Third-Generation Biotechnology: A First Look

In May 2008, the U.S. Department of Agriculture (USDA) published a draft environmental impact statement for the use of genetically engineered fruit flies and pink bollworms in plant pest control programs. According to the statement, the transgenic insects “carry a heritable sterility gene resulting in only male and no female offspring in the field.” As a result, the statement concluded, “The wild population would soon collapse because of the elimination of female flies.”

As a general pest control technique, the mass rearing and release of sterile males to mate with wild females is by no means new. But every significant program that applies this technique today relies on radiation to sterilize the insects. A problem is that radiation compromises mating competitiveness and makes released insects less effective in the field. As a result, more insects must be released, with additional costs. Genetic engineering, on the other hand, minimizes these kinds of problems. Bollworms and fruit flies genetically engineered for sterility are much heartier, sexually active, and competitive than those sterilized by radiation. In recognition of these advantages, the USDA’s draft statement recommends the “integration of genetically engineered insects” into pest control programs.

Genetic engineering now promises a host of new approaches to pest control. Available biotechnology can produce genetically manipulated organisms (GMOs) that are able to protect agriculture by infecting plant pests, promote public health by eliminating disease organisms and their vectors, and preserve the natural environment itself by controlling “invasive” species. The regulatory agencies, however, have yet to develop a response to these emerging pest-control technologies.

Inconsistent regulations

Starting in the 1980s, biotechnologists have produced a variety of genetically modified crops; these plants represent the first generation of GMOs in agriculture. The USDA has approved over 70 genetically modified products in about 15 crop plants. This was not easy; disputes have raged over the possible environmental and health effects of “Frankenfoods.” But genetically modified maize, cotton, soybean, and other crops now dominate agriculture with few ill effects, and concerns about genetically modified crops have largely abated, at least in the United States.

In 2002, in order to accommodate a public initially suspicious of genetically modified crops, the USDA created the Biotechnology Regulatory Service, which has imposed on plants produced by genetic technology stricter and more expensive field tests and other requirements than those that apply or would apply to the same plants produced by conventional breeding. In contrast, the U.S. Food and Drug Administration uses identical criteria to test the safety of food products regardless of the process that created them. Likewise, the Environmental Protection Agency (EPA), which regulates pesticides, applies essentially the same requirements to GMOs as it does to naturally occurring substances.

In the broader picture, no federal legislation deals specifically with the regulation of biotechnology. Agencies rely on a regulatory framework (jury-rigged in 1986 from statutes, some decades old, that did not contemplate genetic engineering) that many commentators believe to be ill-suited to the task.

The adequacy of the current legal framework will be sorely tested by second- and third-generation recombinant products. Second-generation GMOs in agriculture include “functional” plants designed to produce pharmaceuticals, fuels, and industrial compounds. Third-generation GMOs include recombinant organisms engineered to control pests in agriculture, pathogens in human health, and invasive species in the environment.

The absence of a coherent regulatory framework has not deterred the scientific community, which has adopted a “synthesize first-regulate later” strategy. In laboratories and greenhouses, biologists in many countries are experimenting with fungi, viruses, bacteria, nematodes, insects, and other organisms genetically modified to express toxins, to mass produce sterile males, to spread disease among some animals and prevent disease among others, and to be useful in other ways in programs of integrated pest management. Technologists have designed scores of GMOs to eliminate pathogens and pests, vaccinate against them and mitigate their effects, and even replace them in nature with more benign genotypes. The current hodge-podge regulatory approach will be insufficient to address public concerns and still realize the potential of a new generation of genetically engineered pesticidal organisms.

Advanced pest management

As pests have evolved to develop resistance to chemical pesticides and the public has grown concerned about the threats these pesticides pose to public health and the environment, entomologists and other biologists have emphasized ways to integrate novel biological controls into programs of pest management. For example, biotechnologists are working to replace current chemicals with new bioinsecticides that will destroy pests by exposing them to fungal infections. Fungi tend to be host-specific, can be mass-produced on inexpensive substrates, and are thought to be harmless to users and to the environment. But naturally occurring fungi tend to kill insects slowly. Genetic technology holds the promise of producing bioinsecticides based on hypervirulent insect-specific fungi that kill quickly. Moreover, according to one expert, “Genetically engineered pathogens of pest insects might be designed which bring together toxins, virulence, or other pathogen functions from organisms which cannot be mated conventionally.”

As proof of this concept, Raymond St. Leger, an entomologist at the University of Maryland, College Park, has inserted a scorpion toxin gene into a fungus that infects mosquitoes. According to St. Leger, fungi “land on the insect’s outer surface, insert little tubes called hyphae, and grow within the insect…. If you can get the fungus to insert a toxin into the insect, you can kill the insect very quickly. This is what we did. We’re trying to get a supercharged, hypervirulent fungus that will take out the mosquitoes quickly.” The fungus then replicates on the mosquito cadaver and thus will multiply through the environment.

