Chapter 23
Seeds of Change: Applications of Biotechnology to Agriculture
Allen J. Dines and John Mottley
INTRODUCTION
Agricultural applications of biotechnology will eventually touch the lives of virtually everyone in some way. The agricultural sector, while only a subset of the entire universe of the industries biotechnology will influence, contains a full complement of the opportunities, issues and applications created by the technology.
For thousands of years farmers have sought to overcome the many uncertainties imposed by nature on crop and livestock production. More fundamentally, the search for improved plants and animals that produce more valuable products, that give higher yield and that are resistant to pests and environmental stress has been underway since the outset of agriculture.
The search has been extremely successful. The art and, more recently, the science of selective breeding has evolved into an extremely complex body of knowledge. If we think for a moment of genetic engineering broadly as the manipulation of genetic material for the purpose of attaining useful products, then traditional plant and animal breeding could be considered genetic engineering.
This form of traditional genetic engineering has clearly produced results. Breeders and farmers, through painstaking selection and crossing, have developed crop lines that produce larger yields of grains, vegetables and fruits; they have produced plants that survive in harsh environments; and plants that resist attack by disease and insect pests. Similarly, selective breeding of animals has produced livestock with the ability to withstand unfavourable environmental conditions while yielding substantially more meat, milk or fibre than their undomesticated predecessors.
But genetic engineering as we think of it today is different. The term is relatively new and, as generally used, it is not intended to include traditional plant and animal breeding. It refers to a variety of modern techniques of molecular biology ranging from recombinant DNA technology to cell fusion, somaclonal variation and monoclonal antibody technology. Thus, although selective or traditional breeding might be considered genetic engineering in a literal sense, it is in fact imprecise and crude relative to the techniques now emerging to improve agricultural production.
Given this history of accomplishment of "conventional" breeding, "modern" genetic engineering, for all the dramatic novelty the new molecular biology affords, has some distance to go in generating substantial benefit before it matches the contributions to agriculture already made by its conventional predecessor.
The early applications of the new genetic engineering to agriculture seem at first glance merely to offer an extension of techniques, which have already proven to be a vast success.
In this light the new technologies will contribute to further advances, many of which would never be possible using conventional technologies. These advances might generally be characterised as improved products for existing markets. They are not new products for new markets. They offer better solutions to problems that we have been working to solve for many years with conventional technology. For example, chemical pesticides now exist that afford control of most serious pests. Yet substantial improvements in control may be possible with genetically engineered plants that carry their own resistance internally. This would eliminate the adverse environmental effects of chemical pesticides while arguably giving the farmer better control because the threshold of application would not depend on the active participation of the farmer in deciding when to treat the fields.
But this is not the whole story. With the first commercially valuable agricultural products from this technology barely introduced to the marketplace, it would be short sighted to cast the potential contributions of the technology in so limited a light. In this chapter we will examine the likely contributions of biotechnology to agriculture in the broadest terms possible.
First, we will consider the products that have already been developed and commercialised or will come to the marketplace in the next few years. The majority, although not all, these products might be characterised as "incremental improvements'. They are new and improved solutions to problems faced by farmers and ranchers for many, perhaps thousands, of years.
But biotechnology is also leading to whole new approaches for attaining useful products from agricultural commodities. These applications of biotechnology could be classified as new opportunities rather than new solutions to old problems. These opportunities may not even address what we think of today as current problems. We will therefore, in as credible a manner as is possible at this early stage of the development of the technology, examine the longer range opportunity presented by this dramatically new science to agriculture. Here, by definition, the actual products are still quite vaguely defined. They are only in the earliest phases of development or are perhaps only a concept in the minds of researchers.
These near-term and long-term developments comprise the "content" of agricultural biotechnology. Before we move to consider them, we must first review some underlying fundamentals that set the stage and characterise the environment in which the new applications of this technology will emerge. This is the "context" of biotechnology. We will consider how the technology will apply to crop and animal science, where it will fit in the larger scheme of food and fibre production and how the companies developing the technology will relate to agribusiness. Also to be considered is the way these emerging products and companies relate to the public policy arena.
Taken together the content and the context of agricultural applications of biotechnology comprise a dynamic and exciting arena that will see many changes over the next decades. This chapter will provide a brief overview of how that change may take shape and what the contributing components will be.
CONTEXT OF AGRICULTURAL BIOTECHNOLOGY
This section relates the science and business concepts described in earlier chapters of this book to applications of biotechnology to agriculture. At the research level, agriculture will challenge scientists substantially to further refine the techniques of genetic engineering. Products for agriculture involve a broad spectrum of applications including genetically altered microbes, plants, and ultimately, animals.
Some agricultural biotechnology applications will include the use of genetic engineering technologies identical to those in use for production of human healthcare products. These technologies have been discussed in other chapters. Their application to agriculture includes the use of bacteria as living factories to produce new therapeutics for use in animal healthcare and new growth enhancement products to improve the efficiency of meat and milk production.
These production technologies, however, are relatively simple in comparison with the technologies required in applications involving the direct use of modified microbes in farm fields, the use of new varieties of plants developed through genetic engineering, and the use of new lines of improved animals in livestock agriculture.
In most cases the organisms developed for use in agriculture will not have the luxury of the optimised growing conditions of a fermenter in a contained production environment. Rather, in the case of crop agriculture, they compete in farm fields for space and nutrients. And, if they are indeed to be successful in offering a better solution to the farmer's problems than conventional products, they will do so more vigorously than their conventional competitors. These technologies and their role in the agricultural sector are discussed in the first section of this part of the chapter called "The Biotechnologies of Agriculture."
"The Agricultural Sector" is the second part of this section. Agriculture as a component of the economy is immense. The ultimate unit of production is the farm. The farmer is increasingly more than a sole proprietor of a family land holding. More likely, today's farmer is a small or sometimes large businessman who manages assets worth, in some cases, many millions of dollars. The farm sector is thought by many economists to be the closest representation in today's economy to pure competition. This is a sector where many farmers produce largely undifferentiated commodity products for many buyers under a set of uncertain environmental conditions. By virtue of the collective value of production of a nation's farms, the agricultural sector is generally among the top industries in national rankings. Not infrequently, it is the leading industry.
Agribusiness is generally defined as the collection of companies that provides the inputs necessary for farm production. Non-farm inputs include such items as fertilisers, animal healthcare and nutrition products, chemicals, and farm machinery. A major farm origin input that is supplied in large part by a group of specialised agribusiness companies is seed. Included among the principal suppliers to the farm sector are the world's major corporations. The bulk of biotechnology applications to agriculture will reach the farmer through this agribusiness industry. For example, agribusiness will offer products of biotechnology in the form of improved seeds, new products for animal health or new biological pesticides. The companies that comprise this agribusiness sector as well as the agrochemical industry and the animal health products industry are discussed as the third part of this section, called "Companies in Agricultural Biotechnology."
Agribusiness depends substantially on the economic viability of the farmer for the sale of its products. The farmer is in turn subject to world economic conditions, food commodity prices, farm technology, farm management and, perhaps most significantly, government policy issues. In the fourth and final part of this section, "Constraints to Growth," we identify some of the regulatory and public policy issues relevant to agriculture and biotechnology as they contribute to the continually shifting balance of the farm economy.
The Biotechnologies of Agriculture
The agricultural benefits of the new technologies of genetic engineering are widely acknowledged to be enormous. Scientists will develop products that will increase food production, reduce fertiliser use, and decrease the need for costly, and in some cases environmentally questionable pesticides. Fundamentally the new technologies are likely to be superior to conventional technologies because they are faster, more precise and they will afford greater flexibility in selection of genetic traits.
The task facing the scientist charged with the development of new useful products for agriculture is different from that facing the scientist wishing to apply biotechnology to pharmaceutical production. Biotechnology products for agriculture must compete in the environment, attain high yields comparable to conventionally-produced products and deliver the extra desirable trait that was the basis for development of the product in the first place.
Broadly, the technologies differ for crop agriculture and for animal agriculture. In the area of crops, the technologies in development include recombinant DNA technology and plant tissue culture technology. A third technology, that of conventional plant breeding, is critical to this area but as it has already been in widespread use it is referenced here only for comparative purposes.
The area of animal agriculture partially parallels biotechnology applications in human healthcare. This area includes the application of genetic engineering techniques to the production of substances directed at protecting animal health or directed at improving animal performance. The technologies involved in development of these "animal pharmaceuticals" and "animal vaccines" are essentially identical to those described in Chapter 22 of this book.
Beyond this, however, additional technologies are emerging that entail the alteration of the animal genome to attain improved products. This technology is less developed than its parallel in plant genetic engineering but its dramatic potential is currently being widely discussed.
Crop Agriculture-Recombinant DNA Technology
The applications of recombinant DNA technology to plants are not as well developed as similar techniques used to produce novel products based on incorporation of selected genes into bacteria and other single-celled organisms. A number of factors contribute to this.
First, plants are substantially more complex than bacteria. They are multicellular organisms with complex specialised functions capable of myriad responses to changing internal and external conditions.
Second, plants are poorly understood by scientists relative to the knowledge base applicable to bacteria and animals. Animal and bacterial systems have been the subject of extensive scientific study as part of the larger, decades-old effort directed at solving human disease problems.
Third, traits that influence key aspects of plant performance are complex at the molecular level and are controlled by many genes. While a specific protein coded by a single gene may provide a valuable component for a human drug, plant traits that offer a promising basis for yield improvement are often controlled by many, perhaps hundreds, of genes. As researchers learn more about plant molecular biology, yield improvement is increasingly believed to be mediated by a complex series of primary and secondary plant metabolic interactions.
Fourth, the modified plants, even when they produce a particular new protein, still must function as effectively in the field environment as their unmodified counterparts. In other words, the new plant must do everything its best predecessors did as well as exhibit the new trait added through genetic engineering.
Fundamentally, the task of adding new genes to plants is the same as adding new genes to microbes: the scientist must insert the new DNA into the plant cell in such a way as to get the new gene to express (that is produce the protein for which it codes). In practice there are more obstacles that must be overcome as the techniques available for bacterial genetic engineering often require substantial modification to be usable in plant systems. In some cases, they simply have not yet been successfully adapted even after years of research effort.
Genes must be added to cells. In the genetic engineering of microbes, the cells that are the basis for adding the novel genes are also the complete organisms. With plants this is not the case and this poses the first of several hurdles that the plant genetic engineer must overcome. Once a gene has been inserted into a plant cell it must be coaxed to develop into a whole plant capable of reproducing viable seed that can pass on the newly-inherited trait. In addition, the new gene must not disrupt the normal functioning of the cell. This is more critical for plant genetic engineering than for genetic engineering of single cells for production of new proteins in contained and highly controlled systems.
This limited tolerance for disruption is related to two factors:
First, a plant is an extremely complex organism relative to single cell systems. The genetic makeup of the plant comprises a finely balanced system of specialised cells programmed to carry out complex functions in response to developmental as well as situational conditions. Addition of new genes can disrupt these higher functions even though the critical functions necessary for cell metabolism may remain intact. Discovery of such a potential malfunction may not be evident to the research team at the outset. Rather, it may take field growing conditions to bring out the difference and field testing is not likely to be carried out until much later in the process.
Second, the plant will be ultimately required to survive in the field and produce a yield comparable to the best available conventional seed. This means that the genetically engineered plant must not only have its fully functional complement of capabilities, but the plant must also still be capable of performing well under the wide variety of circumstances that typically accompany the field environment.
The steps in the successful genetic engineering of a plant have generally involved five distinct tasks using the most widely practised method of plant gene insertion. Each of them must be accomplished properly, though not necessarily in sequence after the first two steps:
Identification and characterisation of the gene to be inserted.
Insertion of the new gene into the plant cell (this is an imperfect process and only a small percentage of the cells subjected to the insertion of the DNA actually take up the DNA in a functional manner, thus necessitating the next step).
Separation of the cells that have successfully incorporated the new gene from those that have not. This process, called selection, is intended to assure that only those cells that have taken up the new gene will reproduce to form the whole plant. This is essential if the seed ultimately attained is to exhibit the desired new genetic trait.
Expression of the intended trait by the new gene. Generally, genes code for proteins. When the new gene inserted into plant tissue results in production of the desired protein, the new gene is said to be expressed in the plant.
Expression must occur in such a way that the new DNA is integrated into the chromosome of the cell and is stably passed on to future generations. Once a foreign gene has been inserted and stably expressed in plant tissue, the tissue is considered to be "transformed". Plant transformation is the term used to refer to this process.
Regeneration of the cell into a whole plant that carries the new genes in all its cells and expresses them in some or all of them.
An important and sometimes overlooked part of this process is the seed increase that must be carried out to attain a volume of seed sufficient first for testing and then for commercial use. This single regenerated whole plant becomes the basis for production of seed through the traditional techniques requiring a number of generations of field production. At some appropriate stage, field evaluation of the new plant must be carried out. This typically requires two or more seasons to assure adequate experience under a number of field and weather conditions and to allow for the assessment of performance of the new variety in farmers' fields.
If the plant variety used to obtain tissue for the laboratory manipulations was not from a line that will be desirable ultimately for sale, the genetically engineered seed derived from the research must be crossed with appropriate breeding stock in order to attain a marketable product. This is, of course, an extra step adding further to the time-frame and to the complexity of developing a new seed product. However, research on the genetic engineering of crops has advanced so much in the last decade that scientists are now able to successfully transform and regenerate most commercially important crop species and varieties.
Agrobacterium-based Plant Transformation
The oldest and best understood system for transferring new genetic material into plants is Ti plasmid-mediated transformation based on the bacterium, Agrobacterium tumefaciens. A. tumefaciens is a naturally occurring soil-borne bacterium that is, in essence, nature's own genetic engineer. The bacterium is the causative agent for a plant disease known as "crown gall disease". The disease results from infection by the bacteria which gain entry to the plant through a wound or break in the plant's outer protective layer. The symptoms of infection are galls, or tumour-like growths on the infected plant. The unsightly galls are associated with an alteration of the plant's hormone balance which diverts the plant's metabolic function to produce substances the bacteria use as food. The metabolic imbalance is itself the result of an alteration of the genetic make-up of the plant by transfer of genes from the bacterium to the plant.
