A visual based kit for public use

During the design process, we went through multiple iterations of designs. As the need for new antibiotics is higher than ever, we first intended to design an antibiotic screening kit that could be used by anyone, so that as many people as possible can join the search for novel antibiotics. Therefore, initially, the design of our system was based on visualizing bacterial stress by linking stress-activated promoters to chromoproteins, as they are visible to the naked eye. This screening platform could be used to detect substances harmful to bacteria, without any further analysis. It would therefore be perfect for out of the lab usage, enabling anyone to screen for new antibiotics. This would not only be beneficial for the number of substances tested for their harmfulness against bacteria, but it would also raise more awareness in the community of the problem of antibiotic resistance.

However, the RIVM, the Dutch National Institute for Public Health and the Environment, advised us to keep GMO’s in the lab as much as possible. Also, most substances harmful to bacteria will probably be identified in the lab anyway. Therefore, we decided for our next product design iteration to create a kit for lab-use only and to Read more about how we raised awareness by visiting our Education and Public Engagement page raise awareness about the threat of antibiotic resistance using other outreach.

Therefore, our initial design was modelled after an analytical profile index (API) system (Figure 1). The design would contain multiple wells - containing bacteria expressing our stress responsive chromoprotein BioBricks - to which candidate compounds could be added. Should the added compounds be stressful, the wells would change colour. This system would be restricted to laboratory usage only.

Adapting our design to user feedback

During the design of our kit, we involved potential future users to provide us with feedback on our kit design. Coen van Hasselt - an assistant professor at the Leiden Academic Centre for Drug Research who studies the clinical pharmacology of antibiotics - noted that our system came with the disadvantage that a readout would not be quantifiable due to the use of chromoproteins. Contrary to chromoproteins, fluorophores - providing a fluorescent signal - are relatively easy to quantify with a plate reader or using flow cytometry. Additionally, because we were no longer planning to use our kit outside of the lab, the advantage of chromoproteins being visible for any user was no longer as important. Most modern laboratories have access to equipment to measure GFP fluorescence. Therefore, we decided to switch our system to use We decided not to continue with chromo-proteins for our kit. However, we did decide to continue to develop a chromoprotein strain to be used in classical overlay experiments. GFP as a reporter. Because plate reader and flow cytometry experiments are often performed in 96 wells plates, we also decided to provide our kit in a 96 well shaped plate, to facilitate more convenience for fast screenings.

We also spoke to Timo Koopmans, working at Karveel pharmaceuticals. He is involved in the production of antibiotics and advised us that our kit would have to be very easy in use if we want many researchers to adopt our system. One way of ensuring this ease-of-use was by providing our 96 well plate shaped kit with our stress detecting strains included. However, a plasmid - containing our stress detecting BioBricks - can be discarded by the bacteria when they are not growing on selection medium, mitigating the utility of our bacterial cell lines. Therefore, we decided our stress detection BioBricks should be integrated into the genome of our stress detection strains for our next kit product design.

Because we plan to include a living GMO bacterial cell line in our kit, additional safety precautions will need to be taken. Cecile van der vlugt, an expert in the field of regulation and safety at the RIVM, provided us with feedback regarding safe transport of the kit. Particularly, we would have to nullify the risk of our bacteria ending up in the environment. For this, “biohazard safety seals” can be used to minimize the risk. We have decided our kit will contain two seals, one to cover the 96 wells plate, and one surrounding the plate. Moreover, the bacteria will be freeze dried to bring them into an inactive state. The bacteria can only be revived after contact with nutrients and water. To continue reading about the safety measures of our product design, please visit our Safety page.

Based on the feedback of the stakeholders we discussed our project with, we created the rough design of our final product:

• Our system will use fluorophore based stress reporters
• BioBricks will be integrated into E. coli cell lines
• Cell lines will be provided in a 96 well plate format
• Cell lines will be freeze dried and packaged with two safety seals
• The kit should be safe for transport across the globe

Choosing our final promoter repertoire

After determining the rough design goals for our screening kit, we started looking at the details of our product design. Now that we are planning on using fluorophores, we have the opportunity to test multiple stress responses in a single bacterial strain, since different fluorophores can be measured independently based on wavelengths. These fluorophores will be optimized for the most used wavelengths of flow cytometry and plate reader analysis. The wavelength range of most plate readers varies between 340 to 630 nm, with flow cytometers offering similar wavelength ranges. Conversely, the excitation and emission spectra of fluorescent proteins CFP, YFP and mCherry fall precisely within this range and do not overlap majorly with each other. Therefore, we intend on integrating three stress-activated promoters in one bacterial strain, each coupled to a different fluorophore.

During our research, we have isolated and characterized a total of Read about each of our stress promoters on our Parts page 26 stress promoters. Integrating each of these into our system would require nine cell lines, making our kit rather inefficient. Therefore, we categorized our stress promoter repertoire in five already defined bacterial damage responses: cell wall damage, DNA damage, cell secretion/import stress, protein misfolding and cytoplasmic membrane damage. Lastly, we have a group of promoters for which their damage types still need to be further characterized. Of each of these six categories, we will use two promoters. This way we can detect the full stress spectrum using twelve stress promoters, which fit in four bacterial strains. The exact promoters to be used of each category will be determined following further promoter characterization.