The creation of hypervirulence in pathogens that attack pests and then replicate might be worrisome because of possible effects on nontarget species. For this reason, among others, technologists have been fascinated for decades by the opposite strategy: that is, the prospect of developing hypovirulent—that is, relatively benign—strains of fungal blights to replace more damaging forms in nature. In 1992, two microbiologists reported in the journal Science that they had demonstrated the “feasibility of creating hypovirulent fungal strains.” Specifically, they had created a transgenic hypovirulent version of the blight that had felled the American chestnut tree. This transgenic fungus works to protect vaccinated trees, but producing a hypovirulent chestnut blight that out-competes the wild type and readily spreads through forests remains a challenge.

The application of hypovirulent fungi as pesticides shows promise in agriculture. To cite one example, a strain of Aspergillus flavus fungus produces a micotoxin called aflatoxin, a powerful carcinogen that often contaminates seeds and nuts. Scientists discovered many years ago that it is possible to generate an innocuous variant of the fungus—one that differs from the common type because it does not contain the toxin-expressing gene—and more recently they have learned how to mass-produce that variant. When sprayed on plants, the mutant fungus out-competes the toxic version. The hypovirulent spray is now used extensively in the Southwest and has reduced aflatoxin levels in several crops.

The success of this hypovirulent fungus suggests the possibility of deleting genes from other blights, wilts, rusts, and similar troublesome organisms and then mass producing transgenic varieties that when sprayed on plants will replace the deleterious types. The aflatoxin-free fungus may have obtained regulatory approval, however, in part because it can be found in nature. The regulatory hurdles that the USDA imposes on GMOs precisely because they are GMOs could discourage attempts to use gene deletion to produce hypovirulent pathogen variants that would easily gain approval if they could be found even in small quantities in the natural environment. Only the process would differ, not the product.

In another approach, biotechnologists are studying baculoviruses, a large variety of viruses composed of double strands of DNA that act specifically on hundreds of arthropods, including many agricultural pests, but appear to be safe for plants and vertebrates. Two naturally occurring baculoviruses (AgMNPV and HaSNPV)) are now used extensively to protect soybeans in Brazil and cotton in China. But because baculoviruses typically kill much more slowly than do chemical sprays, they have not yet been widely substituted for them. In experiments over the past decade, biotechnologists have increased the killing efficiency of baculoviruses by splicing into them toxin-expressing genes isolated from mites, scorpions, and spiders. A review of the potential use of recombinant baculoviruses in pesticides cautions, however, that “the reluctant attitude of European (but also some non-European) societies to genetically engineered products hampered their introduction or even promoted press and other mass media attacks against field trials.”

Pesticides for your health

Genetic technologies under development to control pests and pathogens in agriculture also may be deployed against diseases and their vectors in programs intended to protect public health. The genetic technology that the USDA has used in working with fruit flies and bollworms is being adapted for programs to combat mosquitoes. These strategies include a male-sterilization approach, a “natural”-enemies method using hypervirulent transgenic pathogens, and a strategy that deploys hypovirulent mosquitoes designed to be disease resistant.

In 1991, the World Health Organization published a proposal to engineer a hypovirulent mosquito that would resist malaria and other diseases. This transgenic “safe” mosquito would either out-compete disease-carrying mosquitoes or transfer genes conveying disease resistance into wild populations. At the level of science and technology, researchers have made a lot of progress in genetically constructing a malariaproof mosquito, and a prototype insect developed in 2005 is being readied for field trials. Many commentators believe, however, that the political obstacles are greater than the technical ones. Will developing nations, where malaria is rife, accept GMOs in programs to either eliminate mosquitoes or to replace them with a disease-resistant type?

Here is where regulatory agencies in the United States may play an exemplary role. If the regulatory community treats products of biotechnology just as it treats products produced by any other means, other countries may be more willing to follow suit. But since the USDA’s Biotechnology Regulatory Service distinguishes GMOs from other organisms on a per se basis—that is, on the basis of the method by which they are produced—this may suggest to other countries that these organisms are intrinsically dangerous. The decision by a federal agency to treat GMOs as a suspect class has ethical, legal, and social consequences that should not be ignored. If GMOs are so intrinsically dangerous that they should be regulated on different terms than other organisms, then the regulatory agencies should be able to say why this is so.

Combating invasive species

In a report to the Fish and Wildlife Service, two prominent fisheries biologists, Anne. R. Kapuscinski and Timothy J. Patronski, have addressed “the feasibility of using genetic methods as a new approach for biological control of non-native fish.” They note that “transgenic techniques designed to produce sterile fish or spread deleterious transgenes to a target non-native target species” are still in the early stages of research, but “these techniques offer a variety of powerful approaches for reducing non-native populations.” This thought represents a view taken by many in the scientific community that non-native species “harm” the environment by competing with native species. Thus, the National Invasive Species Council defines any non-native species to be “invasive” and therefore a pest if it causes “environmental harm.” The council states, “We use environmental harm to mean biologically significant decreases in native species populations.”

