Ensuring Safety

Safety was a significant concern in the design of our project. We wanted to make sure not only that we were safe when working on our project, but that the final system that we designed would be safe to use in the field. We worked to ensure this in a number of ways, which can be outlined below:

Laboratory Personal Safety

Figure 1. Showing off our PPE

All of the students working on the Calgary 2013 team received all of the necessary training as outlined in the University of Calgary’s safety policy. In brief, this included a Biosafety course as well as Workplace Hazardous Materials Information Systems (WHMIS) training. In addition, students took a course introducing them to the necessary laboratory skills where they learned how to perform required techniques both safely and efficiently.

The project that we undertook posed little risk to the students, the only organism used in our project was a commonly used lab-strain of Escherichia coli, which is designated as a biosafety level 1 organism. All necessary chemicals and reagents were handled with care and Material Safety Data Sheets were consulted when dealing with new compounds. Finally, students working in the lab were also supervised at all times by senior lab members, advisors or our lab manager.

Safety in the Design

In addition to the safety of ourselves, the safe design of our system was also a major concern. As our final system will be implemented directly in a feedlot setting, it is very important that we design it not only to be extremely easy and quick to use, but also so that it will be completely safe for the person using it as well as the environment it will be used in. We incorporated safety into our design in a number of ways. First of all, we decided to use an in-vitro system. Although we are engineering our bacteria to contain our genes of interest, we are using purified proteins from the bacteria in our actual test. This not only allows us to make use of a small, test strip that is able to be stored for long periods of time, but it ensures that our genetically modified bacteria will not be released into the environment no matter what happens to the strip system. This will allow for easier disposal of the strips as well as increased safety for the user.

Figure 2. Design of test strip for our system

Our next safety concern was surrounding the functionality of the prototype. Although we would not be using any actual bacteria in it, it was important to have something in the strip that would allow for a positive control, similar to what we see for pregnancy tests. In that system, one blue line appears in the test window in order to tell you that the test is working as it is supposed to, a sort of 'positive control' for the flow of the liquid and for the reagents. In order for our system to achieve similar reliability, we need the same thing. For our strip, the best way to ensure that our test is working the way we expect it to would be to have an actual positive control. In our system this would obviously need to be the shiga toxin gene, in order to verify in each test that our TALEs are binding as expected. Now this poses a problem for us. Not only is this not something that we could amplify from the genome ourselves or get synthesized by a company, but it is not something that we would really want to drench our strips in to sell to users. As such, we needed to find a safe way to do this. So we went and consulted with Dr. Glen Armstrong, a researcher at the University of Calgary who studies pathogenic E. coli.

Dr. Armstong had a number of suggestions for us in terms of refining our TALE binding sequences. He advised us to consider looking for multiple strains of E. coli, essentially creating a multiplexed system. To that end, he helped us to restructure our system to incorporate two different target sequences, as we had initially planned on using only one target sequence for the TALEs. This would allow us to detect a greater breadth in the strains of E. coli that we can detect. Finally, he put us in touch with E. coli vaccine manufacturer BioNiche, who gave us some great insight into the applications of such a sensor as we explain more in depth on our Informed Design page. Importantly, Dr. Armstrong also offered to provide us with DNA to test our final system, which would be critical once a final prototype was developed. Although this would get around our problem of not having the shiga toxin gene, this still left the issue of our positive control unsolved. After some thinking, we identified a small, 18-base pair region of the gene that we could use for this purpose. Based on our TALEs, this should be enough to see binding, however such a small fragment of the gene would not pose toxicity concerns to the public.

Assessment of Accuracy

Outside of safety for the tester, we also wanted to consider one other aspect of safety: safety for the industry. When talking to the feedlot, they brought up the issue of false positives. Although any false positives would hopefully be caught during factory screening, they would have the potential to devalue the cattle that they were found in. This would pose economic risks to not only the feedlot, but to other parties in the chain, including the consumer. In order to mitigate this risk, our test, when and if it reaches a commercial stage, would need to have an absolute minimum rate of false positives. This was part of the reason that we chose to use an in-vitro system, as we hoped that it would decrease variability and improve the robustness of our test.

Although false negatives are also very important, as they would lower the overall quality and reliability of our final product, they would not necessarily pose any new safety risks, as our system is designed to work in parallel with current industry practices. Regardless, we included a line for negative controls in our strip to verify the accuracy of each individual test. This negative control line would contain a sequence of DNA that is not homologous to our stx2 target. If a colour change was observed on this line, this would indicate our system is non-specifically binding to incorrect nucleotide sequences, indicating a problem. In order to have a quality product however, the rate of false positives and false negatives will need to be assessed.

Our Safety Forms

Basic Safety Form
Extended Safety Form - Part 1
Extended Safety Form - Part 2
List of Genes for Safety Form