Enabling both low and high throughput screening

Based on our four bacterial stress detection strains, we can now calculate that one 96 well plate kit can test a total of 23 substances when one negative control is included (96 / 4 = 24). With this design, we visited the biomedical scientist Dr. P. H. Nibbering. He noted that not all scientists perform high throughput screenings. Particularly, he is focussed on the development of new agents for the treatment of infections with multidrug-resistant pathogens. For this, he only tests a limited number of compounds per experiment. Hence, it would be inconvenient to use an entire 96 wells plate at once. However, only using a part of the kit is not safe as this would mean the already used wells would contain residual GMOs, making the partly used kit unsafe for storage.

To tackle this issue, we designed a 96 well plate shaped container into which strips of eight wells can be inserted. The number of strips can be adjusted to the number of tests needed. We chose strips of eight wells as this perfectly fits two sets of our four bacterial testing strains, and because having strips of only four wells would make it too difficult to firmly - and thus safely - attach them into the 96 well plate container. Each strip would be sealed separately, therefore, the number of tests to be performed can be adapted without unnecessarily opening an excessive number of unneeded wells.

To also facilitate easy high throughput screening of our system, strips of eight wells are linked together through tear tape, to create a 96-wells format. Hereby, the 96-well shaped holder can be easily filled. If an experiment does not require 12 strips, the residual strips can be teared off and stored for later use.

Making our kit ready-to-use

Before conducting an experiment, the freeze dried E. coli bacteria in our kit will need to be revived. This is usually done using LB medium, however, fluorescence measurements in LB are undesirable as this distorts the fluorescence signal. This means cells first have to be washed after stressing, before being measured. Therefore, because our cells only need to survive for a short period after revival, complete LB medium is not required. Because of this, we intend on developing a minimal medium which is optimized to allow for bacterial revival, while minimizing noise during fluorescence measurements. This simplifies the use of our kit even further by eliminating time consuming washing steps.

Our minimal medium will be developed in future research. Moreover, optimal freeze drying conditions will be identified to harm the bacteria as little as possible. Although bacterial cell stress caused by freeze drying has not been measured yet, survival rates of bacteria freeze dried in several conditions have been identified. Gram negative bacteria freeze dried in 10% sucrose solution presents the best survival rates^[1]. Therefore, we will attempt the optimisation of freeze drying bacteria in a sucrose solution in future research.

When combined, the features developed during the previous chapter create an extensive and easy-to-use screening system for detecting bacterial cell stress. We have named our system 50S.O.S., an acronym for our iGEM project Fifty Shades of Stress.

50S.O.S. provides scientists with a high throughput system that identifies substances harmful to bacteria, by means of a fluorescent signal.

It includes four different genetically modified E. coli DH5α strains, each expressing three different stress-activated promoters linked to a fluorescent label. The three fluorescent labels are optimized for the most used wavelengths in plate readers and flow cytometry.

In total, the kit includes 12 distinct stress-activated promoters, able to identify five distinct bacterial damage responses, including cell wall damage, DNA damage, cell secretion/import stress, protein misfolding, cytoplasmic membrane damage.

50S.O.S is made up of a 96 wells plate shaped holder in which strips of 8 wells can be inserted. This enables scientists to adjust the amount of wells in the plate to the number of tests they want to conduct.
Each well contains freeze dried bacteria, to allow for long term preservation, combined with a sufficient amount of minimal medium powder. The presence of minimal medium powder facilitate the user-friendliness of the kit. Reviving the bacteria is achieved by simply adding water to the wells.

Watch the video below for a complete overview of the design of 50S.O.S.:

The concept of our 50S.O.S. bacterial stress detection kit is based on literature and scientific experimentation. We have identified and characterised numerous stress-activated promoters and created BioBricks with these promoters linked to fluorophores. Moreover, the functionality of eight of our promoter BioBricks has been proven by exposing bacteria expressing these BioBricks to non-lethal concentrations of existing antibiotics. Additionally, we have shown our system is able to detect stressful compounds by performing a limited screening experiment which identified ascorbic acid as a stressful compound. You can read more about these demonstration experiments on our Demonstrate page.

This proof of concept was based on the expression of our BioBricks in a plasmid. However, 50S.O.S. will contain bacterial strains in which the stress responsive elements have been integrated in the bacterial genome. This can be performed in the future using CRISPR-Cas9 techniques^[2]. Additionally, the kit will contain freeze dried bacteria. Previous studies have shown that bacteria can be revived after freeze drying in 10% sucrose solution, with a survival rate of 80%. However, optimal freeze drying conditions, to minimize bacterial stress, still need to be discovered^[1].

Based on our strong proof of concept, we are confident our 50S.O.S. kit design can be created into a functional tool in the future. Although developments still have to occur through further characterization of promoters, integration of promoters into the E. coli genome and creation of improved freeze drying and minimal medium, we believe these hurdles will be overcome since each of these goals have already been achieved in prior research.