Biosafety is an important issue facing synthetic biology and concerns surrounding synthetic organisms escaping into the environment has prompted the development of novel methods of bio-containment. Many iGEM projects that require an organism to be released from the lab use kill switches to address concerns about the effect of GMOs on the environment. Unfortunately, kill switches - inducible genetic devices that cause cell death - are poorly categorised in the standard registry of genetic parts. There is a distinct lack of quantitative data which prevents them being used with confidence.
After talking to individuals from industry and academia about the strengths and limitations of kill switches, we decided to investigate the effectiveness of different types of kill switch, to quantify their robustness after several generations and investigate the possibility for horizontal gene transfer. We have developed three types of kill switch to cover a broad range of strategies that may be employed in kill switch design. A metabolic kill switch that uses the production of reactive oxygen species to kill the cell, an enzymatic kill switch that uses the production of lysozyme, and DNA degradation mediated by DNAse production.
Our metabolic kill switches builds on previous iGEM projects that have used the expression of highly phototoxic fluorescent proteins to kill the cells by exposing the culture to light. KillerRed and KillerOrange are homologues of GFP which, when irradiated with green and blue light respectively, generate reactive oxygen species (ROS). KillerRed has been shown to effectively kill cells when exposed to green light (540–580 nm) and is much less effective under blue light (460–490 nm) (Bulina, 2006). KillerOrange effectively kills cells when exposed to 450-495nm (Sarkisyan 2015), the range that KillerRed does not. Firstly we aimed to improve KillerRed, an existing registry part, by codon optimising it for E. coli and then improving its characterisation. Alongside this we have characterised KillerOrange. Once we had established how effectively these proteins killed the cells we began growing a cultures containing the kill switches in the ministat.
Preliminary experiments were performed to calibrate the tecan reader. An OD of 0.4 on the cuvette reader was used as the optimum level of growth to induce protein production, this corresponded to 0.26 on the tecan reader.
5ml overnights of the transformed E. coli with the KillerOrange and KillerRed kill switches were used to inoculate five 250ml erlenmeyer flasks covered in tin foil containing 50ml of LB 35µg/ml Chloramphenicol. The five flasks were inoculated with the following conditions.
The optical density of each culture was measured every 1.5 hours until it had reached 0.26 on the tecan plate reader, then 100µl of 0.1M IPTG was added to induce the protein production in the desired cultures. The cultures are then incubated at 37℃ and 220rpm overnight. A serial dilution was then performed to make 10-3,10-4 and 10-5 dilutions of the cultures. 4.5ml of each dilution factor was placed in 10ml falcon tubes one set were placed on their side label down in the light box, the other set were covered in tin foil and also placed in the light box. The temperature was taken periodically inside the box using a thermocouple. After 6hrs of irradiation, spread plates were performed for all the samples. CFU’s were then counted and the dark and light conditions were compared.
Lysozyme is a common enzyme used in laboratories and the gallus gallus form is basis of our enzymatic kill switch. It is bacteriolytic when transported into the periplasm of gram negative bacteria, hydrolysing the glycosidic bond connecting N-acetylmuramic acid and N-acetylglucosamine. Under the control of a T7 promoter we can induce lysis of the cell by adding IPTG. Simply lysing the cell does not prevent HGT it may even exacerbate the problem. We have tested the liklihood of HGT by lysing a culture of cells producing RFP, incubating the lysate in a 90 degree water bath to inactivate the lysozyme. Then growing competent E. coli with the lysate will show if the competent cells take up the RFP gene. Lysozyme EnzChek assaying kit is used to measure lysozyme activity in solution where an increase in fluorescence is proportional to lysozyme activity.
DNase 1 is an endonuclease that non-specifically cleaves DNA. We are creating a kill switch with DNase 1 to address the foremost problem associated with Biosafety - lateral gene transfer. The DNase 1 kill switch, on induction, will degrade DNA and kill the cell. Another biosafety ‘kill switch’ we aimed to test was DNase I. We received a G-block of DNase I DNA and cloned it using the MoClo method. Each time the product from the PCR reaction was transformed we encountered issues. Many times no colonies would form, and in cases when colonies were produced DNase was not present. This could be because colonies containing DNase I has their DNA immediately destroyed. However production of Lysozyme was successful so we continue work with this as a kill switch instead.
