WHEN 140 molecular biologists met at the Asilomar Conference Center in Pacific Grove, Calif., in late February 1975, they faced a startling fact. They realized that ``new [research] techniques, which permit recombination of genetic information from very different organisms, place us in an area of biology with many unknowns,'' to quote the conference statement. A new era had opened in which, for the first time, biological engineers could directly manipulate the genetic instructions that determine the nature and development of living organisms.
The unknowns included questions of environmental hazards, ethical limits to such work, and massive scientific puzzles to be solved as biologists advanced from the first primitive experiments. Nevertheless, the assembled experts knew that the discovery and use of a class of chemicals called restriction enzymes marked a watershed in their science and in its relation to society.
Biologists use these enzymes to snip a gene or complex of genes out of one organism and paste it into that of another. They operate on the level of molecules within a living cell.
Genetic information is carried by the complex, helically coiled molecules of DNA (DeoxyriboNucleic Acid). It is encoded in sequences of chemical units called nucleotide bases, which run along the molecule. There are four of these bases -- adenine, guanine, thymine, and cytosine (A,G,T,C). Taken together in groups of three, they specify the various amino acids that are strung together to make proteins, which govern the structure and function of organisms. This is the genetic code, the chemical blueprint for virtually every living organism on Earth.
Restriction enzymes precisely cut the nucleotide sequence that represents a single gene or gene complex (the smallest unit of meaningful genetic information) out of a strand of DNA. Genetic engineers also use them to snip out a piece from a tiny circular strand of DNA called a plasmid, which exists in a bacterial cell. This leaves the plasmid with two so-called sticky ends. The foreign gene can be inserted into the gap.
Since plasmids normally are passed from bacterium to bacterium, this is a way to get a foreign gene inside a bacterium. Once inside a bacterium, the altered plasmid replicates normally, reproducing copies of the foreign gene in the process.
Genes can also be inserted into viruses, which carry them into animal cells, incorporate them into the host cell's DNA, and give animals new genetic characteristics.
Plant biologists are handicapped in that they have fewer vectors, or carriers, to work with than do scientists working directly with bacteria or animal cells. One of the most useful carriers so far has been what is called the Ti plasmid from the bacterium Agrobacterium tumefaciens, which causes crown gall. It has successfully carried foreign genes into petunia cells. But it won't work with such important grasses as corn, wheat, and rice.
Also, together with scientists working with animal cells, plant biologists lack basic knowledge on how to control genes -- how to turn them on and off. They don't understand how different genes work together to produce multigene effects, which account for most characteristics of interest in plants and animals.
Genetic engineering promises many benefits, as outlined elsewhere on this page. But its practitioners first must do much basic research to gain the depth of knowledge and range of biochemical techniques they need to fully develop this new field's practical potential.
IN deciding how and to what extent the government should regulate genetic engineering, the United States is at a crossroads. The system used until recently, in which the National Institutes of Health (NIH) sets and administers safety guidelines, is suitable only for the research laboratory. Now other agencies, such as the Environmental Protection Agency (EPA), the Food and Drug Administration (FDA), and the Department of Agriculture (USDA), are beginning to assert their authority as genetically engineered products move from the laboratory into field testing and commercial production.
The challenge is to assure the public of adequate environmental safeguards without stifling the burgeoning growth of the new biotechnology industry.
The NIH's guidelines are developed and administered by its Recombinant DNA Advisory Committee (RAC). This committee has clear authority only over federally sponsored research, which in practice means most basic genetic engineering research. Many commercial companies have also voluntarily submitted their research plans for RAC approval.
This essentially advisory committee within what is basically a research-oriented agency can't regulate the manufacture, field testing, and marketing of genetically engineered products. That now is beginning to be done by the EPA under its authority to regulate chemicals, by the FDA when medicines are involved, and by the USDA under its authority to regulate the transport and introduction of alien organisms.
Within the Reagan administration, a Cabinet Council Working Group on Biotechnology has reviewed regulatory needs. It found that such agencies have sufficient authority under existing law to shepherd the new industry and safeguard the public interest.
To ensure coordination, resolve jurisdictional conflicts, and set risk-assessment standards, the working group proposed setting up two overlord groups. One would oversee scientific assessment boards in each agency, which would include the existing NIH, RAC, and a board within the National Science Foundation. Made up of individuals from each of the agency boards, this board would provide a core of scientific expertise to promote a consistent assessment of risk. A second and distinct interagency coordinating committee would oversee risk management and regulation.
The task of assessing the hazards of genetic engineering can be daunting. Risk assessment is not merely a technical exercise, it ``can be very policy dominated,'' says University of Texas Law Prof. Thomas O. McGarity. ``We do not have consensus [among experts]. And when we don't have consensus, then the choice among the [risk assessment] models is going to be determined by who makes that choice,'' he explains.
Thus the way experts go about the scientific task of assessing biotechnology risks is inherently a political matter involving the perceptions and biases of the investigators.
This is why there is no easy answer to safety questions in working with genetically engineered organisms. Even though most experts see little hazard in what is being done or proposed now, there is enough uncertainty for one expert to challenge the reassurances of another.
A decade ago, the concern was to avoid creating and releasing what the press called ``test tube monsters'' -- novel microbes that might be an environmental threat.