Scientists have learned that in the course of infection, the agrobacterium transfers a plasmid called the Ti-plasmid from the cytoplasm of the bacterium to the cytoplasm of the plant's cells in the infected area. Once transferred, the new genes overproduce a plant hormone known as cytokinin. This overproduction distorts the plant's normal cellular metabolism and leads to the formation of the galls which support the proliferation of the bacteria in the infected plant. This natural form of plant genetic engineering provides an excellent model for gene transfer on a broad scale. The objective of scientists is to transfer traits into the plants that are useful to the farmer rather than traits that are useful to the bacteria.
As this disease became better understood, it was apparent that the key to transporting genes into plants via agrobacterium was the elimination of the disease-causing genes and replacement of them with other genes of the scientists' choosing. This has been accomplished and the system has proven versatile and reliable in transforming dicotyledonous (broad leaf) plants such as tobacco, tomato, potato, petunia and sunflower. Tobacco is the most widely used model system and has proven to be an effective tool in the study of new genes and promoters. The Ti-plasmid mediated system has been used in most of the accomplishments in plant genetic engineering to date. These include, for example, disease resistance to crown gall in tobacco, resistance to tobacco mosaic virus in tobacco and tomato, insect resistance in tobacco using the Bacillus thuringiensis crystal protein toxin gene and various herbicide tolerances.
The system is not universally effective in dicotyledonous plants as there remain many species where the transformation has not yet been successful. The Ti-plasmid system has also proven generally largely inapplicable to the transformation of other major group of plants called monocotyledonous plants or grasses. These include such plants as maize, cereal crops, and rice, although limited success using the Ti-plasmid vector in some of these species has been reported.
Other Plant Transformation Technologies
While the Ti-plasmid has proven to be a major tool of the genetic engineer to date, it is limited as to the range of host plants susceptible to transformation. In addition to this, there are other obstacles associated with the dependence of the technique on lengthy regeneration protocols requiring intensive management at every stage of the process. For example, cotton regeneration after transformation via agrobacterium requires a year or more.
In a search for alternative transformation systems that could overcome these limitations, scientists have identified several techniques that offer some promise. To date only one of the alternative systems has been refined to the point where it shows promise for transformation directed at commercial production of improved seed. The following highlights the most significant of these new techniques:
Particle Bombardment
This technique, also called "biolistics", involves coating DNA onto tiny particles and then bombarding plant tissue with them to promote the integration of DNA into selected target plant cells. Substantial success has been achieved using this technique since it was used to produce the first genetically engineered soybean in mid-1988. The method uses an electric discharge apparatus to propel DNA-coated gold particles at rapidly growing soybean tissue taken from immature seeds. Stable transformation was attained and foreign genes have been shown to be expressed in progeny plants. A major advantage of this technique appears to be its applicability to a wide range of species and varieties because it is not dependent on the agrobacterium-based system and it is not dependent on many of the most time- and labour-intensive parts of tissue culture regeneration.
Initial use of this concept was reported by scientists at Cornell University who described a machine adapted to use .22 calibre blanks as the propulsion source to accelerate DNA into onion tissue. The first paper on this technique reported that the particles had carried the DNA into the individual onion cells but evidence of transformation or expression was not attained. Subsequent papers reported progress in both transformation and expression working with improved versions of this same device with a variety of plant tissue targets. Fine adjustment of the device used to accelerate the particles into the tissue is critical to success. Enough DNA must enter the cells to allow uptake of the new genes, yet excessive damage to the cells due to the particle bombardment must be avoided.
There are several major advantages of this technique over the agrobacterium-based method. For one, because it does not use the Ti-plasmid, it does not require transfer of DNA other than that which the experimenter requires to incorporate into the plant. Second, it does not have the host range limitations inherent with agrobacterium and thus should be widely applicable to any plant species. Third, because it can be directed at rapidly growing plant tissues, it does not require the time-consuming steps of tissue redifferentiation associated with tissue culture and it is not limited to the lines that are amenable to regeneration through the tissue culture process.
Electroporation
This technique involves use of an electric current to alter the cell membrane temporarily so as to allow DNA to cross into the cytoplasm and thereby become a part of the cell's genetic code. This direct DNA uptake method, like particle acceleration, has the advantage that only the DNA of specific interest to the investigator is needed in order to conduct the experiment.
Unlike particle acceleration, electroporation requires the use of plant protoplasts. A protoplast is a single plant cell with the cell wall removed through use of enzymes that dissolve the cellulose content of that normally rigid and relatively impervious structure. The electroporation technique was developed originally for use with bacterial and animal cell transformation experiments where no rigid cell walls complicated the process. Enzyme digestion of the cell wall requires carefully controlled conditions to ensure that the cells will be permeable to the DNA upon the passage of the current while at the same time providing sufficient numbers of viable cells to make it likely that the transformed protoplasts will survive regeneration. Because of its dependence on tissue culture regeneration protocols, this technique must be perfected anew for virtually every new plant variety tested as each has unique traits that affect viability and permeability.
Once the protoplasts have been prepared, hundreds of thousands of cells can be treated simultaneously. Cells that express the desired new gene can then be selected for further tissue culture or plant regeneration. Electroporation has been used to transfer genes into several cereal crops including rice, wheat, and maize. As noted above, these monocot plants have been the least responsive to use of the Ti-plasmid vectors in transformation experiments.
There are two principal limitations to the use of electroporation. First, because the technique requires use of protoplasts, lengthy tissue culture manipulations are required and many cells are lost in the subsequent regeneration steps. Some plant species do not lend themselves well to protoplast formation and do not recover subsequently due to initial low viability. Secondly, while the technique has been useful in transforming plant cells, in many cases the transformation has only been transient. Scientists have had considerable difficulty in routinely getting stable transformed plants out of the process at the end of the lengthy plant regeneration period. The method nonetheless shows promise in situations where the problems of transient expression can be overcome.
Microinjection
Microinjection is a technique where the DNA to be incorporated into the cell is physically injected directly into the cytoplasm or the nucleus of the target cell using a special micromanipulator and fine glass micropipettes. The process is an exacting one that must be carried out under a microscope. It demands considerable skill and is quite labour intensive as the DNA must be introduced individually into each cell. To facilitate the procedure, protoplasts must be used. As noted above, this requires the enzymatic digestion of the cell walls and poses the challenge that adequate cell viability must be maintained until the DNA has been successfully transferred. Added time and care are also required in tissue culture to bring these cells back to health and to regenerate whole plants from them. Thus far, this technique has not been very successful in producing transgenic crop plants but, despite its limitations, the technique may yet prove to have utility. It’s future may lie in the transfer of whole chromosomes or organelles, such as nuclei or chloroplasts, rather than individual genes. Indeed, it is one of the most widely used techniques for producing transgenic animals by nuclear transfer.
Except for particle bombardment, common to all of the techniques designed to insert foreign genes into plant tissue is an extensive plant tissue culture process. Depending on the specific technique used, there may be more or less tissue culture effort required to produce whole plants. While tissue culture is thus an integral step in attaining genetically engineered plants using recombinant DNA methods, the tissue culture process itself may be used to develop novel plants without use of recombinant techniques.
Crop Agriculture-Tissue Culture Techniques
The recombinant approaches to altering the genetic make-up of plants described above represent the most elegant and potentially most powerful ways to attain novel variability in plants for use in agriculture. However, a group of techniques collectively known as "cell or tissue culture techniques" offer an alternative approach capable of delivering genetic variation without the complexities associated with use of recombinant methods to introduce foreign genes.
Tissue culture techniques offer advantages over the conventional methods of plant breeding principally because new plant traits can be selected at the cellular level in the laboratory instead of at the whole plant level in the field. This affords advantages of time in that cellular screening for traits requires weeks or months whereas screening of whole plants typically requires entire growing seasons. A related advantage is that the space required for cell screening is substantially less than that required for field screening of traits. More substantively, plants derived from these technologies are said to be capable of greater uniformity because they are derived from single cells in laboratory cultures rather than through sexual reproduction. On the negative side, unlike recombinant methods where the sources of new genetic material can include virtually any living entity, traits attainable using cell culture techniques must be drawn from those which are already a part of the plant's own genetic structure.
The common thread of these techniques is the production of new plants from a few selected plant cells. The techniques comprising this area of development can be grouped into micropropagation, wide hybridisations, haploids, somaclonal variation and mutant selection.
Micropropagation
Clonal propagation, or micropropagation, is a method for production of uniform or disease-free plants. It is now a routine method that most seed companies now employ to propagate at least some of their varieties. The technology involves culture of large numbers of cells or protoplasts obtained from a single plant in a tissue culture and then the regeneration of these cells to fully developed plants (see Fig. 2, bottom) . Because the cells are somatic, that is they contain both the male and female genetic components, the plants grown from these cell cultures are identical or clonal. The technique has been applied to high value crops where uniformity is commercially advantageous. Orchids, oil palm and potatoes derived from clonal propagation are already commercially available. The technique is also useful where seed or planting stocks are likely to carry plant disease. The in vitro development of isolated meristems, or disease-free growing regions of the plant, to the adult plant stage assures exclusion of pathogens. Commercially available disease-free species include potato, sugarcane, and strawberry.
Wide Hybridizations
The techniques involved in wide hybridisation are used to create novel plants through hybridisation of widely related species which would not otherwise cross successfully in nature. Such techniques are described below.
. In Vitro Fertilization
Natural barriers to hybridisation, which occur prior to fertilisation, can be overcome using the artificial pollination of isolated ovaries or ovules. Crops such as tobacco, clover, maize, rice, oilseed rape, poppy and cotton have been hybridised in this way.
. Embryo culture
Post-fertilisation barriers to interspecific and intergeneric hybridisation can be overcome by removing hybrid embryos from the mother plant before they abort naturally and then culturing them to full development in vitro. A wide range of hybrid crops have been produced in this way, including cotton, barley, tobacco, rice, jute, maize and brassicas.
. Protoplast Fusion
By removing the cell walls of plant cells with digestive enzymes, two protoplasts may be joined or fused together in a tissue culture medium. If the fused cells can be regenerated into whole plants, the resultant plant product is called a somatic hybrid (Fig. 2). This hybrid exhibits some of the traits of each of its parent plants.
The selection of somatic hybrid plants after protoplast fusion has been reported in many species, particularly brassicas, citrus, rice, carrot, oilseed rape, tomato, and the forage legumes, alfalfa and clover. However, only a few varieties developed using this technique have proven to be commercially useful. One such example is the tobacco variety ‘Delgold’, which was a product of a fusion between protoplasts from a commercial tobacco variety (Nicotiana tabacum) and a wild relative (Nicotiana rustica). ‘Delgold’ has increased alkaloid content and improved disease resistance and was first isolated in 1978 and released in 1990. In 1994 it accounted for approximately 42% of the flue-cured tobacco acreage in Ontario, Canada, with a value of $US 199 million. Other interesting possibilities also exist. For example, a potato plant resistant to the herbicide triazine has been produced by fusing a potato protoplast with a protoplast of the wild black nightshade. Nightshade is a relative of potato and is naturally resistant to triazine. A less useful, though more widely reported, example of protoplast fusion is the "pomato" plant, produced by fusing a tomato protoplast with a potato protoplast. While the pomato indeed had traits of each plant, and even produced both potato and tomato fruits, the fruits were small and the seed quality was poor.
At present, the technique of protoplast fusion is labour intensive and uncertain in its outcome. It appears that the more closely related the species used in the protoplast fusion, the greater the likelihood of attaining a commercially useful plant. Success has only really been achieved when fusions have been made between genotypes within the same family. The fusion process typically yields a mixture of fused cells of the same species along with fused cells of the two different species. These must be separated and then the desired fusion products must be encouraged to develop into plants. Only a portion of the fused cells usually are capable of this and, once novel plants are attained, there is no assurance that the resultant mixture of traits expressed by the somatic hybrid will actually have commercial value. The main advantages of protoplast fusion over transformation techniques are that, with protoplast fusion, the genes controlling specific traits do not need to be isolated and the traits transferred may be multigenic. It has been valuable mainly in transferring disease resistance into crop germplasm from wild relatives.
----------
Figure 1. Protoplast fusion (top) allows the combination of genes from two different plants, which may be of different species. The resultant single cell may be induced to divide and develop into structures, such as somatic embryos (middle) and small plantlets (bottom), which can then be grown into fully formed somatic hybrid plants. (Photos courtesy of Kazutomo Yokoya)
---------
Haploids
Haploids are fully formed plants but with only half the normal (diploid) number of chromosomes. They are useful in breeding because they exhibit recessive traits or mutations and, once their chromosome number has been doubled, they can be used immediately as inbred parental lines, saving time, expense and space compared to conventional methods. Most crop groups have benefited from a saving of 7 years or more in the development of inbred lines using in vitro methods for haploidisation. Such methods include rescue of embryos after cross pollination of barley cultivars with Hordeum vulgare (bulbosum technique), or the in vitro culture of isolated ovules or microspores in other crops.
Somaclonal Variation
While the clonal propagation technique emphasizes the production of uniform plants from somatic cells, somaclonal variation seeks to utilise what had been observed as a problem in clonal propagation. Scientists observed that in some cases, variants would develop from somaclonal cells, that is, cells that had identical genetic material. This genetic variation of a few cells relative to thousands is typically masked in the whole plant. However, when plants are derived from single cells which express a particular variation, all cells of the new plant contain the variant and the trait thus becomes manifest. Early work on this technology was done on tomato with the objective of attaining tomatoes with high solids content and tomatoes with improved harvesting traits. Work using this technique is now directed at several crops including maize, wheat, celery, carrot, bananas, potato and sugarcane.
Mutant Selection
By combining the cell culture techniques of somaclonal propagation with selection and screening methods, researchers are working to develop new plants that are resistant to herbicides, disease and stress. By exposing protoplasts or dividing cells in cell culture to an herbicide, for example, and then selecting only those cells that survive for regeneration, the resultant whole plants may themselves be resistant to the herbicide. A similar approach can be used to attain plants resistant to disease and plants resistant to environmental stresses. The first plant patent issued for an herbicide resistant maize plant derived from this technique. The maize line was developed by Molecular Genetics, Inc. in collaboration with American Cyanamid and is resistant to American Cyanamid's imidazolinone herbicide. The line was licensed to Pioneer Hi-bred International (see figure 2) which made backcrosses to incorporate the trait into commercially valuable inbred stocks.
