Microbial census hints at biotech hurdles.

Microbial Census Hints at Biotech Hurdles

In what may be the most detailed study yet of genetic variability within a single species of soil-dwelling bacteria, researchers report the microbes are remarkably diverse, and their diversity directly relates to relatively small differences in soil environments. The findings are relevant to current research seeking to assess the safety and usefulness of genetically engineered microorganisms released into the environment.

Last year saw the first controlled releases of gene-altered bacteria into agricultural test plots. Researchers designed the experiments to measure the survivability and genetic stability of laboratory-engineered strains when thrown into the dog-eat-dog world of natural soil-microbe rivalry. Early results indicate the organisms don't survive long enough to disrupt ecological checks and balances (SN: 11/5/88, p.300). But scientists should interpret such results cautiously, suggests Michael H. Smith of the University of Georgia's Savannah River Ecology Laboratory in Aiken, S.C.

Smith and his colleagues measured the genetic diversity of Pseudomonas cepacia in soil cores taken from 40 plots of land in four distinct ecosystems. They tallied a range of diversity -- as measured by differences in enzyme profiles -- a full 10 times greater than anything previously reported within a single species.

"When we talk about biological organisms like this, an order of magnitude difference is really a major statement," says Smith. Moreover, he and his colleagues found that microbial variants were highly specialized -- showing clear preferences for microhabitats characterized by slightly different levels of various minerals and nutrients. "Every single bacterium down there is unique, but each is unique in a way that matches very closely the differences in the environment," he told SCIENCE NEWS. The findings "indicate that the source of strains used in genetic engineering will greatly affect the outcome of planned releases in variable environments," the researchers write in the December PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES (Vol.85, No.24).

Smith says measures of a specific bacterium's survival in a few small plots may not predict its fate in different environs. Scientists may need to compile a large database describing an organism's behavior under a variety of conditions.

While agronomists have gathered some of those data from greenhouse studies, severe restrictions on the number and nature of outdoor experiments have slowed the process, says David M. Weller of the U.S. Department of Agriculture's Agricultural Research Service in Pullman, Wash., where several such field trials are underway. "We're going to have to gather information about bacterial performance, survivability and fate under a lot of different conditions," he agrees. "We're going to have to generate that information, and it's just going to take some time."

Smith suggests that highly specialized, laboratory strains of engineered bacteria may be doomed to short life spans -- except within the narrow confines of each bug's ideal environment. He suggests researchers may have to insert agriculturally beneficial genes into a wide range of bacterial subtypes rather than into single, highly inbred strains to ensure that at least some altered microbes with a given microhabitat preference can survive to perform their genetically assigned tasks.