-free.

Integrating Safety Considerations in our Design

We took several safety considerations into account when we were deciding on how our system should look and who should use it.

1: Safety in our system

Early in our project design process, we chose to avoid using the injectisome as a direct drug-delivery system, since this method has several integrated risks: Controlling the amount and purity of the protein as well as ensuring a correct folded and fully functioning protein.

To avoid this we decided to make an optimized protein production and purification system that naturally has a step between collection by an expert and injection into a patient. The protein product should in theory be pure and have all the advantages that using the injectisome brings, but the product is still testable by experts use by being secreted directly into the collection chamber - instead of the patient's body.

This choice has some drawbacks: The equipment and time needed to transfer the protein from collection-chamber to the patient would be increased?? However, we decided that this method would be superior as it would remove the risk of patients being injected with non-functional protein that is in the worst case scenario toxic.

To avoid GMO contamination of the environment, we designed the bacteria, media etc. to be contained in closed chambers.

No kill-switch was included in our bacteria, since the environment on Mars can act as a natural kill-switch. No E. coli bacteria would be able to survive those conditions.

2: Safety in our choice of user

We considered making our system accessible to the general public. The ability to e.g. produce insulin and other protein drugs at home seemed like a the good goal.

We decided against this approach when we realized the risk of mishandling the system, the bacteria and environmental contamination. Since we designed our system to naturally implement a step meant to be used for testing the protein product and ensuring it is both effective and safe, making the system usable by the entire population seemed problematic.

To ensure the safety of the patient using the protein product we changed our target users to include educated personnel only.

Any astronauts using our system on Mars would therefore be required to undergo training before use.

3: Dual-use concerns

The natural function of the injectisome in bacteria is to produce and inject disease-causing proteins into mammalian cells. Therefore our system harbors the inherent risk of being misused to produce toxic proteins.

To evaluate this risk of dual-use, we reached out to our national institution The Centre for Biosecurity and Biopreparedness (CBB). As stated on their website, The centre "is the national authority that issues licences to research institutions, pharmaceutical companies, hospital laboratories etc. to allow them to work with biological dual use components" and was therefore the perfect candidate to evaluate our project. (4)

Analyst at CBB Jacob Hofman-Bang, PhD (Molecular Biology) was very helpful in answering questions and he conducted a dual-use assessment for our project.

The conclusion of the assessment was that our Protein Printer system does not, in its current form, possess dual-use potential. Though it is stated that at a later stage: Were it is possible to produce large amounts of proteins or to produce currently impossible proteins, we should contact CBB again, as such a system would need to be reevaluated.

Full Dual-use report

From Analyst at CBB Jacob Hofman-Bang, PhD

By an Act of the Danish parliament in pursuance of United Nations Security Council Resolution 1540, materials that could be misused for the development and use of Weapons of Mass Destruction - in casu biological weapons - must be secured against diversion and acquisition for nefarious purposes.

Your project aims at artificially expressing a protein in E. coli and concomitantly purifying it during growth of the bacteria. The Danish Biosecurity Act(5) states that it requires a license from the Center for Biosecurity and Biopreparedness (CBB) to work with genetic elements associated with pathogenicity from any of the biological agents depicted in the control list. As your project does not include genetic elements associated with pathogenicity, you do not need a license to continue your work.

In addition to requirements for work with physical elements such as biological agents and genetic elements, CBB also assesses the dual-use potential of such a project, that is, the potential of the system to be used for production of harmful agents or toxins. As your specific system (theoretically) is suited for production of a toxin, we need to consider the amounts of toxin that can be produced, how easy is it to produce and the knowledge and skills which are needed in order to assess whether the technology is dual-use and thus subject to control.