Animal Agriculture-Health Care Applications and Products to Improve Productive Performance
Just as conventional pharmaceuticals developed for human applications have found applications in animal health, biotechnology-based animal health applications in agriculture largely derive from development of recombinant products for human health care purposes. Products in development for sale in the animal health market in therapeutic applications as well as vaccine applications utilize the same technology as those that form the basis for human health care products discussed in Chapter 22. Indeed, some of these products are the identical proteins or, in some cases, the animal analogue of the same form produced for human consumption. Examples include use of interferons or interleukins to enhance the animal immune response as well as the use of a species-specific growth hormone to enhance milk or meat production.
--------
Figure 2. A scientist from the Biotechnology Research Department of Pioneer Hi-Bred International, Inc., applies herbicide to an experimental plot of corn. Pioneer has been involved in the development of new hybrids resistant to the imidazolinone family of herbicides produced by American Cyanamid Company, under license from American Cyanamid. (Photo courtesy of Pioneer Hi-Bred International, Inc.)
--------
Animal Agriculture-Transgenic Animals
Introduction of new genetic material into animals to produce transgenic animals may be applied in a number of areas ranging from applications to attain higher quality meats by virtue of alterations in the fat to lean ratio of the animals' muscle, to the production of useful proteins secreted into milk or eggs. The technologies to attain these results lag somewhat behind other areas of agricultural biotechnology. Transgenic mice, rabbits, pigs and sheep have been routinely produced by direct injection of DNA into the pronucleus of fertilized eggs.
The Agricultural Sector
Agriculture is said to be the only truly indispensable industry. As such, agriculture is the underpinning of developed as well as developing economies. In virtually every country of the world, agriculture is the largest single category of economic activity. In the U.S., agriculture touches nearly every sector of industry, even though the value of farm commodities is small relative to the value added downstream in the industrial processing sector.
The agricultural sector is comprised of the nation's physical production facilities, the farms and the people that operate them. In addition to the farms, the input industries that serve agriculture with input products such as agricultural chemicals, seeds, machinery or financing are also integral to the farm sector. The principal forces that affect the farm sector are public policy, exogenous variables and technology. Included in the public policy variables are price supports, direct payments, commodity loans and acreage diversion programs. Collectively, these public policy choices are called "the farm program" and as such are a topic of constant debate among legislators, farmers, input suppliers and consumers. The program, whatever the specific components, seems always to be under critical review for substantially failing to accomplish its intended objectives. Exogenous variables comprise a wide variety of factors including weather, disease, pest infestations, economic conditions and international trade factors. The consequences of these variables on the farm sector often provide incentive for further adjustment of public policy, thus leading to ever-changing dynamics of the farm sector. Technology defines the level of capability of the farm sector and, to an extent, the nature of the output of farms. Technology has had a major influence on the farm sector this century, accounting for a total transformation in the way food and fiber are produced.
Food and Fibre Consumption Trends
U.S. households spend more on products derived from agriculture than any other category of goods and services, including housing. This is true despite the U.S. being noted for low food costs in comparison to other countries. In 1984, consumer end use expenditures for farm origin foods was $332 billion. The value of the farm component of this expenditure was only 27% or $89.5 billion, while the balance amounting to $243 billion comprised the marketing bill. This $243 billion is made up of packaging, processing, transportation, wholesale and retail markups, advertising, taxes, interest, repairs and other factors. Labour, at the various stages in the food marketing chain, is the largest single component of the marketing bill, about 33%. In general, the value of the farm component of retail food expenditures is declining as a percent of the total.
In 1986, for the first time, the value of the processing component of food purchased at the checkout counters of foodstores in the U.S. exceeded the value of the food itself. The USDA Economic Research Service reports that about 30 cents of every dollar spent in foodstores in 1986 went to cover the cost of food production while about 31 cents paid for the processing (including processing labour) of the food after it left the farm. This trend toward declining farm share of food value as a percentage of consumer food expenditure is about a decade old. The farm value component of food expenditures had been relatively constant at between 30-40% of the consumer food dollar through the 1960s and 1970s. Since 1979, that share has been steadily declining. This suggests either that the processing sector is undertaking an increasing role in food preparation or that overproduction of farm commodities is depriving the farmer of a share of increasing consumer food spending. In fact, both of these trends are known to be occurring.
In 1984, 5.8 million people, about 2.4% of the U.S. total, lived on farms. This population comprises the workforce for direct production of U.S. crop and livestock products. Another 19% of the U.S. workforce is employed in manufacturing, processing, transportation and retail industries related to farm equipment supplies and services. When non-food uses of products from the agricultural sector are included, the market value generated by the farm sector measured at the end use level is about $575 billion. Of this, farm sector income amounts only to $174 billion. Other than food products, the components that make up the balance of the more than half trillion dollar value are exports, non-food goods such as clothing, shoes and industrial goods, tobacco and tobacco products and farm origin inputs for agriculture. Farm origin inputs include seed, animal feeds and livestock for breeding.
Size and Nature of Modern Agriculture
U.S. farming has changed dramatically in the last five decades. The most noticeable difference is in farm structure and organisation. Many studies have analysed and characterised these changes. The following factors are usually cited in most studies as principal indicators of the changes that have occurred:
. Fewer and larger farms
. Increasing concentration of production
. Increasing specialisation
. Growth of multi-ownership farms
. Use of capital requirements with commensurately more complex financial structures
. Greater concentration of wealth and income among the larger farms
. Fewer young people entering farming than older operators leaving
Many factors aided in bringing about these changes. Some of the frequently cited influences include:
. Inflation
. Increased exports of farm products
. Availability of capital-intensive new technologies
. Non-farm employment opportunities
. Availability of institutional credit for land purchase and capital goods
. Commodity programs supporting prices for farm products
. Tax laws applicable to incomes and estates
Changes in the Farm Sector
The most significant and perhaps most obvious structural change of those mentioned above is the decline in the number of farms and the increase in average size of the farms remaining. The number of farms in the U.S. decreased from 6.5. million in the 1930s to about 2.4 million in the 1980s. Total cropland use has declined slightly only recently. In 1979, the amount of land in crops in the U.S. was about the same as it was in the 1930s, about 370-380 million acres. In recent years, with the acreage reduction programs of the federal government in effect, land in crops has dropped to little more than 300 million acres. Thus, with the decline of the number of farms has come the consolidation of small farms to form more economically efficient entities. A good part of this consolidation occurred between 1930 and 1960. The rate of decline in the number of farms has slowed since that time. In the 1950s farm numbers dropped 30%. From 1949 to 1970 the rate of decline was substantially less at 1.1% per year. By the early 1980s, the decline had ceased but with the low prices and agricultural instability of the farm sector brought on by overproduction, the number once again began to drop.
Average farm size doubled from the 1950s to 430 acres per farm in the 1980s. The number of farms greater than 500 acres in size doubled from 1960 to 1978 while the number of farms less than 100 acres in size declined twofold during the same period. In 1978, the number of farms having greater than 1,000 acres accounted for only 7% of the U.S. farms, the total acreage of these farms comprised 90% of the total U.S. cropland.
While the rate of change in the total number of farms slowed in the 1980s, the rate of change was still significant within certain segments. Small farms, defined by the USDA as farms having annual sales of $2,500 or less, and large farms, those having sales of $100,000 or more, both increased in number while mid-size farms, those with sales between $20,000 and $40,000 declined. Thus, mid-size farms have been under considerable pressure to become larger or to become smaller. If this trend continues, U.S. agriculture will be characterised by a concentration of very large farms on the one hand, providing most commercial output and occupying the vast majority of the land and, on the other hand, a large number of very small farms occupying a small percentage of the land collectively and producing very low output. USDA data indicates these small farms are owned increasingly by the so-called hobby farmer. These are part-time farmers who have regular, and perhaps significant, off-farm income sources and embrace an affection for the quality of life afforded by a rural environment. For these farmers, the farm serves primarily as a rural residence. Sale of farm commodities appears to be only a secondary activity and is only nominally pursued by farmers in this category.
A phenomenon that was first noted in the early 1980s, which has received considerable attention, is the so-called "superfarm". Superfarms are defined by USDA as farms having sales in excess of $500,000 annually. Table (1) provides a summary profile of the superfarm as described by USDA.
The Farmer as Customer
Because a significant portion of the applications of biotechnology to agriculture will take the form of products designed to improve the profitability and productivity of the farm, the purchaser of these products is generally going to be the farmer. As the ultimate buyer of seed, chemicals, implements and other farm services, those marketing biotechnology-based products will have to gain the farmer's attention sufficiently to convey the unique features these new products offer and why they afford benefits over currently available alternatives. An advantage held by many of the companies developing new products for agriculture is a long history of serving farmers and farm markets. Even with this customer familiarity, some companies will find marketing biotechnology products a new challenge as numerous products already populate the market. The sheer number of such products has heightened price competition as companies fight for market share in a current era of declining acreage due to over-production of farm commodities. In addition, the increasing sophistication of farm use products has created a sceptical buyer who carefully evaluates the probable pay-off of each prospective purchase before buying. Nonetheless, farmers have generally demonstrated a strong interest in the potential of biotechnology and its likely contribution to farm productivity. Farmers have proven themselves capable of widely and rapidly adopting new methods, practices and products once they are shown to be valuable. As noted above, these improvements have revolutionised farming over the last 40 years. In the U.S., for example, according to USDA the average farmer produced enough food for 14 people in 1947. In contrast, by 1986 the average farmer produced enough for 93 people. Such change has not come without a strong sense of experimentation by farmers towards new technology.
--------
Table 1
Superfarm Profile
--------------------------------------
* Although superfarms account for only 1% of total farms, product sales from these farms accounted for nearly 30% of total farm sales. This amounted to about $30 billion in 1978.
* Average size is 4,700 acres
* Total land in superfarms was 84.5 million acres or 7% of total farmland. Cropland accounted for 25% of superfarm land, pasture and rangeland accounted for 64% and other land uses for 11%.
* Superfarm crop sales totalled $10 billion, while livestock and other farm product sales totalled $20 billion in 1978.
* Net superfarm income amounted to $14.3 billion in 1982, accounting for 60% of U.S. net farm income.
* Corporations owned about 40% of superfarms and nearly half of superfarm land in 1978. Sole proprietorships accounted for 37% of superfarms and partnerships for 20%.
* Superfarms are concentrated in California (about 20% of total superfarms) Texas (7%); other states, including Florida, Iowa, Nebraska, Kansas, Illinois, Colorado, Washington, Arizona and North Carolina, comprise 33%. California and Texas together accounted for one third of the superfarm sales.
* Superfarms are most concentrated in speciality crops. Percentage figures indicate percent of total production in that crop category which originates on superfarms:
Vegetables 62%
Nursery and greenhouse products 55%
Fruits, nuts, and berries 44%
Grains 7%
* In livestock, superfarms are less concentrated.
Cattle and calves 44%
Poultry and poultry products 42%
Dairy 13%
Swine 9%
* Superfarms typically are more efficient than smaller farms relative to level of expenses as a percent of each dollar of gross income ($0.69 per dollar gross income for superfarms versus $0.84 per dollar gross income for farms having between $200,000.00 and $499,000.00 per year sales).
----------------------------------------------------------
While on the one hand farmers have generally proven themselves to be willing to try new products, the price they are willing to pay for them is typically only a fraction of the benefit they are likely to derive from their use. In the industry this is referred to as the "benefit cost ratio". For example, a 4:1 benefit cost ratio indicates that a farmer is willing to pay $1.00 for a product that will generate $4.00 of added value to his operation. In the U.S., farmer benefit cost ratios vary from less than 2:.l in the case of some highly effective and well accepted pesticides to greater than 10:1 for some implant products for animal growth enhancement. Typically, a benefit cost ratio of 3:1 or 4:1 is common for agricultural inputs such as pesticides and seed.
The principle factor that influences the benefit cost ratio is risk. Farmers usually purchase products early in the season and hope for returns on their use by the end of the season. Many uncertainties will intercede between product purchase decision and payoff. Most of these uncertainties are beyond the farmer's control. Weather and market price for his crop are the two principal categories for this uncertainty but other performance factors associated with the product itself, with the field conditions, with farm equipment interfaces and other factors also contribute to the discount the farmer applies to the value of a product.
The net effect of this for companies developing products, whether they be biotechnology-based or otherwise, is that the value of the benefits to farmers associated with use of a particular product is substantially greater than the available market for the product.
The Agricultural Input Industry
The point has already been made that, the agricultural input industry comprises the companies that provide the inputs necessary for the farmer to produce farm commodities. These range from crop inputs such as fertiliser, seed and pesticides, to animal inputs such as animal health products and breeding stock for animal production. Other inputs include farm machinery, financial services, and, increasingly, information services and computer hardware and software dedicated to farm applications. The sectors of this industry most generally applicable to marketing of biotechnology-based products are companies engaged in production and marketing of seed, agricultural chemicals and animal health products. The following briefly highlights these three key sectors in terms of the participants and the nature of the businesses they operate:
Seed
The seed industry world-wide accounts for about $25 billion in sales annually. The commercial value of seed used by farmers globally has been variously estimated at between $45 and $60 billion. The difference in these values is due to the widespread farm practice known as plant-back. Farmers often plant seed saved from the prior year's harvest for the next year's production instead of purchasing certified seed from a seed company. This practice is particularly common in the major field crops such as wheat and soybean. To a lesser extent, it is common in some production regions for cotton. Because these crops are open pollinated, parent plants transfer traits uniformly through seed to progeny plants. The progeny will have the same genetic potential for yield and other traits associated with crop performance as the parent plants. Thus, for these crops, as long as the farmer is satisfied with the performance of his prior year's crop, there is limited incentive again to purchase seed. Generally, the price of open-pollinated seed is a small premium over the commodity value of the seed. The premium is based on the cost of cleaning and conditioning the seed plus a charge to offset breeding of improved varieties. Only in the hybrid crops, where progeny segregate into non-uniform populations with varying traits relative to their parent plants, is the practice of plant-back seldom found. The yield loss associated with these dissimilar field operations eliminates any economic incentive to plant with saved seed. Major hybrid crops include maize, milo, sunflower and tomato. Because of the limitations on plant-back, hybrid seed can be sold at a greater premium over the basic commodity price. Also, because of the hybrid vigour associated with hybrid seed, the crops are higher yielding and hence of greater value. Thus, the seed price per acre of hybrid seed in comparison to non-hybrid seed is usually substantially higher. As a result of this greater price flexibility, greater performance potential and inherent protection of the commercial value-added, companies spend considerably more in research to develop improved lines for the crops that can be hybridised than for the crops that are open pollinated. The seed industry is, however, segmented by crop and by geographic area within the various segments, so there are significant differences that limit the usefulness of any generalisations.
