Project Safety

Overview of Safety Measures

As our goal is to create a fully implimentable IBD therapy, making sure that our product will eventually be safe enough for human consumption was of great concern to us. Find out below what approaches we used to tackle our safety challenges!

E. coli Nissle 1917: An Unconventional Chassis

Our first, perhaps most noteworthy, safety feature was choosing a non-pathogenic chassis, Escherichia coli Nissle 1917. We decided on this chassis as it is non-pathogenic and commonly used as a probiotic in patients with IBD (1). Its use as a therapy makes it adequate for human consumption and may add to the therapeutic effect of our treatment. Furthermore, it is an E. coli strain, which is reasonably easy to work with in the lab.

Genome Integration

The second security feature we worked on was genome integration. We decided to integrate our genetic construct for two main reasons: To avoid horizontal gene transfer and forego antibiotic resistance.

As bacteria can exchange genetic information by swapping plasmids through a process called horizontal gene transfer, our GMO could also pass on our plasmids to other strains. As we are unsure how our plasmids behave in other strains, we wanted to avoid this. Furthermore, with antibiotic resistances present in the plasmid, horizontal gene transfer could lead to the spread of antibiotic resistance in the gut microbiome, including to potentially harmful bacteria (2). Genome integration not only to prevents the spread of resistance genes but instead gets rid of them altogether. As constructs inserted into the genome are more stable than plasmids and cannot simply be extruded, there is no need for constant selection pressure, as is usually done with antibiotic resistances. Thus, having our construct integrated into the chromosome of E. coli Nissle would allow us to have an antibiotic-resistance-free drug. (3)

For integration, we decided on the 'clonetegration' method presented by St. Pierre et al. (4), as it is faster than traditional genomic integration and was included in the iGEM distribution kit. We used the pOSIP-KO plasmid, which is supposed to integrate at the bacterial attB site, a conserved transposon landing site. The integration proved to be problematic. First, we could not transform bacteria (neither E. coli Nissle nor Mach1) with the pOSIP-KO from the iGEM kit. After a transformation was finally succesful, it was sent for sequencing and did not match the desired sequence for the pOSIP-KO plasmid. Because of this, the plasmid was ordered from Addgene.

After we got the plasmid from Addgene, we were finally able to transform bacteria and continued following the integration protocol. However, even after multiple attempts and tweaks to the original protocol, we did not get any integrants. To ensure that the problem was not the size of our construct (consisting of promoter, nanobody and secretion system), we also tried integration with only the promoter, GFP and NorR. This approach did also not yield any integrants.

Due to this, we were not able to successfully integrate our genetic construct into the genome. We hope that future teams will continue working on this issue. A next step would be to test how genomic integration affects our construct's NO-sensitivity and GFP expression.

Kill Switch

We also planned to include a kill switch in our construct so as not to release genetically modified organisms into the environment. We had decided on a two-input kill switch presented by Rottinghaus et al., which can be triggered by either the presence of Anhydrotetracycline or a drop in temperature below 35°C (5). The kill switch works by cutting the host strains' genome (E. coli Nissle 1917) at multiple sites using CRISPR-Cas9, thus destroying all templates of one genome. Bacteria cannot repair a double-strand break without a template,therefore cutting all genomic copies of one gene is lethal for the host. Since kill switches impose enormous selective pressure on the host, bacteria usually evade kill switch mechanisms quite quickly, whether that be through developing resistance or getting rid of the plasmid. Though the kill switch by Rottinghaus et al. has not specifically been tested for plasmid retention, other CRISPR systems have shown stable plasmid retention over 17'000 generations, compared to the 140 generations reached by a toxin-antitoxin killswitch by Sterling et al. Further, our chosen kill switch retains high killing efficiency after 28 days in vitro (5).

As we reached out to the paper's authors, they were kind enough to agree to send us the plasmid. Unfortunately, the material transfer agreement got delayed, so we did not get our strain in time. However, we hope future teams will consider this system for their projects.

Lab safety

To keep our lab members and the environment safe, we always followed the safety procedures put in place by Professor Jinek's and Professor Seeger's lab. This included proper separation and disposal of waste containing antibiotics, cell culture waste etc. We made sure to follow all safety requirements of a biosafety level 1 lab.

References:

  1. Kamada, N. et al. Nonpathogenic Escherichia coli strain Nissle 1917 inhibits signal transduction in intestinal epithelial cells. Infect Immun 76, 214–220 (2008)
  2. Burmeister, A. R. Horizontal Gene Transfer. Evol Med Public Health 2015, 193 (2015).
  3. Hägg, P., de Pohl, J. W., Abdulkarim, F. & Isaksson, L. A. A host/plasmid system that is not dependent on antibiotics and antibiotic resistance genes for stable plasmid maintenance in Escherichia coli. J Biotechnol 111, 17–30 (2004).
  4. St-Pierre, F. et al. One-step cloning and chromosomal integration of DNA. ACS Synth Biol 2, 537–541 (2013)
  5. Rottinghaus, A. G., Ferreiro, A., Fishbein, S. R. S., Dantas, G. & Moon, T. S. Genetically stable CRISPR-based kill switches for engineered microbes. Nat Commun 13, (2022)