Before starting the project we spoke to Prof. Robert Beardmore EPSRC Leadership Fellow in the Mathematical Biosciences at Exeter University. Much of his research has been into antibiotic resistance. We talked about how high selection pressure is applied by prolonged use of antibiotics and how kill switches may be analogous to this. It is clear that cells which develop a mutation that inactivates a toxic protein would be strongly selected for. The length of time for this to happen was estimated to be quite short and if this was the case could have strong implications for the effectiveness of a kill switch over time. To test this we decided to use the ministat to perform a continuous culture. The ministat was developed in the Dunham lab at the University of Washington (Miller et al), each ministat chamber is fed from its own media container via a peristaltic pump and the culture volume is set by the height of the effluent needle in the chamber. Air is bubbled through flasks of water to hydrate it and then used to agitate the culture. We inoculated chambers with freshly transformed E. coli (BL21 DE3) and took samples to test if the kill switch was still viable. Our kill switches were all under the control of the T7 promoter, which we knew to be leaky. It was highlighted to us by Dr Markus Gershater, chief scientific officer at Synthace Ltd, that any leakiness in a system would likely produce sub lethal doses of the toxic proteins accelerating selection for mutants without the kill switch. He also pointed out that chemical and physical means of bio-containment were cheaper and more effective in industry. To investigate this, the ministat was used in our project to simulate on a small scale how a kill switch might be maintained in a large chemostat culture.
References Bulina, M. E. et al., 2006. A genetically encoded photosensitizer. Nature Biotechnology.24(1). Sarkisyan, K. S. et al, 2015. KillerOrange, a Genetically Encoded Photosensitizer Activated by Blue and Green Light. PLoS ONE.10(12) Miller, A. W. et al, 2013. Design and Use of Multiplexed Chemostat Arrays. Journal of Visualised Experiments. (72).
While plasmids are widely used to carry genetic parts, integration into the host genome could prove a more robust approach to introducing genes into organisms. Genome integration removes the need for a selectable antibiotic resistance marker as the parts will be faithfully replicated and the variability of copy number is removed. We are investigating whether integration into the E. coli genome will affect the efficiency of our kill switches and whether they will remain functional for longer in a continuous culture. We have used the lambda red recombination method to integrate our parts into the arsB locus. Integrating in this locus does not affect E. coli growth (reference kiko paper). We used the pKD4 plasmid as a vector to carry our parts, this had a pst1 site and two xba1 sites. Using site Q5 site directed mutagenesis kit we removed these illegal restriction sites. We planned to test how the robustness of a kill switch that was integrated into the genome was different to the same kill switch gene that was incorporated on a plasmid. To do this we aimed to use the Lambda red genome knock in method, using the pKD4 plasmid to integrate KillerRed and KillerOrange into the E.coli DH5α genome at the arsB location. However the pKD4 plasmid we had available contained EcoRI and XBal restriction sites corresponding to the enzymes used by iGEM teams for parts. To resolve this we decided to carry out site directed mutagenesis to change one nucleotide base pair in each sequence of the restriction sites so as to remove them. Primers were designed for use with the Q5 kit. The first attempt using this kit involved at 2 step PCR reaction, this was shown by gel electrophoresis of the product to have been unsuccessful. The protocol was changed to a 3 step PCR reaction and a successful product was produced. The PCR product underwent a KLD reaction and was transformed into DH5α E.coli. The transformation was unsuccessful and so mutagenesis was carried out again and re-transformed. Each time the transformation was unsuccessful. We decided to use another mutagenesis kit, QC multi, which used all forward and reverse primers in separate reactions and could produce multiple mutations at once. Unfortunately, this kit was also unsuccessful. Therefore we decided to focus our efforts on other tasks within the lab.