That notwithstanding, there are generally two broad types of companies in the seed business. On the one hand are small, typically family-owned seed companies that focus on serving a regional market with seed for crops grown in that area. These companies generally specialise in varieties particularly well adapted to their marketing regions. They have limited research capability and often rely on public germplasm or germplasm licensed from foundation seed companies. On the other hand, a group of large diversified, often multinational seed companies have emerged with significantly larger marketing power than the small regional family operations. Some of the growth of these companies has come as a result of, and perhaps a reward for, the development of superior lines often associated with hybrid technology. A number of the other large seed companies have grown through acquisitions of prominent small regional companies to gain broader market access. Many of the larger companies have been acquired by major agrochemical companies in the last few decades. At present, it is not unusual for a major agrochemical producer to have a seed company subsidiary that holds a significant share of the seed market in one or another major world crop.
Agricultural Chemicals
The pesticide industry has traditionally been comprised of basic producers and formulators who often also act as wholesalers to the distribution system. The pesticide industry emerged as a major supplier to farmers after World War II, largely as a result of the successful application of synthetic organic chemistry to the development of farm pest control products. The introduction of such new products contributed to major changes in the economics of farming by allowing farmers to farm more land and preserve a greater share of the crop for harvest. Pest control afforded by the new chemistry provided a major advantage to farmers and thus rapidly grew into a major industry. The estimated 1986 sales of the worldwide agrochemical industry were $17.4 billion. The vast majority of agrochemical sales are for farm use.
The basic producers who made early commitments to research efforts to develop new chemical pesticides generally became the most successful in the marketplace and typically dominate the industry today. The companies best positioned to undertake this research were already the major chemical producers who had access to the appropriate feedstocks, the production facilities and technical expertise necessary to bring the products to the market. Today, most of the principal agrochemical basic producers are also major chemical producers. As the industry was becoming established, regionally specialised companies purchased the active ingredients from the basic producers and prepared formulated material for sale into the farm distribution channels. Often the companies operated distributorships as well as formulation plants. As the industry matured, the control of the formulation function increasingly shifted to the basic producers. This shift was encouraged by the increasing competition within the industry as more products crowded the market. With the greater competition, formulation became an integral part of product market positioning. The basic producers began to integrate forward into formulation as a means of improving margins and gaining more control over the channels of distribution. Formulators increasingly turned to producing off-patent active ingredients as a means of maintaining sales of their formulated products. In some cases, they would market these products themselves through their own distribution system but they would also produce the chemicals under private label for sale by others.
As biotechnological applications to agriculture have developed, the entry of new companies into the pest control marketplace has speeded up. Chapter 24 of this section provides a more detailed discussion of the pesticidal applications of biotechnology insofar as actual pest control materials such as biopesticides are concerned. There is likely to be an assured role for the agrochemical companies in the new biotechnology as most of the major players have already made commitments to seed company acquisitions, to biopesticide research and to development of plant genetic engineering skills.
Animal Health Products
In 1995, world animal health product sales were reported at $10 billion annually, with pharmaceuticals accounting for more than half of this. The products that comprise these sales include animal therapeutics, animal vaccines and nutritionals. The provision of animal care products to veterinarians is dominated by companies who are already major participants in the human health care business. That is, a company already committed to the development of human pharmaceuticals is in a good position to create a division charged with finding applications of the developed technology for animal care markets whether that be marketing to producers of farm animals or to the veterinarians who provide animal care. Biotechnology applications to animal health appear at present to fit well within this industry structure. Many of the most significant efforts in the development of new biotechnology products in this sector have been pursued in small start-up companies who have special research capabilities in the new techniques; these companies are frequently allied through funding or marketing arrangements with the major animal health companies.
A number of studies have attempted to inventory the activity in this rapidly changing area. Vaccine production is the fastest growing component of this market and it has been reported that there are several hundred companies and other organisations active in developing products for just the veterinary vaccine market.
Constraints to Growth: Regulatory and Public Relations Issues
This section highlights the major constraints to growth applicable to the use of biotechnology in agriculture. Many of the factors involved are the same as those discussed in earlier chapters of this book; thus they are not covered again here other than to note aspects of these issues specific to biotechnology in agriculture.
The success of the agricultural biotechnology industry, while clearly dependent on the progress of the research and the demand for new biotechnology products in the market-place, is also critically dependent upon public policy and public perception. Agricultural applications of this technology will require widespread use in the environment on thousands, indeed more likely millions, of acres. To be effective, the products must survive and compete in the field environment whether they are seeds or microbial products applied to crops or improved vaccines for use in animal health applications. In the view of the regulatory agencies there is no clear line which distinguishes the degree of viability that a crop plant or an agricultural microbe must have to be effective in the field and the degree of viability that would be undesirable in a weed or an out-of-control microbe.
The history of the debate about the appropriate level of regulatory scrutiny to be applied to securing clearances to market, and indeed even to conduct small scale field tests of biotechnology products, has been characterised by a concern that unpredicted adverse effects may accompany their introduction. Proponents of a slow approach to approvals of biotechnology field tests have argued that past experience with chemical technology has demonstrated the value of proceeding with the utmost caution. The record of the regulatory debate contains many references to the dangers of introduced species. The suggestion is that our experience with such introductions should teach us that genetically engineered microbes or plants released to the open environment without appropriate containment may lead to unexpected environmental problems. The heritage of the environmental movement of the 1970s, the argument continues, dictates that the only rational way to proceed is to conduct a careful assessment of the likely environmental and ecological effects of each introduction before the introduction is actually carried out.
While arguments like these have an elementary appeal to the casual participant, there are several problems they pose for the encouragement of innovative research in the area of agricultural biotechnology. First, there is not yet the ability effectively to predict the behaviour of even known organisms in the environment. This lack of so-called predictive ecology has been held up as reason for caution in proceeding quickly to introduce new products of biotechnology. On the other side of the issue is the conviction that predictive ecology is an illusive goal at best and more probably is never attainable (somewhat like the goal of predictive economics or predictive psychiatry). The argument is that no matter how much additional science is known and how much additional data can be assembled, there is little likelihood that accurate predictions about the behaviour of complex organisms in even more complex environments will result.
Second, with or without predictive ecology, it is not clear that one can assess a priori the problems likely to derive from an environmental introduction. The most frequent analogies cited are to toxic chemicals and to hazardous pesticides. The more extreme analogy is that of nuclear power. The argument is that had we known the downside effects of introducing these new technologies we would have proceeded more cautiously. The basis for the suggestion that these analogies are appropriate appears to be that, as events turned out, we learned things about these technologies later that we might have wished to have known earlier. This appears somewhat of a circular argument. While it may be reassuring to be able to say we know now all the downside risks and unexpected outcomes that are ever going to befall society as a result of the adoption of new biotechnologies, such an outcome is clearly never possible. Certainly nations will suffer problems from biotechnology as there have been problems from any new technology. Certainly we would prefer to avoid as many such problems as possible. But clarification of the problems themselves does not evolve from a theoretical debate before experience has been gained. We have to experience a problem in order to understand it so that we can solve it. To talk of solving problems before they emerge borders on the fanciful. We can take positive steps in programs of accident prevention as, for example, in the case of highway automobile safety by careful study of prior experience and making adjustments in highway construction, traffic laws, auto construction or driving practices to eliminate conditions that have been associated with accidents. This is appropriate because accidents are by nature unpredicted events with undesirable outcomes. What is being urged for an ex ante analysis of biotechnology now is, in effect, analogous to the process of attempting to define the complete traffic accident prevention program prior to the emergence of mass highway travel by motor car. In 1920, for instance, there was simply not enough experience with the automobile to establish any such program. Any effort to develop a complete one during that timeframe would have been a hopeless failure in the light of affairs now and in the future.
Third, it is not clear that the risks of products of agricultural biotechnology are comparable to the technologies with which biotechnology is being compared. As noted before toxic waste problems, synthetic organic chemicals and nuclear power are the "bad" experiences that have been cited by the critics of biotechnology. Because of problems with these technologies, they argue, we must proceed especially carefully in adopting biotechnology. But this argument by analogy fails to demonstrate what, in fact, the similarity is between biotechnology and the selected other early technologies that led to unforeseen adverse results. The similarity asserted by the proponents of this argument appears to be simply that all were new at one time and that new technologies invariably have unpredicted, adverse consequences. When reduced to a syllogism, the flaw of this argument is more apparent. It rests on the premise that "all new technologies have unpredicted ... etc.... consequences". Support for this premise is the citation of a few technologies that have indeed resulted in unanticipated problems. But there has been no exhaustive demonstration to show that all new technologies have such problems. Many, indeed, have become integral parts of our daily lives with never a thought to a heritage of adverse consequences. Consider, for example, technologies such as computers, telecommunications, the adoption of jet engines for commercial airlines, and the use of electric lighting. While one can certainly imagine negative outcomes associated with all of these advances, they are largely accepted by most as, at worst, benign to society and more likely as contributory to substantial benefit.
A review of the safety record of the biotechnology industry and research establishment reveals no suggestion of a troublesome technology. The decade-and-a-half of safe recombinant DNA research in the laboratory is augmented by several decades of safe experience with extensive and often random genetic manipulations in plant and animal breeding, in the field for the purpose of improved agricultural productivity. Indeed, this conventional work has been carried out without notice by those concerned with environmental degradation. The most recent relevant experience is the favourable experience of the field tests that have been carried out on genetically engineered plants and microbes. In none of the tests conducted as of this writing has there been a reported problem.
Perhaps because of these intuitive factors, it will be merely a question of time before the extreme scrutiny to which recombinant DNA experimentation has been subjected in the press and at the state and federal regulatory agencies will subside to a level more commensurate with the risks associated with the experimentation. In fact, the regulatory climate has improved to some extent during the last decade. In 1986, field tests of a bacterium genetically engineered to protect plants from frost damage proposed by researchers at the University of California, Berkeley, and by their colleagues at a small biotechnology company called Advanced Genetic Sciences of Oakland, California, were repeatedly deterred by regulatory delays, court challenges and tactical errors in testing. Opponents of biotechnology were largely responsible for the delay. Similarly, also in 1986, a proposed field test of a microbe genetically engineered by Monsanto was dropped. Monsanto scientists had engineered the microbe to produce an insecticide that has been in common use in agriculture (the gene inserted into the microbe coded for the crystal protein toxin gene from the bacterium Bacillus thuringiensis). The U.S. Environmental Protection Agency determined that Monsanto must obtain an Experimental Use Permit under provisions of an interim policy adopted by the agency to review small scale field tests of biopesticides developed using biotechnology. The agency declined to approve the permit in 1986 and Monsanto eventually dropped the project.
--------
Figure 3. Agracetus scientists inspect the progress of young tobacco plants during the first outdoor test of a genetically engineered plant in 1986 near Madison, Wisconsin. The plants were genetically engineered to be resistant to crown gall disease, a bacterial infection. (Photo courtesy of Agracetus.)
--------
Three separate tests of genetically engineered plants were conducted in 1986 (Figure 3) and again in 1987 further testing went forward for plants. All the plant tests were carried out without incident. In the area of field testing of genetically engineered microbes, two tests of the bacteria designed to prevent frost damage were conducted in 1987, along with a third test of a microorganism carrying a marker gene. The latter test proved that the microbes did not spread in the environment as the opponents of the tests had feared. By 1988, additional field test activity proceeded with both plants and microbes without widespread attention or incident. In general, the novelty of these tests has worn off to the point where regulatory agencies and the media no longer dwell on them. As a result, the general public is largely taking the view that these tests are mundane and unlikely to lead to environmental disruption. In this more reasonable atmosphere of public opinion, opponents of the technology have turned to other areas in an effort to plumb the imagination of the general public to take issue with the progress of biotechnology. Labelling of products derived from genetically engineered crops is one such issue. The 1996 controversy in the EU states regarding Monsanto’s glyphosate-tolerant ‘Roundup Ready’ soybean stemmed from pressure by several European trade associations and other interest groups. Despite formal EU approval for importation, these organisations further insisted that transgenic soybean products used for food or feed should be separated and labelled. Other issues for the industry in this regard are the patenting of animals and the long-term social and economic effects of adoption of biotechnology on the small farmer
THE CONTENT OF AGRICULTURAL BIOTECHNOLOGY
This section covers the substance of what is coming in developments for the agricultural sector. Applications of biotechnology to agriculture may be divided into two broad categories:
Near-term Developments
Products in development now that will comprise the initial applications of agricultural biotechnology. This section is based on announced products in testing, development or commercialisation by various companies in the ag-biotech industry.
Long-term Potential
These are products that will follow the initial applications described under near-term developments. Second generation ag-biotech products will generally require development of additional scientific understanding or further refinement of current techniques before they become a commercial reality.
Near Term
Many of the products of agricultural biotechnology now in development address problems for which solutions already exist rather than problems for which solutions are yet to be found. This is in contrast to biotechnology applications in human healthcare where most new therapeutics are being developed to solve problems that have not been amenable to treatment using conventional pharmaceutical technology. For example, medical science has yet to find a cure for cancer, but the recent advances in medical research made possible through biotechnology has crystallised the prospect that new drugs will soon be available that will cure selected cancers and will alleviate the severity of the disease in other instances. Considerable work is underway in this area directed toward commercialisation of new products which will substantially address this type of unmet need.
