Preliminary Design Considerations

FRED and OSCAR have been tasked with jobs that require them to be outside of a laboratory environment. Our discussions with industry experts emphasized the need to design a system that minimized the chance of bacteria escaping into the environment. Despite our belief that due to the increased metabolic load FRED and OSCAR are undertaking they would not be able to outcompete any native bacteria, we took these concerns to heart when we designed our project. We have designed multiple layers of controls for each system, utilizing both biological and physical controls.

Physical: The first line of defense

The best way to prevent FRED and OSCAR from spreading into the environment is to make sure that they cannot get to it. As such both our bioreactor and biosensor prototypes involve isolating the bacteria in closed systems. In our biosensor we seal the tubes with a one way valve with FRED trapped inside. The tailings sample is added through the one way valve and then when the testing is done the cap is twisted slightly to release bleach into the sealed system. After the bleach is added the tube is disposed of in a safe manner.

OSCAR presents more of a challenge, as he needs to remain in one place for an extended period of time to perform his tasks. For this we have created a bioreactor house for him. In this sealed system filtered air is bubbled in to keep oxygen levels optimal while a HEPA filter is used to screen air coming out. To extract any hydrocarbons from the reactor a belt skimmer is used that selectively picks up oil while leaving bacteria behind. When the oil is separated from the belt skimmer it is exposed to UV to kill any bacteria, and then is placed into a fractional distillator that heats to 400°C to separate the hydrocarbons from each other.

Biological: Preparing for the worst

Although industry experts felt a genetic safety mechanism was important, they felt that it needed to fit into a cost effective remediation solution. It was stressed that price is a paramount factor to favor adoption at industrial scales, and so an inexpensive system would have a greater likelihood of being implemented outside the laboratory. Experts were also more concerned about the spread of DNA over the death of our organisms. As such, we needed a system where we avoided lysis of our cells, so as to prevent possible uptake of DNA by other surrounding organisms in soil and water, something that has been criticized in previous iGEM systems.

We settled on an inducible ribo-killswitch system. Although inducible systems have been shown to have a tendency to mutate, rendering them ineffective and allowing possible escape, they are more cost effective than other strategies. An auxotrophic marker could have been used for example, where a deletion in the genome would make the organism dependent on an externally supplemented metabolite. Although a mutation restoring the metabolite would be sufficiently complex as to be rendered improbable, metabolic supplements are considerably more expensive than the glucose and metal ions that our systems require. As such, we used an inducible system, but took several approaches so as to mitigate the risk of mutation.

Firstly, we engineered redundancy into our system. By using two kill genes, both would have to be rendered inoperable for the kill switch mechanism to malfunction. Knudsen et al. (1995) proposed that active kill switches containing a single kill element were subject to a mutation rate of 10^-6 per cell per generation, but that a second kill gene under a common regulatory element reduces this value by two orders of magnitude. The mutation rate of a system with two kill genes under independent controls might approach 10^-12 (Knudsen et al 1995). Secondly, the kill switch mechanism is, of course, only a failsafe measure for controlling our organism's spread. The primary means of preventing its escape is through the multiple layers of mechanical security provided by our bioreactor and biosensor. Only when these measures fail will the kill switch be required to function. We felt that this system would also mitigate the concern that an auxotrophic control mechanism would kill the organism without degrading its genetic material, thereby making possible horizontal gene transfer to other organisms.