Our assessment of your system is thus:

The biomimetic membrane you plan to use is, at this point, not available from a commercial supplier and therefore not readily useful for creation of a biological weapon. Should it one day be possible to buy such membranes from a commercial supplier, it will be a limiting factor in terms of the biosecurity aspects of the project.
The liposomes, that you will synthesize yourselves, can only be produced in limited quantities. They are therefore a limiting factor for the technology at this point of time.
The amount of protein produced via these injectisomes would roughly require a membrane area of 30 x 30 cm in order to produce multimilligram amounts of a protein/toxin. The membrane is therefore also a limiting factor at this time.
Proteins secreted through injectisomes will have to unfold and refold and therefore proteins having disulfide bridges are not likely to fold correctly once secreted. That will limit the dual-use potential of the system as toxins containing disulfide bridges are not likely to be produced with the system.
Toxins without disulfur bridges may theoretically be produced by this system, but it is not known at this point.

Our assessment is therefore that your project does not, at this stage, possess dual-use potential. Should the project advance to a stage where it is possible to produce large amounts of protein or to improve the ability to form disulfide bridges, you should contact CBB, as the project would then need a reevaluation.

Safety for the patient

We identified several safety risks inherent in the more “traditional” use of the injectisome system already exemplified by Team HokkaidoU 2010 , who aims to use it as a direct from bacteria to cell delivery system. These identified issues caused us to ultimately take our system in a different - and more safe - direction. There are several dangers in enabling direct delivery of proteins produced by bacteria which are introduced directly in to a patient’s system and the following is a list of issues, in no particular order(6):

Differing secretion efficiency means that ensuring the correct and regular dose of protein injected by the T3SS system is difficult to ensure. And different proteins might have different secretion rates. The secretion efficiency of identical proteins might diverge due to effects of the host environment etc.
Ensuring proper folding in 100 % the produced target-protein is essential to ensure correct doses of medicine and avoid negative immune reactions being triggered by improperly folded proteins or protein-aggregates.
Specific targeting is very important to ensure the system targets the correct cells.
Secretion timing is important for patients needing different quantities of proteins at different times.
Ensuring a stable level of bacteria is a concern. Too high or low levels of bacterial propagation would mean that the number of functioning bacteria in the patient’s system would be difficult to keep stable, in effect creating further dosage regulation problems.
Designing a bacterial killswitch after proper dosis has been given or problems has occurred.
Triggering of immune responses would be difficult to predict and deal with.

Furthermore, there are innate problems with live therapeutic systems:

Evolution of bacteria leading to at best the loss of therapeutic effect and at the worst the gaining of a harmful effect. It is difficult to ensure the bacteria’s genetic stability. Transfer of secretion ability to harmful bacteria - or gain of toxic ability from surroundings. Introducing GMO bacteria into nature carries risks.

Because of these safety concerns we decided that our project should, while having the same goal of exploiting injectisomes for medicine and useful protein production, separate this production from the delivery in to a patient. We wanted to allow for collection of pure protein to ensure proper testing before injection into the patient. This removes the direct protein delivery ability from our system but prioritizes the patient’s safety.

Sources

(1) Cornelis, G. R. (2006). The type III secretion injectisome. Nature Reviews Microbiology, 4, 811. Retrieved from http://dx.doi.org/10.1038/nrmicro1526

(2) Australia Group (July 2017). LIST OF HUMAN AND ANIMAL PATHOGENS AND TOXINS FOR EXPORT CONTROL. Viewed 17:37 11/09/2018. URL: https://australiagroup.net/en/human_animal_pathogens.html

(3) iGEM (2018). White List. Viewed 17:35 11/09/2018. URL: https://2018.igem.org/Safety/White_List#FAQ

(4) The Centre for Biosecurity and Biopreparedness (n.d.). Viewed 18:00 11/09/2018 URL: https://www.biosecurity.dk/

(5) Annex 1 to Executive Order no. 981 of 15 October 2009 with subsequent amendments and Annex 1 to Order no. 803 of 22 June 2017

(6)Walker, B. J., Stan, G.-B. V., & Polizzi, K. M. (2017). Intracellular delivery of biologic therapeutics by bacterial secretion systems. Expert Reviews in Molecular Medicine, 19, e6. http://doi.org/10.1017/erm.2017.7

Team:UCopenhagen/Safety

Safety

Safety Tour First