In the case of agriculture, the near-term products of biotechnology are largely new solutions to problems that already have remedies even though the remedy may have several disadvantages. The new solutions will be better because they allow the farmer more flexibility, because they afford more complete management control of the production environment or because they may cost less than conventional solutions. Nonetheless, they are likely to be incremental improvements in the view of the farmer and perhaps also in the view of the investment and business community.
While product sales for many of these near term applications of biotechnology may be in the multi-million dollar per year range, they are not likely to be grossly different from the size of markets for current conventional products. The value of each product will likely depend on the degree of improved efficiency afforded by that product over its conventional rival.
A guide to the near-term applications of agronomic biotechnology are the number and class of notifications to introduce genetically modified plants to the US Health Inspection service (APHIS). These introductions include field tests, US importation and interstate movements. From 1987 to October 1997, the number and percentage of the total notifications for different classes of genetically engineered phenotypes were herbicide tolerance (1396)(31%), insect resistance (1286)(29%), product quality (957)(21%), virus resistance (251)(6%), agronomic properties (238)(5%), fungal resistance (193)(4%), bacterial resistance (25)(1%) and others (149)(3%). A description of some of these applications is given below. The target crops thus far have been mainly those farmed on a large commercial scale and are listed in Table 2. The predominating crops are clearly those of vast economic importance, particularly food crops, such as maize, potato, soybean and tomato. Cotton and tobacco are the only important non-food crops which have been genetically modified on a significant scale so far.
-----------------------------------------
Table 2. Total Notifications of Transgenic Crops to APHIS up to October 1997
________________________________________
Crop Total
Notifications
________________________________________
Alfalfa 2
Beet 13
Brassica oleraceae 5
Carrot 5
Maize 1989
Cotton 360
Cucumber 1
Flaveria bidentis 2
Melon 6
Nicotiana benthamiana 1
Pepper 1
Pineapple 1
Poplar 9
Potato 425
Rapeseed 49
Rice 10
Soybean 462
Squash 1
Strawberry 2
Sunflower 4
Tobacco 178
Tomato 469
Walnut 7
Wheat 23
______________________________________
Herbicide Resistance Traits in Plants
One of the major improvements to agricultural production practices realised over the last 40 years has been the development of effective chemical herbicides for weed control in crop production. The importance of herbicides, as well as other important agrochemicals, and their applications to agriculture are covered in chapter 24 of this volume. The net effect of these chemicals has been to free the farmer from reliance on mechanical cultivation or hoeing for weed control. While cultivation employing disks or chisel plows pulled behind a tractor is still used physically to eliminate weeds in conjunction with herbicide spraying, it is now largely regarded as a minor tool used to supplement an herbicide program. The problems with cultivation were that the practice did not provide long-lasting control and it required a major investment of the farmer's time. Weeds could grow back within days and the amount of land a grower could farm was limited in part by the farmer's capability to cultivate mechanically.
Modern herbicides changed all that. Control was long-lasting and spraying was a simpler operation than cultivation. This change contributed to making large scale farming operationally more feasible, particularly in the large acreage field crops such as maize and soybean. Freed from these cultivation limitations, farmers expanded their acreage to take advantage of the economies afforded by the larger scale of operation.
As a result of the demand for weed control chemicals, herbicides became the major segment of the agricultural chemicals market, accounting for $8,600 million or 43% of world wide agrochemical sales in 1987. Today there are multiple herbicides available for virtually every crop grown. In response to the large market potential, many companies are active in development of new herbicide products and there is considerable competition on performance, price, product flexibility and convenience.
Despite the foregoing, herbicides are still far from perfect. Some weeds are not well-controlled in some important crops. Some herbicides, when applied in sufficient quantity to control weeds, lead to crop damage. There are many complex weed-crop interactions that often require a farmer to use several products over the course of the growing season to attain adequate weed control. Timing of application may be critical to successful weed control in many situations while carry-over effects of the herbicide on the following year's crop can become limiting factors in selection of a weed control chemical. By the time the ideal mix of products is applied, the cost may be so high that the crop is no longer profitable. In addition, adverse environmental effects have been associated with some herbicides ranging from human health problems to impact on non-target organisms. High rates of application of some of the older herbicides have led to concern about spray drift, worker exposure and ecological effects of the herbicide program. Some of the newer chemicals now entering the market require as little as .02 lbs per acre. In comparison, many of the older chemicals required typically 5-10 lbs per acre.
The ideal herbicide would be one that provides long-lasting effective control of all the plants in the field except the crop plants. Although considerable effort has been directed at this objective by the agrochemical companies, synthetic chemistry has yet to attain such a perfect herbicide. Rather, a wide variety of herbicides have been developed with different levels of activity against different weeds and different degrees of phytotoxicity to various crops. The lack of a perfect herbicide has led to developmental efforts using genetic engineering to improve the tolerance of the host crop to selected herbicides. This is, in effect, using biotechnology to improve the safety margin of herbicide use and thereby extend the markets for the herbicides for which resistant crops can be developed. The market for the sales of herbicide tolerant crop varieties in the U.S. has been estimated to have been $50 million to $200 million up to 1995. Analysts predict that herbicide sales will increase for those products where crop seed resistant to that herbicide becomes available. Such sales increases are expected to be at the expense of other, less versatile, herbicides.
The incentive for development of herbicide resistant crops generally comes from the companies offering the herbicide products themselves. Many of the most widely sold herbicides are popular because of their broad spectrum of weed control. However, in some cases, the wide use of an herbicide on some crops is not possible because of crop phytotoxicity problems. Also, a continuing stream of new product introductions offers farmers new weed control choices, possibly resulting in lost sales of a highly successful established product. Availability of crop seed carrying resistance to a particular herbicide product provides the company with a defensive tool to prevent loss of market share to new competition. At the same time, herbicide resistance can provide a strong offensive weapon to expand the market for a particular herbicide into crops that had not been possible before due to the chemical's phytotoxicity to that crop.
The potential benefit of herbicide resistance for the farmer is simple and effective weed control available with fewer chemical applications and less risk of crop damage. In some situations, the farmer may also gain added flexibility in the choice of herbicide or in the use of a more desirable crop rotation strategy. This is because some herbicides have the potential to damage crops the following season due to residues left in the field over the winter. With herbicide resistant varieties, the farmer could avoid this problem by choosing a seed variety that would not be affected by herbicide residues. A type of crop production that benefits particularly from herbicide resistant crops is no-till agriculture. Minimising or eliminating tillage stops soil erosion and reduces pesticide runoff but weeds become a serious problem if herbicides are not used aggressively. The use of herbicide-tolerant crops gives farmers lower herbicide inputs but effective weed control. The benefit for the company marketing a broad spectrum chemical with an already established reputation is the prospect of strengthened sales to the extent that farmers choose to use seed varieties carrying resistance to the herbicide.
As result of these incentives, there has been strong interest in biotechnology products able to address this need and products from this research have been amongst the earliest to reach the farm. In many cases, herbicide resistance is a single gene function. Traits for resistance to a particular herbicide can be engineered relatively easily once the mode of action of the herbicide is understood. Usually the herbicide is effective against plants because the active ingredient, attacks an enzyme critical to the plant's metabolic activities. If a gene coding for an alternative to this enzyme is added to the crop plant, and if the gene expresses the enzyme at sufficiently high levels, the plants should be unaffected by field applications of that herbicide sufficient to eliminate weed pests. Alternatively, if to the plant a gene can be added that produces an enzyme to break down the herbicide itself before it caused damage, the plant should be able to tolerate herbicide applications.
During the last decade there has intense work on herbicide tolerance in most of the major global commercial crops and there has been commercial success in this area for those major crops comprising a high percentage of US acreage, such as soybean, maize, cotton and rice. This work often involves a joint venture that may include a seed company, a chemical company and a crop genetics company. The major announced research programs on herbicide resistant crops are:
Imidazolinones
Imidazolinones are low dose, highly effective and environmentally safe herbicides developed by Cyanamid. "Scepterr" is the trade name for one of this family of herbicides that is labelled for use on soybean. Cyanamid worked with Molecular Genetics to expand the use of this herbicide to maize by developing maize hybrids resistant to Scepter. Transgenic potato and cotton varieties resistant to one or more of these herbicides have now been developed.
Glyphosate
Glyphosate is considered a highly effective and environmentally safe herbicide. However, it is non-selective, meaning it kills virtually any type of plant, although the amount of chemical necessary to impart a lethal dose varies from one plant to another. Glyphosate's use, therefore, has been limited to situations where it can be kept out of direct contact with the crop. The herbicide has also been effective in industrial applications such as utility rights-of-way and roadside weed control where crop phytotoxicity is not a factor. Despite these limitations, glyphosate has become one of the largest selling herbicides worldwide. Monsanto, the manufacturer of this product marketed as "Roundup", was the first company to successfully produce glyphosate resistant crops using recombinant DNA technology. In 1987, the company first field tested tomato plants having this trait but have genetically engineered several other major crops since then.
Separately, Calgene, of Davis, California, has also developed plants resistant to glyphosate using recombinant DNA methods. The Calgene approach used a different gene from the one developed by the Monsanto researchers. Calgene has field tested tobacco plants as a model for the resistance. Figure 4 shows the resistance level attainable with these recombinant plants. Calgene is now a subsidiary of Monsanto so the parent company profits from sales of both resistant seed and glyphosate. The company also has the benefit of being able to control the filings for expansion of the glyphosate label to be compatible with the status of development of additional resistant crop varieties. During the last decade, transgenic glyphosate-resistant varieties of most of the major US crops have been developed, field tested and, in a few case marketed, including maize, tobacco, soybean, tomato, cotton and rapeseed.
Atrazine
Atrazine is an older, inexpensive, and off-patent product that has been used widely for broad-leaf weed control in maize and sorghum. Ciba-Geigy of Basel, Switzerland, first developed tobacco plants that exhibit resistance to trace levels of atrazine in the soil. Ciba-Geigy has also developed soybeans resistant to residual atrazine so that soybeans following maize in rotation are not be damaged by residual soil levels of atrazine. This can be viewed as a defensive measure for Ciba-Geigy in that the availability of the resistant soybean varieties will make the use of atrazine a more viable choice for maize growers who might otherwise select a different product. Allelix of Mississauga, Ontario, Canada, in conjunction with the University of Guelph has also developed rapeseed resistant to atrazine. Several atrazine resistant crops have been found to have a yield penalty over non-resistant equivalents. However, the advantages associated with weed control outweigh these disadvantages.
Sulfonylureas
DuPont of Wilmington, Delaware, developed a family of low dose, environmentally safe pesticides in the class of sulfonylureas. Products on the market include "Glean," "Assure", "Classic" and others. Several companies have developed sulfonylurea-resistant crops including maize, potato, tobacco and, especially, cotton.
----------
Figure 4. Glyphosate resistant tobacco plants: the three plants to the left have not been transformed. The three plants to the right are genetically engineered to be resistant to the broad spectrum, non-selective herbicide, Glyphosate. The front two plants have been sprayed heavily with the herbicide. The plants in the middle row were sprayed moderately, the two plants in the back row were not sprayed with any herbicide. (Photo courtesy of Calgene, Inc.)
------------
2,4-D
2,4-D has found special use in cotton despite its reported residual activity and environmental toxicity. Transgenic 2,4-D resistant cotton has been developed by several US companies, including United Agri Products and William Seed.
Bromoxynil
Bromoxynil (BXN) has traditionally been used to control broad-leaved weeds in cereal crops. It has also found a role, however, in transgenic BXN-resistant crops such as cotton, tobacco and potato.
Phosphinothricin
Phosphinothricin and a related analogue bialaphos (marketed as "Basta" by Hoechst AG, Germany) are broad spectrum contact herbicides which are non-toxic to mammals. Maize and soybean are the two major transgenic crops made resistant to this herbicide.
Other herbicide-resistant crops
Cotton is the main transgenic crop, developed by Boswell, made resistant to imidazoles. Monsanto has developed chlorsulfuron-resistant maize, and isoxazole-resistant transgenic soybean has been developed by Limagrain.
During the last decade, there has been increasing awareness that the development of herbicide-tolerant crops has some potentially undesirable environmental consequences. It is quite clear that many of the newer herbicides to which crops are being made resistant are environmentally safer than their traditional counterparts. However, there is the danger that the increased use of a single chemical-based weed control method, at the expense of other weed control measures, may accelerate the development of herbicide-tolerant weeds and hence make the method obsolete. For example, triazine-resistant weeds are common in areas where atrazine has been extensively used. To date, the number of herbicide-tolerant weeds stands at about 180 species worldwide. Already, 26 weeds are known to be resistant to members of the sulfonylurea and imidazolinone classes alone, and in 1996 there was the first report of a glyphosate-tolerant annual ryegrass. The direct transfer of herbicide-tolerant genes from transgenic crops to their wild relatives would speed up this process considerably and in 1997 the first reports of the transfer, by simple outcrossing, of the bar gene, which confers resistance to glufosinate, from rice to red rice, one of its troublesome weeds. Another potentially risky transfer might involve glufosinate-tolerant oilseed rape and Brassica campestris, a weedy relative to which it readily hybridizes.
A second environmentally undesirable aspect of some programs for developing herbicide-tolerant crops is that the herbicide to which the crops are being made resistant are not necessarily harmless. Bromoxynil is known to be toxic to fish and can lead to defects in laboratory animals. Likewise, atrazine and 2,4-D are much more toxic and persistent than the newer herbicides currently available but they are still a focus of interest in some herbicide-tolerance programs as mentioned earlier.
Insect Resistance Traits in Plants
The development of insect resistance has been arguably the second most successful application of agricultural biotechnology so far. Insect control has been a priority target of agricultural biotechnology for several reasons. First, control of insect pests is a major market, amounting to an annual worldwide cost of $3.5 billion in 1995. Each year in the USA, $400 million is spent on lepidopteran pests alone, and the coleopteran pests Colorado potato beetle and corn rootworm cost US farmers $1 billion annually. Even with these expenditures, farmers still report losses of 5-15% and more of their crops to insect pests depending on the crop, the location, the timing of the pesticide application and the degree of insect infestation.
