Survival of genetically engineered microorganisms in the environment and a test of the ability to predict survival of Pseudomonas chlororaphis 3732RN-L11 in soils across Canada.

摘要

The survival of a genetically engineered microorganism (GEM) is an important consideration in assessing potential risks posed by an environmental release. A scientific literature review identified 102 records that contained the results of 719 experiments comparing the survival of genetically engineered bacteria relative to their parental microorganisms. Experiments were predominantly short-term (e.g. 100 days) laboratory experiments simulating terrestrial agro-ecosystems in northern temperate climates. The survival of the GEMS was the same or lower than parents in 92% of experiments. GEMS showed a lower survival relative to parents in 36% of experiments, and an enhanced survival in 8% of experiments. The results reject categorical hypotheses that GEMS are less fit or more fit than their parental microorganisms.; Linear multiple regression models were developed to examine the ability to predict survival rates of Pseudomonas chlororaphis 3732RN-L11 in bulk soil and wheat rhizosphere. Models were developed from GEM survival data generated in the laboratory using intact soil core microcosms and physical, chemical, and biological data on twenty-three soils measured at the time of GEM inoculation. The models were tested by comparing predicted GEM survival rates with observed GEM survival rates from environmental releases at five field sites across Canada over two years. No single soil parameter like % clay proved very useful for predicting GEM survival in soil or wheat rhizosphere. The best multiple regression model for soil selected fungi colony forming units (CFUs), % soil titanium (positive correlations) and % soil magnesium (negative correlation) for predicting GEM survival slope. The best model for wheat rhizosphere selected soil bacteria CFUs, soil nitrate, % soil potassium (positive correlations) and soil magnesium (negative correlation) for predicting GEM survival slope. While the models were statistically significant, they were not able to reliably predict the survival rate of the GEM in field trial releases. The results demonstrate there can be considerable uncertainty associated with predicting GEM survival for widespread release into soils.