"

From 2011.igem.org

Human Practices

Our team has decided to go down a novel route in tackling the human practices issues surrounding not only our project but also iGEM in general. Instead of sticking to the established routes of either proposing complete containment or relying on "kill switches" to prevent spread of GM bacteria, we have decided to engineer a containment switch that will not kill our AuxIn bacteria but all other microorganisms that take up the auxin-producing plasmid. In addition, we have consulted many experts and will conduct experiments that demonstrate the safety our device. The true scope of many iGEM projects can only be fulfilled if release is possible and we will be attempting to take a first step towards making this possible for our project.

Historically, iGEM teams have tried to control GM release via two mechanisms: complete containment and "suicide" mechanisms in which the bacteria kill themselves in the absence or presence of specific stimuli. However, when considering the eventual use of many projects – be it bioremediation (e.g. Peking 2010’s project), crop-enhancing projects (e.g. Bristol’s 2010 project) or other applications – full use of synthetic biology organisms will only be achieved by release and full containment is often not a realistic option. In addition, kill switches may be effective to an extent but they are easily selected against by evolution as they present a strong selective disadvantage. Stress defence mechanisms such as the SOS response in E coli add to this effect. In addition, transgenes can be transferred to other bacteria in the environment using naturally occurring mechanisms such as conjugation of plasmids. Finally, while it may be argued that engineered lab strains will quickly be outcompeted, bacteria with GM markers have been found in the environment more than a decade after they were released REF. In some cases, endurance of the bacteria in specific environments may even be desirable.

In light of these issues, we have decided to engineer Gene Guard, a containment switch that will lead to the lysis of natural soil bacteria that take up plasmid DNA from our engineered bacteria. We have consulted experts and the literature about the implications of our project and used this information to design an effective containment switch. However, we also tried to address all possible problems and complications arising from the impossibility of absolute control. Accordingly, we used the information we gathered to influence our release strategy and design.

However, we also acknowledge that this containment switch is never going to be completely effective. Accordingly, we have consulted ecologists and other experts on auxin, plants and soil to ensure that our device is as safe as possible and we can justify release. We researched other organisms such as soil microbes and earthworms that may be affected by the AuxIn bacteria and, with the help of the experts we consulted, devised experiments to test the safety and impact of many aspects of our project.

References:

Human practices panel discussion

In order to discuss the possible implications and consequences of our project, we held two human practices panels.These panels were extremely helpful in informing the design and implementation possibilites of our project. Many experts in synthetic biology but also social sciences kindly agreed to attend our panel meetings and advise us on the human practices aspects of the project.

The first panel consisted of Prof Richard Kitney, Dr Tom Ellis, Dr Guy-Bart Stan, Charlotte Jarvis and Kirsten Jensen. The panel addressed many different questions that we later used to inform our design.

Could the bacteria impact the germination of the seeds?
The coat itself would not be prohibiting germination. It is possible to design the coat sufficiently well to ensure that this would not happen. In addition, seeds normally germinate in soil full of bacteria that do not prevent germination.

How can we ensure that the auxin does not kill the plants?
We will be able to vary the inoculum of bacteria in the coat. We will get an experimental estimate of the auxin production, which will help us estimate the ideal number of bacteria to be contained in the seed coat. While a weak promoter may be better for constitutive expression of auxin, it will be easier to weaken the promoter later. We have used an insulator sequence to separate the promoter from the RBS so that we will be able to replace the promoter very easily. This may also contribute to the fine-tuning of expression and thus help us make sure that auxin is expressed at ideal concentrations. The worst case scenario consists of the auxin producing genes being transferred to other bacteria that become pathogenic. However, this could be tested exclusively beforehand and the infrastructure for this separate development and safety testing stage is already in place. In addition, unlike synthetic auxin, natural auxins such as IAA have a short half-life and degrade rapidly.

What is the risk-benefit relationship of our implementation?
In our implementation, we are trying to improve already existing practices. We do have to take a certain risk to combat desertification. However, is putting GM bacteria into soil worth speeding up the acacia tree planting process? How much does this really help? While GM bacteria may pose a risk, introducing foreign plant species that also show drought resistance and grow faster than acacia trees can be extremely risky and introduction of foreign species into ecosystems has already had negative consequences all over the world. This effect is likely to be worsened by the fact that we would be introducing the foreign species into an already damaged ecosystem. We may also be able to plant other fast growing plants at the same time as planting our coated seeds to hold the soil down while the seeds are growing.

Should we be using B. subtilis or E. coli as our chassis?
B. subtilis spores spread very easily over long distances and may thus be blown into different ecosystems where they may have negative effects. On the other hand, its spores would be easier to integrate into a seed coat. E. coli is not as easy to integrate into the seed coat. However, it does not form spores and is therefore very likely to stay inside the ecosystem we introduce the microbes into. We have already shown that E. coli is able to survive in non-sterile soil for more than two weeks and that it can pass on its plasmid to other bacteria, enabling them to express GFP and antibiotic resistance. By using E. coli as our chassis, we can be sure that the bacteria will survive in the soil for a reasonably long period of time but not spread as rapidly as B. subtilis spores would. At the same time, we will be preventing plasmid transfer using the BacTrap. We should be able to overcome the technical challenge of putting E. coli into the seed coat.

Would we be able to get rid of the bacteria once they are in the soil?
The kill switch is never 100% effective and the bacteria will lose the plasmid. In addition, bacteria killed by kill switches still leave behind DNA that can be conjugated by other, naturally occurring bacteria. We may not be able to take the bacteria back out of the environment after they have been distributed into soil. Instead, we will be aiming to prevent spread of the plasmid. We will be using E. coli as our chassis, which should be outcompeted in the soil and our BacTrap device will be used to prevent plasmid conjugation.