A second factor that has led biotechnology companies to focus on insect control is that substitutes for pesticides are increasingly considered environmentally desirable. Thus, there is pressure from both the consumer and from regulatory agencies to find better solutions to pest problems. Plants with built-in pest resistance will not leave pesticide residues, will eliminate problems of spray drift during application, will not adversely affect beneficial organisms and will not have problems of limited worker re-entry and groundwater or stream contamination.
A third factor is that built-in pest control is perhaps the ultimate application strategy. For the farmer it is convenient and built-in insect resistance eliminates the task and cost of field scouting and the choices inherent with timing of application.
Finally, though this is not proven as yet, with proper resistance management strategies, there is a good chance that genetically engineered plants that carry their own built-in pest control, may at least retard the onset of pest resistance. According to the World Resources Institute, there are more than 400 species of harmful insects resistant to chemical controls.
The delta-endotoxin of the bacterium Bacillus thuringiensis has received the most attention in the insect resistance area. The toxin, for convenience sake, is often simply referred to as "Bt" even though that is only the organism from which it is derived. This toxin is highly specific to certain classes of insects and has been used in its natural form for two decades as a biopesticide. Scientists have cloned the gene from many varieties of Bt and have found there are differences in selective toxicity of the various Bt toxins. Genes isolated from the majority of strains are toxic to lepidopteran insects in their larval stage. Commonly known as caterpillars, these worms are among the most significant agricultural pests on certain crops because they are chewing insects that can devour a plant in a relatively short time. Some species of caterpillars are known to consume as much as 10,000 times their own weight in two weeks. Pests in this category are the tobacco hornworm, the cotton bollworm, the beet armyworm and the gypsy moth. Other varieties of Bt have been found to be selectively toxic to coleoptera, or insects of the beetle family. Diptera, including mosquitoes and certain flies, are susceptible to other strains.
Thus, in one species of bacterium, biotechnology researchers have found a rich lode of insect control potential. Their task has been to exploit the Bt genes in two main ways. Firstly, new bioinsecticides based on genetically modified microbial strains have led to forms with a wider host range and more potent and more stable activity. The farmer, therefore, needs to make fewer applications of these modified Bt bioinsecticides making them cheaper to use than the conventional strains.
Secondly, Bt genes have been transferred directly into crops. Early work in cloning and expressing the toxin gene was done at a number of laboratories in the early 1980s. By 1986, a field test of tobacco expressing the Bt gene was carried out in the U.S. by Rohm & Haas Co. in collaboration with Plant Genetic Systems N.V. of Belgium, whose scientists were the first to report successful expression of the gene. Now there are about a dozen companies developing transgenic Bt crops, ranging from high acreage crops like maize, sorghum, sunflower, oilseed rape, wheat, alfalfa, tobacco, potato, rice and tomato to horticultural crops like apple, eggplant, cranberry and walnut.
Figure 5 illustrates the protection from insect attack afforded to tomato plants by incorporation of the Bt gene. At present, Bt crops grown on a large scale include cotton, maize, and potatoes, which were grown commercially for the first time in 1996. Bt cotton, resistant to bollworm, tobacco budworm and pink bollworm, currently accounts for 12% of US cotton production (1.8 million acres) and is estimated to reach 80-90% within 5 years. Current measured yield increases of Bt cotton compared to non-transgenic varieties are 8-10%, while the yield increases of Bt maize, resistant to the European corn borer, are up by 5-10 bushels per acre
Farmers have to pay a technology fee to grow transgenics, which is currently $32 per acre for Bt cotton and $7.50 per acre for Bt maize. They, therefore, have to predict the potential pest problems in any given year and decide whether the likely level of pest infestation makes it worthwhile to invest in Bt crops.
In the first three decades of use as a conventional biopesticide, there were no field reports of pest resistance to this toxin. More recently, however, two major insect pests have shown resistance in the field and over 10 have shown the potential to develop such resistance in experiments. Companies involved in developing Bt resistant crops have, therefore, been concerned about the long-term stability of their products and have developed field resistance management strategies to slow down the buildup of resistance in target insects. These strategies include rotations, refugia and mosaics. Rotations involve the temporal growth of transgenics and non-transgenics. Refugia are crops of non-transgenics spatially separated from transgenic crops. Mosaics involve the cocultivation of resistant and non-resistant crops in the same fields.
All three strategies are intended to allow the survival and mating of susceptible insects which then have the opportunity to mate with Bt resistants. As the Bt resistance gene in insects is a single recessive one, resistant/susceptible hybrids are susceptible, at least to high selection pressure. These strategies are also enforced by appropriate authorities. The Environmental Protection Agency currently requires that cotton growers plant at least 4% of their crop with non-transgenics untreated with insecticide. In addition, Monsanto requires growers of its Bt seed to abide by agreed resistance management plans as a condition for seed purchase.
It is too early to predict accurately how effective these resistance management methods will be over long times periods under field conditions. However, in the absence of large populations exposed to repeated sprayings of pesticides, there is likely to be a slower selection pressure applied through transgenic pest control compared with that of chemical spraying after field pest populations have risen past the threshold of economic damage.
Although use of toxin genes isolated from various strains of Bt has been the major focus of early research to develop plants resistant to insects, other approaches are emerging to demonstrate that development of insect resistance traits in plants will involve substantially more complex science than the simple incorporation of a particular Bt gene.
Several laboratories have identified other genes active against selected species of insects. For example, researchers at Monsanto have discovered a new insecticidal protein from streptomyces, which function as a cholesterase oxidase and provides an alternative to Bt technology. In another development, the gene for alpha-amylase from wheat has been transferred to rice, and that from bean has been transferred to peas. In both cases protection of stored grains from burrowing insects is obtained by preventing the digestion of starch by the insects. Recently, rice and potatoes have been engineered with snowdrop lectin genes to confer resistance to sucking insects. Researchers at Ciba Geigy have also developed maize with resistance to several insects using lectin genes from other plant species. There are some reports from the UK, however, that ladybirds fed on aphids, which themselves were fed on crops transformed with lectin genes, were adversely affected. This is an example of where entry of biological pesticides into the food chain may kill beneficial insects.
One area of intensive research has been in the development of trypsin inhibitor genes from various plant species. A proteinase inhibitor from potato has already been isolated and tobacco and rice plants expressing the gene have been attained. Both species showed significant resistance to insect pests, such as the pink stem borer in rice. One major advantage of these non-Bt genes is that they naturally occur in plants and are easily destroyed on cooking. This goes part of the way to alleviating some of the concerns by consumers that ‘unnatural’ poisons are being engineered into food crops. In addition, the spectrum of insects controlled is broader when these non-Bt genes are used in combination with Bt genes. For example, Monsanto has reported a synergistic effect on insect killing activity in plants that express both the trypsin inhibitor gene and the toxin gene from Bt. Additional work is expected to enhance the expression of the combined gene strategy. A further benefit is that plants carrying two insect resistance genes of a different fundamental nature are believed to have a strong likelihood of delaying the onset of pest resistance. To this end, workers at Michigan State University have developed transgenic potatoes containing a ‘gene pyramid’ of two insect resistance genes with differing modes of action, namely lectin and trypsin inhibitor.
----------
Figure 5. Tomato plants illustrate protection afforded by an insect resistance gene. The plant on the right is transgenic and contains an insect toxin gene from Bacillus thuringiensis. The plant on the left, stripped by caterpillars, is a control that was not genetically engineered. (Photo courtesy of Monsanto Agricultural Company.)
-----------
Yet another area of recent research focus is modification of the Bt insect toxin gene to enhance its expression in the plant. Several laboratories have reported that modifications to the coding sequence of the gene can result in improved expression of the gene in plant tissue even though the protein sequence remains unaltered. Considerable research has also been undertaken towards the identification of promoters that are selectively active in specific parts of the plant, at specific times in the plant's developmental life cycle, or on being wounded by a feeding insect. For example, for the control of the budworm/bollworm complex of pests in cotton, a toxin gene active only in the boll and only when being attacked by insects could provide an efficient approach to control since those pests confine their attack to the cotton boll.
Agronomic Properties
A considerable effort has gone into the development of crops with traits which allow them to survive and yield in adverse environments, such as those which are too dry, salty or frigid to support normal crop growth. An equally strong effort has gone into causing crops to develop differently within the same environment. Traits which control such characters are usually controlled by many genes, often arranged in linkage groups. This means they are difficult to cross-breed for and equally difficult to genetically engineer. However, a few of these traits have been found to be amenable to biotechnological manipulation, particularly some forms of stress resistance, such as vegetative growth and flowering.
Low temperature tolerance is an important target for biotechnological improvement of plants. Crops engineered in this way can be grown outside areas in which they are normally grown, such as in higher latitudes or higher elevations. In addition, growing seasons in temperate zones can be extended further into the cooler seasons, with less risk of damage to crops from sudden frosts. The first approach to solving this problem used ‘ice-minus’ forms of Pseudomonas syringae developed by Advanced Genetic Systems to prevent ice crystal formation on plant surfaces at temperatures below freezing. First tested on strawberries in 1987, it caused a public outcry because of the possible environmental risks attached to the release of the first genetically engineered microorganism. Since that time, cold tolerance has been successfully engineered directly into many crops using antifreeze proteins from marine fish that live in icy waters, such as wolf fish, winter flounder and sea raven.
Salt, cold and drought resistance have also being engineered into such crops as rice and tobacco using bacterial genes which increase the production of solutes such as betain, inositol, mannitol and sorbitol. These chemicals protect plant cells against the ravages of various stresses by preserving the osmotic balance, by stabilizing protein structure and by protecting the photosynthetic apparatus. Likewise heavy metal tolerance has been considerably increased in some crops, such as tobacco and cabbage, by their transformation using mammalian metallothionein proteins. The transfer of bacterial genes which cause crops like tobacco, papaya, rice and cotton to produce and release excess citrate increases their tolerance to aluminium toxicity and acid soils. Such advances should enable crops to be grown on land which is presently unsuitable, an important consideration when one third of the world’s land surface suffers from soil acidity and aluminium toxicity.
Male sterility is another trait that is of considerable importance to breeding companies involved in the production of F1 hybrid seed. Existing chemical methods of sterilisation are often unreliable. A technique developed by Plant Genetic Systems in Belgium in the late 80’s makes it possible to genetically engineer crops that are male infertile so that two inbred parental lines, one of which is the male infertile female parent, can be securely crossed without the risk of self-pollination.. It is also a method for preventing potential gene transfer through pollen transmission from genetically engineered crops to wild relatives. It is even possible to engineer another gene into the male pollen donor plant ensuring that the resulting F1 hybrids are fertile so that seed is produced and growers can preserve the parent strains for future use. This genetic engineering approach to controlling male sterility is being actively developed in many crops, such as tobacco, maize, sunflower, sorghum, cabbages, lettuce, oilseed rape and tomatoes.
Finally, the aesthetic quality of ornamental plants can be enhanced by molecular methods. The vase life of cut flowers can be substantially increased using genes which counteract the effects of bacterial infection and ethylene production. New colours and patterns are being designed into the flowers of some species. The era of the genetically engineered brick-red petunia and Florigene’s violet/mauve carnation have already arrived. It has been estimated that Calgene Pacific’s collaboration with Suntory of Japan to invest $10 million into engineering the elusive true blue rose could well command a return of up to 10% of the $5 billion annual worldwide rose retail market.
Disease Resistance
Most of the advantages noted before in the section on insect resistance are equally relevant to the provision of incentives for the development of biotechnological approaches for the control of plant disease. Diseases of plants have long been major problems for farmers, but controlling disease loss is much less straightforward than controlling insect damage. Today, disease control poses a vexing problem for the farmer as well as for the companies contemplating products to assist him in choosing ways to control plant pathogens. While the problem is substantial, the best strategy to achieve effective control is not always apparent.
The advantages of built-in disease resistance are in fact analogous to those of built-in insect control:
. Plant disease has been a contributor to yield loss in the production of virtually every crop. Estimates of the extent of loss attributable to plant disease range from almost negligible though chronic yield impairment to total crop loss in occasional dramatic situations. In the U.S., expenditures of farmers to control plant disease are less than the sums spent for the control of weeds or insect pests, but they are nevertheless substantial. In Europe, expenditures for chemical fungicides to control plant disease are proportionally higher than in the U.S. suggesting a higher priority for plant disease control there. Wood Mackenzie & Co. estimate that for 1987 fungicides for plant disease control accounted for 21% of world agrochemical sales or $4.1 billion. This is in contrast to expenditures of $8.6 and $6.1 billion, respectively, for herbicides and insecticides, accounting for 42% and 31% of world agrochemical sales. But these expenditures by farmers do not purchase complete control. Studies of the extent of disease loss in major field crops often cite yield losses on the order of 12-14% due to plant disease even with availability of chemical controls. It is clear that some level of disease infestation in a farmer's fields will be tolerated as long as the pathogen populations do not become large. However, in some situations effective chemical controls are not applied in time to prevent crop damage. In other cases, effective chemicals simply are not available.
. As in the case of insecticides, there is growing pressure from consumers, and hence from government agencies, to limit the use of some currently registered fungicides as a result of environmental and safety concerns. Use of several of the most inexpensive and widely used fungicides have been cancelled in the last five years and several others are under scrutiny for cancellation or for significant use limitations. Concerns of the regulatory agencies range from ecological effects to human safety problems associated with residues on foods or with worker exposure. Unlike herbicides and insecticides, where a few field crops comprise the major use category of those chemicals, fungicides are used primarily on fruits and vegetables. These crops tend by the very nature of their production requirements to be more likely to involve worker exposure risks because of the frequent use of field labour for harvesting. Similarly, in the case of some fresh fruit and vegetables direct exposure of humans to chemical residues is a greater risk due to the absence of processing steps that can assure a safety margin through minimum washing or preparation operations performed by the food processor.
. As in the case of built-in insect control, which offers the advantage of an application method where problems of timing spray operations and effectively delivering the control agent to the pest are eliminated, built-in disease control carries parallel advantages. The presence of the trait inherent in the plant, whether it be incorporated by recombinant DNA methods or methods of tissue culture, affords the assurance that the disease will be controlled without the farmer having to decide that pest levels have risen sufficiently to justify application.
