Spread of organisms with novel genotypes: thoughts from an ecological perspective.

One category of objection to the release of organisms produced by genetic engineering is based on the fear that such organisms may persist in the environment and damage existing ecosystems. An assessment of environmental risk thus involves an ecological question analogous to the introduction of exotic species which has been known to produce serious ecological disruptions. An investigation of the literature on exotic introductions reveals, however, that foreign species do not invariably produce adverse changes. Ecologists believe that only a fraction of immigrating species actually produces ecological dislocation while the majority probably fail to penetrate existing biotic assemblages. Stressed or simplified environments are, however, more vulnerable to successful invasion. Unfortunately, because very little information has ever been collected to document the number or causes of failed introductions, it is impossible to quantify the probability that any introduced species will or will not cause serious disturbance purely on the basis of historical evidence. The development and spread of genotypes that confer resistance to chemical control agents in insects and microorganisms is also analogous to genetic engineering in that human activity contributes to the spread of new genotypes. In both groups of organisms, resistant genotypes can come to predominate in even geographically widespread populations with great rapidity. Resistance to pesticides in insects is usually found to be determined by single genes. In bacteria, antibiotic resistance genes are usually, if not always, associated with the extrachromosomal genetic elements known as plasmids. Bacteria seem to be able to transmit plasmid-borne genes between species and genera with facility. The ease with which new genes can be inserted into bacteria via plasmid vectors in recombinant technology is thus a two-edged sword. It may be very difficult to keep inserted genes isolated in single bacterial strains. The evaluation of the literature on which this report is based suggests that an ecological approach for risk assessment is appropriate. Microorganisms, for which genetic engineering is of most immediate importance, exhibit the same ecological properties as higher organisms. The proportion of an organism's genome which is novel has no direct correlation with the magnitude of impact such a change may have in economic, medical, or ecological terms. Meaningful probabilities for persistence of engineered organisms in the environment will have to be generated by experiment, probably with model microbial ecosystems.