Genetically-engineered microorganisms: I. Identification, classification, and strain history.

We have argued that accurate identification of the microorganism will form a cornerstone of the assessment of potential hazard. Appropriate methodology for identification exists, and is continually under development and refinement. Organizations such as the American Type Culture Collection will perform certified identifications for relatively low cost. Thus there appears to be little reason that an organism should not be identified insofar as current microbiology allows prior to submission for PMN review. We suggest that a complete microbiological characterization be considered an essential element of an acceptable PMN. To accomplish this, however, current institutional arrangements for the protection of trade secret information needed in the process of identification may need to be improved. An accurate identification of the strain will often provide access to important information with which to evaluate its ecology, pathogenicity, biochemistry, and genetics. Specialized texts, the scientific literature, and professional consultation are ready sources of such information. However, a major effort should be made to establish a data base that can specifically address the needs of biohazard evaluation. This could be done, in part, by collecting information about the construction, and about the behavior in the environment of genetically-engineered microorganisms that are now under development and will soon be tested or used. Identification information may also eventually be useful for the formulation of hypotheses about possible modes of harm or about relative safety, based on phylogenetic relationships. This is a very difficult undertaking at present, however. Microbial taxonomy is currently in a process of radical reevaluation as new macromolecular sequence information reveals previously unsuspected phylogenetic relationships, and disturbs categorizations based on older types of traits such as morphology, etc. This means that both inferences about relative safety and about possible modes of harm from taxonomic relationships must be highly tentative based on current information. Regulatory authorities may wish to consider requesting confirmatory DNA hybridization data or other macromolecular sequence comparisons in cases where strong arguments related to safety must be made from taxonomic information in relatively poorly studied groups of organisms. Detailed strain histories would provide valuable information for safety evaluations.(ABSTRACT TRUNCATED AT 400 WORDS)