While the above are all incentives that encourage the development of disease resistance traits, the extent to which these traits will equal herbicide resistance and insect resistance in market size remains unclear. Despite the large loss numbers and the imperfect control afforded by current preparations, the benefit of a specific disease control product is likely to be limited. Unlike herbicides and insecticides, for which a few major crops and a few major pests account for large expenditures for agrochemicals, plant disease problems are spread widely among many crops and over many disease pathogens in each crop. The significance, on average, of any one of them is minor in comparison to the significance of, say, an insect such as cotton bollworm to cotton or maize rootworm to maize. Moreover, the incidence of problems is highly variable from year to year and from region to region, and it depends also on weather, crop management practices and soils. Because of this combination of factors, losses accrue to the farmer typically in two ways:
. Major loss of the crop associated with an epidemic outbreak of disease that occurs before adequate chemical controls can be applied.
. Subtle yield loss which, while due to pathogens, are in fact invisible to the farmer and are likely to be attributed to normal field and weather-induced agronomic variability. These circumstance do not lend themselves to the creation of market potential and hence do not encourage product development by companies seeking commercial opportunities. Since farmers cannot anticipate the outbreak of disease problems leading to catastrophic loss, they are unlikely to pay more than an "insurance" value at the beginning of the growing season for the control of a particular disease during that year. Similarly, the farmer is unlikely to pay for the value of a subtle yield loss that he is unable to distinguish from a loss associated with normal agronomic variability.
Despite all these negative factors, disease resistance is in fact a near-term application of biotechnology to agriculture and several products are already being marketed or well along in development. Even though the market sales of any one disease resistant seed product may never rival those of seed products with herbicide or insect resistance, collectively farmers are likely to benefit from their availability. Some companies will probably establish market positions based on niche opportunities afforded by offering effective disease control to growers of certain crops for which no reliable disease control was previously available.
Disease resistance has been an objective of plant breeders for decades and indeed farmer use of disease resistant varieties has been an effective substitute in some situations for the use of chemical controls. The principal advantage the new techniques of biotechnology bring to the development of disease resistant varieties is selectivity and, once the genes and plant transformation capability is in place, speed. The conventional plant breeder's task is painfully slow because the desired disease resistance traits are developed over many generations of plant crosses in a gradual process that relies on artificial selection and natural laws of hereditary segregation. Often the desired traits, when they can be identified, reside in plants that are distant relatives of the commercial crop varieties that have the otherwise desirable yield and quality traits. Once the breeder has crossed the two, he is faced with the time-consuming and tedious task of selecting repeated generations of progeny of the initial cross to eliminate the undesirable traits that he was initially forced to incorporate with the desirable characteristics.
The new methods of biotechnology changed this: recombinant DNA techniques allow the transfer of only the desired trait into the otherwise commercially optimal line. Similarly, techniques of tissue culture, when used in combination with artificial selection at the cellular level, permit scientists to accomplish in weeks what previously took years.
In one early example from the 1980’s, research efforts at Plant Genetics of Davis, California before it merged into Calgene, were directed at developing lettuce resistant to corky root disease using cultures of lettuce cells and artificial selection at the cellular level. Scientists isolated the disease-causing toxin from the microorganism responsible for the disease and applied it to growing cultures of iceberg lettuce cells. While most cells exposed to this artificial selection pressure died, the few remaining cells were regenerated and tested for resistance to corky root disease. In the last decade, researchers at several company and university laboratories have attained transgenic plants resistant to many viral diseases, several important fungal diseases and a few diseases caused by bacterial pathogens. Some of the more significant molecular strategies currently employed for producing transgenic disease-resistant plants are described below.
Virus resistance
Virus resistance has so far been the simplest disease resistance to genetically engineer into crops because several single gene resistance strategies are possible. Such has been the success of engineering crop virus resistance that APHIS approved 2380 field trials of transgenic virus resistant crops at 9467 sites between 1987 and July 1996. Nearly three quarters of these approvals have occurred during the last 2 years of this period. The most important transgenic method for conferring viral resistance to crops during the last decade has been the use of viral coat protein genes. With this approach, a crop is transformed with genes coding for viral coat proteins, along with appropriate attached expression signals recognized by plants. The cells of the crop then produce viral coat proteins which, in a way which is not entirely clear, competitively inhibit the infection and/or spread of any invading native virus. Almost all major staple and vegetable crops have now been developed, field tested and marketed by several public bodies, universities and/or companies using coat protein genes and the method has proved efficient and stable for conferring resistance against specific viruses.
Other methods developed for conferring virus resistance include the use of "satellite RNA" genes, which are only of use in a restricted number of viruses, such as cucumber mosaic virus and tobacco ringspot virus. These viruses contain special RNA sequences that reduce the virulence of the virus. Transformation of plants with genes to produce these specific RNA molecules protects the plant from attack by the viruses. In another approach, ‘antisense’ genes cause the silencing of viral gene expression in plants by causing the production of mRNA which is complementary to the mRNA transcripts of the invading virus. The resulting hybrid duplex RNA formed cannot be translated on the host’s ribosomes so viral spread is ‘neutralised’ by the plant using the transferred antisense genes. A few crops, such as potato and tobacco, have already been engineered with antisense genes against viral coat protein genes by Monsanto and North Carolina State University, respectively.
Other, potentially valuable methods involve interference specifically with the replication of invading viruses. Plants transformed with genes for enzymes such as proteases, replicases and ribosome inactivating proteins confer virus resistance, but further work is required to prove their efficacy in the field.
Finally, a technology quite exciting and unique has been used recently in plants to confer very specific viral resistances. This involves transforming plants with genes that cause the production of antibody proteins against the target viruses. This method is relatively new and has only been applied in a limited way to a few crops. Tobacco has been engineered in this way to resist artichoke mottled crinkle virus and tobacco mosaic virus. ARS recently (Dec 1996) issued an interstate movement notification to APHIS for tobacco resistant to tobacco mosaic virus and TSWV using antibody technology. Further details of "plantibody" production is given in the next section.
Several fears have been expressed recently concerning the potential risks associated with the genetic engineering of virus resistance. These fears include that of a high rate of recombination between the RNA of invading viruses and the mRNA from transgenic plants leading to the creation of new plant pathogenic viruses. Another is that of the crop becoming a problem weed because its numbers would not be checked by one of its natural enemies. A further fear is that the virus resistance genes may become transferred by natural hybridisation to one of its weedy relatives. Such fears led to opposition, and temporary halt, in 1994 to the release of Upjohn Co’s "Freedom II" transgenic squash variety which had been engineered with 2 viral coat protein genes against different cucurbit potyviruses.
Fungal resistance
The second class of plant disease against which genetic engineering is beginning to play a major role is that caused by fungal pathogens. During the last few years, several such approaches have been taken to stem infection in both major and minor crops. An array of genes have been used but the ones which have been the most popular during the last decade have been those coding for the enzymes chitinase and glucanase. These enzymes occur naturally in plants and are used by them to disrupt the cell walls of invading hyphae, which contain chitin and beta 1,3-glucans. The genetically engineered crops become resistant to a wide range of fungal diseases, including phytophthora, verticillium and rhizoctonia in potatoes, alternaria in carrot and tomato, frogeye leaf spot in tobacco and maize, phoma in tobacco, crown rot and powdery mildew in tomato and verticillium in strawberry.
Bacterial resistance
Bacterial disease has been less successfully engineered so far than those diseases caused by viruses and fungi. The genes available for use against these diseases are few in number and are targeted at specific diseases in a few major crops, such as tomato, rice, potato and tobacco. The genes used so far code for enzymes which degrade bacterial cell walls, such as lysozyme from hen egg white, produce antimicrobial cysteine-rich thionins, detoxify bacterial toxins or code for toxin-resistant target enzymes. In most cases, crops transformed with these genes showed enhanced resistance to important pathogens. However, the most fruitful phase of the application of biotechnology to this area of disease control has yet to come but is just around the corner.
Nematode resistance
Genetic engineering for nematode resistance is a relatively unexplored area of disease resistance but there has been some progress on a few crops. So far, tobacco, walnut and sugar beet have been successfully engineered to resist problem nematodes. This work is aimed at reducing the need for chemical treatments and rotation practices and also the use of wild relatives of these crops in breeding programs. One recently developed approach to nematode resistance, as well as resistance to other pathogens, is that of artificially programmed cell death. This strategy involves the expression of genes which cause death of plant cells, under the influence of pathogen-specific promoters. The cells around the infection site in the host plant tissues are then killed on infection, thus preventing further spread of the pathogen.
Animal Vaccines
Much of the current interest in veterinary vaccine research has been spurred by the promise that biotechnology holds for improvements in vaccine safety and efficacy. Several products have been developed and/or marketed over the last decade using various innovations based on genetic manipulation, such as the use of various live viral vectors against viral diseases hitherto difficult to control.
Vaccinia virus, attenuated by engineering to prevent development of the disease, has been a popular choice as a viral vector for carrying suitable immunity-inducing genes to target animals. Using this vector, immunity to several diseases has been induced, including rinderpest, peste d’petite ruminants for small ruminants, such as sheep and goats, Japanese encephalitis for swine and rabies vaccine for raccoons. Disarmed live viruses, other than vaccinia have also been used. For example, swinepox virus has been used to carry pseudorabies vaccine to pigs and fowlpox virus has been developed to carry fowlpox, Newcastle’s disease and Marek’s disease to poultry. Such vectors are very precise, cheap and safe methods for delivering vaccines compared to more traditional methods, such as the use of killed or attenuated viruses.
Other viral carrier systems include that of baculoviruses. These viruses only infect insect cells and have been used to carry vaccines, such as that for bluetongue of sheep, into insect cells which are then used as small-scale ‘vaccine factories’. Likewise, baculovirus systems have been used to produce vaccines in whole silkworm larvae. These insect vaccine factories have proved to be up to 10,000 fold more productive than mammalian cell systems.
A third innovation for the vaccination of farm animals that has been developed over the last decade has been the use of DNA vaccines. These involve the direct intramuscular transfection of non-replicable plasmids containing sequences that code for vaccines. Transient expression of these genes in the tissues of target animals gives rise to immunity. Such a system has been used for the delivery of bovine herpes virus 1 glycoprotein and influenza virus hemagglutinin glycoprotein to chickens. Direct transfection of DNA vaccines reduces the cost associated with the production and purification of recombinant protein vaccines, although they can be used in the production of such proteins in cell cultures.
But development of new biotechnology-based vaccines which offer improved efficacy or safety over current products on the market, and which would provide protection against diseases not controlled by traditional vaccines, are not the only promises offered by applications of biotechnology to veterinary biologicals. Antigenic materials might be developed with or without modification so that the appropriate antibodies can be induced in the animal receiving the vaccination. In parallel, work is also under way to improve delivery systems and to develop better adjuvant technologies.
Vaccine delivery has always been a problem for veterinary vaccinations because of the difficulty and inconvenience of extra animal handling usually associated with their administration. The vaccination of entire herds or flocks is often impractical due to time and labour costs. In the absence of a specific disease threat to a herd there may not be sufficient economic incentive to encourage the livestock producer to vaccinate. Research is currently directed at developing new delivery systems which may offer savings in time and labour as well perhaps as providing safer and more effective inoculation. Some of the approaches which show promise include:
. A ballistic delivery system that would deliver a freeze-dried vaccine in a biodegradable bullet via an air-powered gun that can be fired at some distance from the animal.
. An intra-nasal delivery technology. A few products have already been approved for the market in this area. Side effects are reported to be lower on animals receiving vaccines administered via this route in contrast to conventional ones.
. Aerosol formulations have potential advantages for confined facilities such as those used in poultry rearing and increasingly in swine production. Field research in the poultry area has yielded promising results indicating that vaccine administration by aerosols can give effective immunity and produce minimum side effects.
. Embryo vaccination is being examined for applications in the poultry industry where chicks must be vaccinated within a few days of hatching. In this way, immunity could be conferred to the developing bird before it hatches.
Vaccine adjuvants have traditionally been used to enhance the antigenicity of the vaccine. Practically speaking, typical adjuvants can be characterised as formulation improvements or perhaps as a part of the formulation itself. For example, the vaccine antigens are sometimes adsorbed on minerals such as alum, phosphate or aluminum hydroxide. In other cases, the antigens are emulsified in mineral oil and sold in a water-oil emulsion. Current adjuvant research can be classified into two groups. One is directed at refinement of the current adjuvant technologies so that they are better adapted to the new vaccines being developed by biotechnology. The other employs a novel strategy of combining vaccine antigens with immune system enhancers to increase the efficacy of the vaccine.
Many current adjuvants have been developed for use with traditional whole-cell vaccines but these approaches may not be effective when used with new subunit antigens. Thus, some other work is directed to developing new adjuvant formulations that will improve the efficacy of the subunit vaccines. A separate area of research involves the use of entirely new types of adjuvant materials which themselves have been found to have immune system enhancement capability. For example, the interferons and interleukins have been shown in animal and human clinical studies to heighten the immune system response to disease challenge. Interleukin-2 improves the protection afforded by vaccines in mouse and swine animal models.
Bovine Somatotropin
Bovine somatotropin (BST), often less precisely referred to as "bovine growth hormone", is a naturally occurring protein produced in the pituitary glands of cows; it has been known to scientists for decades. During the 1950s, university researchers learned that administration of BST to lactating dairy cows produced an increase in the amount of milk given. In the 1960s Monsanto established a research program to apply BST technology to improve the efficiency of milk production. Although the performance of BST was promising, that research was eventually discontinued because of the high cost of producing the hormone in large volumes. The advent of recombinant DNA technology has dramatically changed the production economics and allowed the research to proceed far more widely. In 1980, scientists at Genentech and Monsanto succeeded in expressing the BST gene in Escherichia coli. This accomplishment allowed the production of sufficient quantities of the material to conduct more extensive studies of its efficacy, with the prospect of a cost-effective production process now at hand.
