The beginning of

something beautiful

The idea of a partnership with the Wageningen 2022 iGEM team, Colourectal, quickly arose after finding out that both projects must deal with the same challenge: confining a probiotic GMO within the body of its host. Both teams had already done extensive research regarding the topic of biocontainment and concluded to use temperature as a variable to activate our kill systems. After meeting again during the iGEM meetup in Utrecht plans were made to set up a video call (figure 1) to discuss what grew out to be a fruitful partnership as characterized by the event in the timeline below (figure 2).

Figure 1. First online meeting to discuss our partnership. 3 members of each group were present.

Figure 2. Timeline of the partnership with team Wageningen.

Lab Cooperation

During this first meeting we found out that we shared the use of the same temperature sensitive promoter pTlpA, developed by Piraner, et al. (2017) [1] and converted into a biobrick (BBa_K2500003) by iGEM 2017 Zurich. However, our utilization of this sensor is quite different in principle. In our system the pTlpA(36℃, 39℃, 41℃) is used to control the expression of a TetR gene, which in turn inhibits the expression of a fatal CRISPR-Cas9 casette. The TetR gene should be repressed at temperatures lower than the chicken's body temperature which activates Cas9 and causes cell death.

Colourectal uses the temperature sensitive pTlpA(36℃) promoter to control the expression of the antitoxin ccdA, while the toxin ccdB is under a constitutive promoter. When the temperature drops, the promoter blocks the expression of the antitoxin ccdA, leading to cell death. Therefore, we decided to compare the two systems regarding effectiveness and leakiness.

We did this by both replacing the kill-mechanisms of our temperature sensors by fluorescent reporter genes, with Colourectal choosing GFP and Nanobuddy using mRFP1, the latter of which is explained in more detail in the Engineering page.

We both intended to test our constructs at 26℃, 30℃ and 37℃ in temperature-controlled microtiter plate-readers, measuring OD and fluorescence over time using the same optimized settings. And then use the same data processing methods and an internal standard to compare our constructs. However, due to time constraints we were not able to optimize the settings, acquire the internal standard on time nor test all 3 temperatures. Instead we both individually tested our constructs as much as possible, but with different measurement settings, making comparison less reliable.

In the end, Wageningen tested the three temperatures, while we tested ours at 37℃, allowing only for the rough comparison of the general trend, which can both be seen on the results page and below in (figure 3) and (figure 4).

Figure 3. The results of the temperature sensor fluorescence experiments performed by the Wageningen iGEM team 2022 (Colourectal).

Their construct used the pTlpA(36℃) promoter to control GFP expression in E. coli Nissle 1917 at 26℃, 30℃, and 37℃. Fluorescence was normalized against an M9 blank, and then divided by measured OD to get the relative normalized fluorescence depicted on the y-axis.

Figure 4. The results of the temperature sensor fluorescence experiments performed by our team (Nanobuddy).

The trend in the data from Colourectal is that a clear peak in fluorescence forms, which then quickly drops to 0 within the span of approximately 8 hours, very much unlike our own data, where the trend is a continuous increase in fluorescence/OD over-time. Though it is uncertain why, it appears that E. coli Nissle 1917 is not capable of maintaining the fluorescence from this construct at any temperature, including at the optimal 37℃. This led to their conclusion that their construct doesn’t function and they would use the PcspA instead. For us in E. coli DH5α the mRFP1 is retained more effectively, so the organism might affect fluorescence significantly. Additionally, another obvious peculiarity of our data is the initially low starting point of our relative fluorescence, probably caused by the extremely low cell-density at the beginning of the experiment. Therefore utilizing higher cell concentrations is recommended, as well as the repeating of the experiment with a wider range of temperatures.

Finally, to allow for a more comprehensive data-comparison it is recommended to use a shared internal standard, for example using the serial dilution of the fluorescent dyes from the Interlab study. It is also possible to use a well characterized constitutive promoter for comparison, e.g. using constructs containing BBa_J23100 or BBa_J23102, which we have also constructed. Resulting in a fluorescence fold-change relative to previously mentioned standard. However due to time-constraints this could not be performed at this time.

Further Shared Experiences

Beyond individual lab-work and comparing data, we also regularly met online throughout the summer to discuss and troubleshoot the two biocontainment projects. And when the Wageningen team was not satisfied with the backbone they used for their killswitch, the proposed solution was to try our pTRKH3 plasmid that contains a broad-range host ORI. We sent the plasmid and shared our protocols and snapgene files through a shared google drive folder.

Finally, we also formed a collective Whatsapp group for organization, real-time discussion of various matters, and meme-sharing. We also met physically in Groningen in June, and Hamburg and Utrecht in July to socialize and share ideas. This brought our teams more closely together, and brought our iGEM experience to new heights!

Sonia (Wageningen): This partnership allowed us to troubleshoot the pTlpA construct and to test it with a different backbone. Unfortunately, we haven’t managed to make our construct work, making pcspA is the best option for our project. The monthly meetings and updates with NanoBuddy made us discover how important is to discuss your project and results and get an outside point of view!

Sander (Groningen): Together with Wageningen we managed to more reliably determine the functionality of the temperature sensor component of our kill-switch, and if we lacked the time-constraints we could have streamlined both our kill-switches far more than we ever could alone. Our partnership and regular meetings helped us speed up the design and testing of our constructs by granting us access to a greater pool of knowledge and experience! Not to mention the moral support given, when experimentation in-silico and beyond was difficult! They raised our iGEM experience to greater heights!