Concern about “environmental harm,” which is distinguished from both economic harm and harm to human health, is particularly strong in Australia. For at least a decade, researchers there have attempted to control or eliminate rabbit populations, which are not native and are considered invasive on that continent, by developing more lethal transgenic variants of viruses that cause myxomatosis and other diseases in rabbits. At the same time, in Spain and other parts of Europe, where the same rabbit is welcome as native and therefore ecologically correct, biotechnologists are designing a genetically modified virus that will spread in the wild population to immunize the animal against these diseases.

The irony that the same viruses should be genetically engineered in different ways to eliminate a species in one place and to protect it in another has not been lost on observers who note that the same “waskally wabbit” is loved in one place and hated in another. An international study acknowledged that genetically modified biocontrol agents might cross national borders. The study raises the concern that a creature valued in one country could be harmed by “GMOs originally designed to eradicate or reduce that animal in its target country. Similarly, concern has been raised over the risks of a GMO designed to preserve a species in one country compromising pest-control programs in another.”

The possibility that researchers could genetically manipulate the natural enemies of non-native species to kill them faster and better, thus protecting the “natural” environment, attracted public notice in 2001, when a team of virologists, seeking to control non-native rodents in Australia, announced that it had had added an interleukin-4 gene, which suppresses the immune response, to ectromelia or mousepox virus. When infected with the recombinant virus, all the mice in a test population quickly died, including those inoculated against the disease. At the University of California, Berkeley, researchers showed that they could achieve the same quick and certain kill rate by engineering a hypervirulent strain of the disease organism that causes tuberculosis in mice. A team at St. Louis University has designed and constructed an even more virulent mousepox and cowpox virus. It seems possible to make virtually any disease organism lethal by attaching genetic instructions that circumvent the immune system of the target animal. This can serve as a powerful weapon against non-native species or, indeed, against any organism.

Questions for regulation

While offering these and other potential benefits, third-generation biotechnology also confounds the USDA and other agencies with at least three conundrums.

First, these agencies will have to revisit the relevance of the process-product distinction. Should regulatory agencies require the same kinds of data from all products proposed for use in pesticides? Should agencies instead treat GMOs as a suspect class? If so, why?

Second, the regulation of GMOs in agriculture has largely been seen as a nation-by-nation affair, since crops can be grown and sold in some places but not in others. But biopesticides present a different scenario. National borders may not constrain fungi that replicate in the cadavers of insects and spread, hypovirulent insects that drive desirable genes into wild populations, or hypervirulent poxes that transmit themselves between organisms. As a result of globalization, regulation may have to take on an international dimension.

Third, the possibility of creating a transgenic pesticide to kill nearly anything raises the question, “What is a pest?” In the past, regulatory agencies did not need to confront this question because the market answered it: Farmers and others who demanded the compounds decided which species to target on their own property. The production of a new generation of biopesticides will be driven not just by private consumers but also by public agencies. The pesticides will be applied not just on private property but in the environment at large. These pesticides will target human pathogens and their vectors, such as mosquitoes, as well as invasive, non-native, or otherwise ecologically incorrect species. It may present a conflict of interest that the same agency that commissions the pesticide identifies the pest.

No statute or ordinance defines, or could define, in scientific terms what constitutes a “weed” or “pest” species. For the purpose of regulating pesticides, the Federal Insecticide, Fungicide, and Rodenticide Act has since 1972 given the EPA the authority to declare to be a pest “any form of plant or animal life (other than man and other than bacteria, virus, and other micro-organisms on or in living man or other living animals) which is injurious to health or the environment.” Because the statute offers no guidance as to what may be deemed injurious to the environment, a determination whether a creature constitutes a pest represents not a scientific but a political judgment when it is not a consumer-or market-driven private one.

In moving ahead, the USDA should take a lesson from the National Institutes of Health (NIH) and its Ethical, Legal, and Social Implications (ELSI) research program, initiated in 1990 to help prepare the public for technologies that could be expected to result from the Human Genome Project. Because of nearly 20 years of a largely government-initiated democratic deliberative debate, the public has overcome the initial hysteria that surrounded the sequencing of the human genome. To a large extent, the work of the NIH ELSI program can be credited with the broad social understanding and acceptance of advances in genetic engineering in medicine.

In contrast, the USDA has systematically avoided sponsoring inquiry into the ethical, legal, and social implications of biotechnology. One reason may be that agency officials have associated these issues with demagogues who early on inveighed against genetic technologies. A belief that informed public debate about agricultural biotechnology is not possible—an inference that might have seemed reasonable at the time—may have become a self-fulfilling prophecy. At some point, the USDA will have to confront the significant ethical, legal, and social questions that third-generation biotechnology raises, at least to avoid the regulatory train wreck that affected first-generation agricultural biotech.

But there may be hope. Chris Wozniak, national program leader for food biotechnology and microbiology at USDA, has written that “those creating transgenic insects for pest management purposes will be wise to learn from the lessons already experienced by others in agricultural biotechnology…. We do not need to repeat those lessons!”