Environment

One may think that engineered microorganisms pose little threat, since, after all, environmental problems caused by them are unheard of after decades of recombinant DNA research. Fact is, the risks that synthetic organisms and their release in the environment pose are still not well understood and are difficult to predict (1). Could they spread without control, out-competing native species? Could they interfere with normal ecosystem functions in other, more subtle ways? Our team has considered these questions since the beginning of our project.

A constraint in the use of biocontrol agents (BCAs) is the possibility that it will persist in the environment and spread, perhaps affecting non-targets. It should be said that, unfortunately, literature on the transport of microorganisms in field conditions is scarce. Measures that prevent dissemination from happening are desirable, and a possibility to achieve this is the use of a induced lethality mechanism, also called kill switch, which would cause the death of the microorganism once it is not useful anymore. Our team has been working on a kill switch to overcome this barrier. It must be considered, however, that this kind of safeguard is not entirely fail-proof: spontaneous mutations may inactivate it. Cells with a deactivated kill switch could have a selective advantage and end up making the bulk of population (5).

One of the areas of concern when introducing genetically modified microorganisms is gene transfer. Self-destructing organisms still pose a problem in this regard, since their genes may still be taken by other organisms. The three known horizontal gene transfer (HGT) mechanisms, (conjugation, transduction and transformation) occur in the soil, and are known to be dependent on the characteristics of the soil (3).
It must be noted that soil poses physical and nutritional barriers to gene transfer. The structure of the soil hampers the mixing and movement of microorganisms, phages and DNA (which is continuously released into the environment through cell death), which need to get in contact for these mechanisms to happen, and oligotrophy also prevents the expression of systems involved in gene transfer. However, the rhizosphere might be a hot spot for gene transfer. Plant roots are the major site of input of carbon into soil, and rhizosphere bacteria show enhanced growth and activity due its availability. Furthermore, roots induce water flow in soil, which may enhance bacterial chemotaxis towards them. All this would result in more bacteria-bacteria, bacteria-phage and bacteria-DNA interactions (3). However, the role of roots in these processes remains unclear.
Some authors have considered that the frequency of these events is low enough to be negligible (3), although there are many technical difficulties in studying gene transfer in the soil (for example, due to inability to culture many of the bacterial species living in it). Furthermore, cells seems to be unable to be transformed with DNA that has been in the environment for a period of time that can range from hours to days (6). Adding a system that prevents such occurrences, rare as they might be, is still a sensible choice to prevent gene transfer as much as possible.
For this reason, the team decided to implement a toxin-antitoxin interdependent plasmid system to prevent HGT. These constructs are, however, not a complete reassurance, as mutations that inactivate one of the toxins would allow for gene transfer (5).

Plant-growth promoting rhizobacteria, such as our chassis, Pseudomonas putida, have not been observed to cause damaging effects to the environment. However, it might happen that our synthetic bacteria has a decreased fitness in the soil environment compared to the wild-type, because of, for example, a greater metabolic load due to the expression of our constructs. It is known that even wild-type pseudomonads lose ecological competence in the field after being cultured in vitro (4). In fact, most literature on the ecological impact of bacterial BCAs seems to focus on low competence and how to improve it (2,4,8). Because of this, a scenario where a BCA overcomes native populations in the rhizosphere seems unlikely.
This does not preclude it from having other, more subtle effects. For example, an unforeseen effect of introducing a bacterial BCA can be the addition of food resources for native consumers; it is not unusual to observe predation upon BCAs. This is not merely an interference with the biocontrol, but is actually an interference with food webs. Many organisms are food limited, and so-called “food subsidies” can restructure food webs at various trophic levels, which has unpredictable and possibly important indirect effects in the ecosystem (7).
It is clear that, before our product could be used it the field, it would be necessary to extensively study its ecological effects, including its competence in the rhizosphere (root colonization and survival), its impact on microbial population (specially fungi, since it produces antifungal compounds) and soil functions or its possible dissemination to ground water.

1. Four steps to avoid a synthetic-biology disaster. Dana et al. Nature 483, 2012.
2. Significance of ecology in the development of biocontrol agents against soil‐borne plant pathogens, Deacon. Biocontrol Science and Technology, 1:1, 5-20.
3. Molecular Ecology of Rhizosphere Microorganisms: Biotechnology and the release of GMOs. Edited by F.O'Gara, D. N. Dowling, B. Boesten. VCH Verlagsgesellschaft mbH, 1994.
4. Multitrophic Interactions in Terrestrial Systems. Edited by A.C. Gange and V.K. Brown. Blackwell Science, 1997.
5. Preparing synthetic biology for the world. Moe-Behrens et al.Frontiers in Microbiology. Vol 4, article 5. 2013.
6. Monitoring and modeling horizontal gene transfer. Nielsen et al. Nature Biotechnology 22, 1110–111, 2004.
7. Indirect effects of host-specific biological control agent,. Pearson et al. TRENDS in Ecology and Evolution. Vol.18 No.9 September 2003
8. Biological control of soilborne plant pathogens in the rhizosphere with bacteria, Weller. Annual Reviews of Phytopathology, 1988. 26:379-407.