Second panel

We held a second panel two weeks after the first one. Several people from diverse backgrounds took part in our panel. Dr Stephan Güttinger and Alex Hamilton from the LSE BIOS centre, as well as Charlotte Jarvis from the Royal College of Art and Dr Janet Cotter, a scientific advisor for Greenpeace attended the panel. In addition, Prof Paul Freemont, Dr Geoff Baldwin, Dr Tom Ellis, and Dr Guy-Bart Stan from the Synthetic Biology centre at Imperial joined the panel.

Interview with Dr Alexandru Milcu

Dr Alexandru Milcu is an expert in above-below ground interactions who works at Imperial’s Silwood Park campus.He is a research associate and leads the research conducted in the Ecotron facility. He kindly agreed to meet Nick and me at Silwood to discuss the ecological implications of our project. We discussed some of our ideas about inoculating the bacteria into the soil and re-establishing vegetation in areas affected by desertification. He gave us a lot of very valuable advice and pointed out strenghts and weaknesses of the project.

Dr Alex Milcu demonstrating the Ecotron experimental set up (picture courtesy of Dr Milcu).

We discussed our idea of using a capsule to place our bacteria into the soil. Dr Milcu thought this idea was feasible and suggested making the capsule water soluble as one of the main limiting factor for bacterial growth is soil humidity. Another option we discussed was to inoculate the bacteria after rainfall.

An issue with dispersing bacteria in the soil is that they may be eaten by protozoa that naturally occur in the soil. Most bacteria are probably digested in this way. Grazing by protozoa has been shown to to increase plant growth. There are competing theories that this may be due to promoting auxin-producing bacteria or by increasing the nitrogen in the soil. However, it is becoming more likely that the latter theory is true. To test the impact of protozoan grazing on our bacteria, Dr Milcu suggested ensuring that protozoa be in test compost with our bacteria and then measuring auxin levels. This adds to the complexity of trials but gives a more realistic measure of density and auxin secretion than by simply plating out our bacteria on agar. Realistic conditions are also important for plant experiments.

Dr Milcu thought it was a good idea to plant trees at the same time as putting our bacteria into the soil. He also confirmed our theory that the project would not be very helpful with agriculture, especially because agricultural plants would be uprooted again. In addition, heightened root growth may lead to decreased growth above ground, which would be problematic for crops but should not have a lot of negative consequences for trees. He recommended using a mixture of different plants to re-establish a diverse ecosystem. However, this may be problematic if we would later skew the population if auxin is proven to differentiallty affect different plant functional groups.

Altering the plant community compositon a problem on its own. We do not want to further endanger threatened species or negatively affect diversity. However, it can be argued that we would be targeting disturbed areas with very little naturally occurring diversity. Nevertheless, we would have to ensure containment because we would not want the bacteria to spread to other, more diversely populated areas and alter native plant communities. While our kill switch would be preventing spread, it would not ensure complete containment. To address this issue, we are planning to model the spread of bacteria.

In addition, auxin concentration is a big concern. It should not give the bacteria an invasive advantage or act as a herbicide (although auxin as a herbicide is normally sprayed on leaves). It is therefore adamant that we not only model auxin secretion by our bacteria but also test it.

To test the effects of heightened auxin on different kinds of plants, Dr Milcu suggested setting up lab experiments with different plant functional groups. These would be planted in compost with an increased auxin concentration (using synthetic auxin) and observe how they respond. This may give us a better idea of how different plant functional groups are affected by auxin.

An issue we need to consider is that E. coli may not be able to survive in soil. This issue is complicated as we want the bacteria to have a lasting beneficial effect on the plant population by secreting auxin. However, at the same time, we do not want the bacteria to spread to other areas and affect diversity or crop yield. Dr Milcu thinks it is a sensible idea to use naturally dominant beneficial soil bacteria to ensure that they will last in the soil. However, these bacteria would not be outcompeted and may therefore be more likely to spread to other areas.

While we will be aiming for uptake of the bacteria into plant roots, Dr Milcu pointed out that auxin will mainly act in the rhizosphere and the bacteria will not have to invade the roots for plant growth to be stimulated. Dr Milcu thought the idea of controlling root morphology by controlling the distance of bacteria from the root is an interesting concept but unlikely to be feasible.

It is currenlty not known how auxin affects soil fauna. Auxin should not affect fauna dramatically. However, invertebrates with cutaneous respiration such as earthworms may be negativelyaffected. Earthworms are very important for soil structure and are often described as “ecosystem engineers“. While the areas that we will be targeting may not have earthworms, we do not want to negatively affect these important soil engineers. Below-ground invertebrates are responsible for nutrient cycling. This may be slowed down if the organisms are negatively affected. To test the effect of auxin on earthworms, Dr Milcu suggested keeping a couple of earthworm species in compost with a heightened auxin concentration, ideally equal to that which we are expecting our bacteria to produce, and observe the effect of auxin on the worms.

Another thing we will need to consider is the impact of our project on the carbon budget. Increased root biomass leads to more carbon storage so that we will probably create a bigger carbon sink. However, we will need to do more research into this.

Dr Milcu also suggested to conduct split root experiments to observe the effect of auxin treatment. In this type of experiment, the root of a plant is split in two and two different treatments are applied to both parts of the root.

In summary, Dr Milcu endorsed our idea of putting plants into the ground alongside our bacteria. He did not think there would be a lot of negative effects on soil fauna and that our project may positively affect the carbon budget. We will need to weigh out if we want the bacteria to persist in the soil against containment issues associated with this. The three big concerns Dr Milcu named were containment, the effect of auxin on soil invertebrates and changing the compositon of plant communities. We will be addressing these issues as well as possible by conducting further research and experiments.

Correspondence with Dr Robert Griffiths