Now, with uniform safety guidelines administered by the NIH (and widely adopted by other nations) and with early fears having proved to be greatly exaggerated, that concern has faded.
Risk analysts today are focusing on factories that use genetically tailored microbes to produce chemicals and on the deliberate release of genetically engineered organisms such as bacteria on corn plants in field tests.
Professor McGarity speaks for many biotechnological experts when he says that factory use of the organisms poses no new or especially tricky safety problem. The use of microbes in fermentors to produce chemicals is well established, as are the means to assess and cope with its risks. New products, such as human insulin, produced in this way can be evaluated for safety just as are other new drugs and chemicals. Thus any uncertainties due to genetic tinkering with the microbes should be relatively easily handled.
But large-scale release of novel organisms is a new challenge for risk assessment. And nobody is sure how to go about it.
Cornell University microbiologist Martin Alexander warns that the ecological consequences of introducing novel organisms into the environment is unpredictable at this point.
Moreover, experts in the new biotechnology usually are narrowly specialized. They lack the breadth of knowledge to deal with ecological questions. Dr. Alexander explained his reservations during a symposium held last February at the National Academy of Sciences and in the current issue of the academy journal Issues in Science and Technology.
Even when a new species invades a habitat naturally, he explains in Issues, ``ecologists are unable to predict which introduced species will become established and which will not, and it is often not possible to explain successes or failure after the fact.''
This ignorance dictates caution in field testing or using genetically engineered organisms even when the experimenters themselves say they are convinced there is no risk.
In fact, Alexander notes, ``The scientist in a given field or the industrial manager hoping to exploit that field is often in no better position to evaluate the consequences of the endeavor than a hen to comment on the edibility of her egg.'' This is why he and McGarity have stressed the need for extensive research to build a better scientific basis for ecological prediction.
Thus, the US is still feeling its way in regulating what promises, within a decade, to be a major new industry. Whether Congress will agree that existing laws are adequate for now or will enact new ones is unknown.
Meanwhile, the courts are beginning to define environmental responsibility for the regulators. Last May, Federal District Judge John Sirica enjoined the NIH from approving any field tests of genetically engineered organisms.
In a suit brought by activist Jeremy Rifkin, Judge Sirica halted a University of California field test of bacteria engineered to prevent frost on crops. He agreed with Mr. Rifkin that the NIH had not made an adequate environmental impact study of the proposed test, in which altered bacteria would be sprayed on a 200-foot row of potatoes.
On Feb. 27, the US Court of Appeals for the District of Columbia unanimously upheld the specific ban on the ``ice-minus'' bacteria test until ``an appropriate environmental assessment is completed.'' But it told the NIH it could begin approving other field tests, again with ``appropriate environmental consideration.''
The three-judge panel commented that the NIH, and by implication all regulatory agencies concerned with this new technology, ``should give greater consideration to the broad environmental issues attendant upon deliberate release of organisms containing recombinant DNA.''
THE proliferation of genetic engineering companies is a hallmark of the potential payoff from recombinant DNA technology. The Industrial Biotechnology Association estimates there are 200 or more small companies trying to exploit the techniques. Major drug and chemical companies have also gotten into the act.
Industrial investment in the new techniques exceeded $1 billion two years ago, according to a congressional study. Market projections for this new industry's products anticipate annual sales volumes of $1 billion by 1990 and $10 billion or higher by 2000.
The benefits of genetic engineering are potential and actual. In the latter category are a handful of vaccines for veterinary medicine and, what may currently be the star product, human insulin. These are the forerunners of many useful chemicals to be made by microbes that have been given the genetic equipment -- including human genes -- to make them. Experts envision altering microbes to detoxify chemical wastes and clean up oil spills.
Many experts predict that an agricultural revolution, based partly on genetic engineering, may soon be on the horizon. One of the most publicized prospects is that of giving grains, such as wheat and maize, the ability to fix their own nitrogen fertilizer from the air.
But there is still much to learn. Ernest G. Jaworski, director of Biological Sciences for the Monsanto Company, explained during a recent National Academy of Sciences (NAS) symposium that geneticists know very little about how genes work in plants, let alone how to manipulate them. He and other biologists emphasize that, for some years to come, genetic engineering will be a tool used to unravel these mysteries. Its practical fruits on the farm are a long-term development.
This also is true of using genetic engineering to replace damaged or missing genes in humans. W. French Anderson of the National Heart, Lung, and Blood Institute told the NAS meeting that only one limited form of such therapy shows even a glimmer of practicality at this time. It involves possible therapy with somatic cells -- body cells other than germ cells -- to correct genetic deficiences, especially in bone marrow.
But Dr. Anderson said that the more exotic possibility of redesigning humans by tinkering with the genetic makeup of germ cells will be technically infeasible for an indefinite time, as well as being beyond what he considers the ethical pale. A number of experiments have shown that new genes can be incorporated into germ cells of animals such as mice. They then are inherited by the animals' offspring. But the uncertainties of doing this in a practical and controlled way with mice, let alone humans, are so great -- the genetic engineer's ignorance is so large -- that creating a brave new world of ``improved'' humans is pure science fiction. ``In terms of things that can't be done, that is one of the things which can't be done,'' Anderson said. -- 30 --