The ensuing years of research and testing have demonstrated that the recombinant product works. Its molecular structure is virtually identical to the naturally-occurring form of the growth hormone so that the genetically engineered version of the protein behaves in the animal just like the native product. When administered to mature lactating dairy cows, BST causes a 10-25% increase in milk production while the feed intake of the cows increases only by about 15%. This significant increase in feed efficiency is the principal factor that has company officials and research scientists excited about the value BST could have to the dairy farmer. In tests carried out to date BST has not been found to have any effect on humans or other primates; milk naturally contains trace amounts of BST and its levels in milk from cows treated with recombinant BST are not substantially different from those in milk from untreated animals. In 1993, the FDA finally approved the commercial use of Posilac, Monsanto’s BST, after considerable delay awaiting the outcome of further research to determine effective doses of BST and the collection of long term safety data derived from BST administration to several types of dairy cows in different geographical locations (see Figure 6).
In addition to Monsanto, other companies that have developed BST products for dairy farm use include American Cyanamid, Elanco (the agricultural products division of Ely Lilly), International Minerals & Chemicals and Upjohn. There is obviously considerable competition in this segment of the market. Other transgenic forms of BST, such those from fish and pigs which increase growth rates considerably, are also undergoing research and development by various companies. However, in 1993, Pittman-Moore decided not to pursue the commercialisation of their transgenic porcine BST to avoid the public fallout which dogged that for Posilac.
----------
Figure 6. A research scientist uses a minicomputer to record feed consumption data on cows involved in bovine somatotropin research at Monsanto Company's research farm near St Louis, Missouri. (Photo courtesy of Monsanto Agricultural Company.)
---------
The arguments used against transgenic BST spanned the full gamut of issues from the economics of the small family farm to safety to farm policy. Critics of BST have vociferously asserted that it is a product that neither the farmer nor the consumer wants. Overproduction in the dairy industry, the association of milk with child and infant nutrition and consumer fears of food additives and "unnatural" products have all served to enhance the arguments of activist groups seeking to prevent the introduction of BST as a product.
The reporting of positive results of BST testing and the announcement of considerable research expansion occurred at about the same time that the U.S. Department of Agriculture was initiating a dairy herd buy-out program in 1987. The program was intended to reduce the milk surplus problem by the purchase of cows from active dairy farmers with payments conditional on the non-replacement of the cows once sold. The proximity of these events was not overlooked by those seeking to halt BST technology. A concerted effort was made by the activist groups bent on stopping the technology to label the effort to improve dairy productivity a misallocation of resources. A good part of the debate over BST has involved the relative advantage the new product may confer on the large dairy operation versus the small dairy farmer. Critics of the use of BST have alleged that it will strongly favor the large, sophisticated dairy operator at the expense of the small farmer, thus driving the latter out of farming altogether and forcing an undesirable concentration in the dairy industry. Ultimately, it is suggested, BST and similar biotechnologies will hasten the loss of the small family farm - an institution considered by many to have substantial intrinsic value to the nation.
Many company and university researchers maintain that BST is merely one additional input factor that dairymen consider for use among the many already available: milking machines, dairy herd improvement rationing, improved antibiotics and other management practices. They point to government policy as the principal determinant of supply because the price support level establishes the guarantee that milk produced will have a market at a profitable price. Further, they argue that it is quite possible that dairymen will use BST to reduce the size of their herds by producing the same amount of milk with fewer cows at lower input costs.
Long Term
While the initial product applications of biotechnology to agriculture largely offer improvements that modify existing agricultural practices, the longer term developments will open up entire new commercial ventures utilising different processes to produce food and fibre and using existing food and fibre crops to generate brand new products.
As research with genetic manipulation progresses, and plant and animal systems become better understood, genetic engineering techniques will become obvious mechanisms to enhance the development of agricultural products to create new uses for old crops and permit new crops to become sources for old products. This should steer us away from dependence on oil-based chemicals and reduce the production of surplus food, such as cereals, along with the subsidies inherent to such surplus production.
Our ability to introduce new genes from diverse sources, including bacteria, animals and exotic plants will be complemented by our ability to silence endogenous genes. This will enable us to modify the natural developmental pathways of our crops, such as ripening and senescence, to add vital nutrients to our food and to remove toxins. In addition, our ability to restrict expression of genes to certain tissues, to preferred developmental stages, to specific environmental conditions and stress situations using gene promoters that only work under these conditions will help us to ‘fine tune’ desired characteristics to a level that was not even imagined a decade ago. There will also be rapid advances in the use of crops as biomanufacturing systems to produce expensive commodities more cheaply, such as pharmaceuticals. Our ability to modify single gene traits in plants will soon be overtaken by our need to modify whole metabolic pathways that are controlled by many genes, perhaps linked together. Technologies to achieve this are already being put in place. The overall result will be to significantly increase the yield of crops, rather than just improving the quality or cost of production.
Finally, the technology developed for large-scale commercial crops in well developed countries, will be transferred to cash or staple crops in less developed countries, via international research institutes and public-funded bodies. Crops such as cassava, sweet potato, yams, coconut, bananas, groundnut, cowpea and possibly sorghum and millets, will become much more productive under adverse conditions. The ripening of fruits, such as papayas and bananas, and losses through spoilage of many fruits and vegetables will be controlled. Some of these crops are also likely to be further developed as biomanufacturing systems for the production of pharmaceuticals that do not need to be extracted and purified: they could be used locally without the need for expensive technology and human resources. With techniques such as these, the biotechnological control of diseases of humans and tropical farm animals will improve. A start has already been made with plant- produced vaccines and vaccinia viruses aimed at rinderpest control, as noted previously.
Improved Quality and Processing Traits
Quality traits are a major factor in determining a crop's value as a food commodity. The importance of biotechnology in modifying the quality of plant-derived food is covered in chapter 25 of this volume. For grain crops, relative protein, fibre and oil content, all taken in conjunction with price, are key in the selection of one commodity over another. This is equally true whether the intended use is in animal feeds or for the production of oil, flour or meal for human consumption.
Several genes have already been identified that are associated with specific proteins or oils. Expressing these genes in selected crop plants has involved both recombinant methods and tissue culture techniques. For example, cereals, such as maize have very little of two important amino acids, lysine and tryptophan. These are essential for human and animal growth and their availability in maize would markedly change its value for use as a food or feed commodity. Similarly, legumes, such as soybean, which are high in overall protein content, are deficient in cysteine and methionine, the latter another essential amino acid. Incorporation of these amino acids into soybean would provide a more balanced food source and increase the value of the meal.
Using conventional breeding, numerous attempts have been made to improve the protein content of maize and soybean. Maize which has significantly higher levels of lysine has indeed been developed this way but unfortunately the maize hybrids carrying the high levels of the otherwise deficient amino acids did not yield as well as the standard hybrids. Thus, absent the incentive of a higher price offered by grain elevators, farmers have not adopted use of these lines. Yet clearly before us is the prospect of new techniques of biotechnology leading eventually to cereals that have both high yield and improved levels of the critical amino acids. Likewise, a considerable effort is being made worldwide, such as with the USDA Agricultural Research Service, to genetically engineer an increase in the glutenin content of wheat grains to improve baking quality of the flour.
Similar genes have been identified which improve the oil and protein content of soybean. Calgene has done considerable work on oil content and has developed oilseed rape varieties expressing a recombinant gene regulated by a seed-specific promoter. The promoter was isolated from a highly expressed seed storage protein and fused to a marker gene. The result was a genetic modification that was integrated into all the tissue throughout the plant but expressed only in the seed tissue. The company has used this promoter to control the expression of certain desirable fatty acid biosynthesis genes on which they are now working. Their goal is to develop new oils for the food and speciality industrial oil industry. For example, it is now possible to genetically engineer oils in plants which vary in their fatty acid chain length, have altered position and number of double bonds and modified attached chemical groups. All these characteristics improve the oils for both food and non-food uses. For farmers, this would translate to new markets for growers who produce selected crops such as oilseed rape (Figure 7). Pioneer Hi-Bred has developed sulfur protein-rich soybeans by transformation with brazil nut S-rich proteins. Unfortunately, after a public outcry over the potential allergenicity of brazil nut proteins, the varieties were withdrawn and other sources for such proteins like sunflower are being actively pursued.
During the last few years, several important quality components, besides proteins and oils have been genetically modified in several major crops. These newly altered traits include carbohydrates, pigments, and vitamins. The purpose of changing the composition of crops in this way is to aid in processing, as well as improving the nutritional quality and taste. Potato, maize and tomato have all now been genetically modified for increased carbohydrate content, using different genes involved in carbohydrate biosynthesis. This affects the solids content in tomato and the starch and fat content of potato products, both extremely important factors in processing and nutritional value. Likewise, the flavour, colour and vitamin content of fresh tomatoes has been modified by introduction of genes, along with fruit-specific promoters, for increased solids and carotenoid content. The latter genes have also been transferred to rice. This could well go towards alleviating the vitamin A deficiency which results from eating de-hulled rice grains as a staple food and which causes 250,000 children in SE Asia to become blind each year. Potatoes varieties with altered browning enzymes to prevent discoloration from bruising have also been developed using genes from, amongst others, the larva of the greater wax moth.
Arguably the most significant genetically altered agricultural product to hit the market in the last few years has been that of Calgene’s Flavr Savr tomato. This, in 1994, was the first genetically altered whole food to become available to US consumers. Tomatoes are a difficult crop for industrial harvesting because fruits on the vine ripen at different rates and once ripening has started it accelerates rapidly. This leaves little time for the producer to get them to the supermarket shelves before they over-ripen, soften and rot. Consequently, immature fruits are harvested and allowed to ripen in transit or are induced to ripen in storage, a process which usually leads to a firm, but rather tasteless, product. The Flavr Savr tomato was modified using the antisense gene for the enzyme polygalacturonase, which is involved in fruit softening during the ripening process. This allows the fruits to be left longer on the vine to ripen more fully and evenly into tastier, more attractive fruits, without softening and rotting so readily.
Calgene’s tomato was also the first visible sign that genetically engineered food was ready for marketing on a large scale, after so many years of development, testing and regulatory approvals at a cost of $25 million. Biotechnology had penetrated to the heart of wholesome food, the humble fresh tomato, with its $2.5 billion annual US market. The biotechnology industry had to get it right this first time - so much depended on its acceptance by the public in the face of a critical onslaught from opponents of genetically engineered food. After considerable delay, the FDA finally gave approval and the public were let loose on the new product. In the event, the new product did not suffer from consumer rejection, but was eventually removed from the shelves because of poor fruit yields and disease susceptibility that made it unprofitable.
At about the same time, other companies were developing similar tomato varieties for improved processing, especially increased solids. Zeneca has also developed slow softening tomato varieties similar to Calgene’s, but the purpose in this case was for improved processing, rather than fresh fruit sales. The resulting reduced rate of softening during ripening considerably increases the solids content at the point of processing, reducing overall costs associated with high water content, such as transport and evaporation. The resulting tomato paste derivatives have been a considerable commercial success even in Europe where the public has shown some antipathy towards genetically engineered foods.
The tomato has led the way for other fruits and vegetables to be similarly engineered to improve their storage life, nutritive value or organoleptic qualities. A recent report from Europe shows that the seedless character of the fruits of tobacco and eggplant can be genetically engineered. This paves the way for the incorporation of seedlessness into other, more commercially important, crops.
Improvements in processing technology have already led to considerable attention to non-food uses for crops: they include the production of biodegradable plastics, namely polyhydroxybutyrate (PHB). Genes for this plastic have been transferred and successfully expressed in Arabidopsis thaliana and oilseed rape. Researchers at Monsanto, Zeneca and other companies expect that by 2005, there will be transformed crop plants producing PHB at a commercial level. Agracetus, a subsidiary of W.R. Grace & Co., has also transformed cotton with PHB genes, using fiber-specific promoters. The resulting fibers are stronger, longer, less water absorbent, more heat retentive, resistant to wrinkling and shrinkage, and bind better to dyes. In fact, Monsanto are currently developing designer blue-gene cotton!
Process improvements have also made possible the development of environmentally safe road de-icing materials, and inks for newspaper printing based on the use of soybean oil. These inks have several advantages over petroleum-based varieties in giving better sharpness and depth of colour, better adhesion properties and thus lessened susceptibility to being rubbed off. Because of their vegetable oil base, they are easier to clean up and disposal of wastes is environmentally more sound.
These process improvements are available today and their development is not associated with advances in genetic engineering. However, such process techniques are expensive given the cost of today's alternative feedstocks and, until the price of the conversion drops, are likely to be, adopted only in specialised situations. The application of genetic engineering to plants will greatly enhance the opportunity to reduce the processing cost of these and similar products. As more is learned about the makeup of plants and their genetic control, scientists will find out how to manipulate the plant material so as to optimise the production process.
Similarly, the production economics of many crops will undergo vast change with improved understanding of the physiological and genetic basis of plant systems. As techniques of genetic engineering are applied to these plants, the food value of the commodities produced may be dwarfed by the value they hold as production systems for new materials. Whole new industries are likely to emerge and the farm as we know it today will likely experience substantial change as a new base of production for high value materials that had previously to be extracted or synthesised in industrial facilities. Such new industries might employ genetically modified plants as sources for pharmaceuticals, flavours, fragrances, colouring agents or other industrial products.
The inverse of this situation is a possibility that foods previously produced on the farm might one day be grown economically using cell culture technology without the need for reliance on the vagaries of nature and the extensive land commitment inherent to agricultural production. Scientists at Escagenetics, Inc. have done extensive work on producing a natural vanillin using genetic engineering and tissue culture techniques. The new product would offer a high quality substitute for artificial vanillin and would be usable in some situations where presently only the high cost natural extract is possible. Other researchers have already developed a laboratory method for producing orange juice from cultured juice sacs obtained from orange cells. This approach may never compete with the naturally-produced orange juice either on the basis of taste or cost, yet specific applications may emerge where this way of getting orange juice is indeed a viable cost alternative.
----------
Figure 7. Oilseed rape, an important oil crop in Europe, Canada and Asia, is among the crops scientists are studying to improve oil quality traits for industrial and edible oil applications using recombinant DNA techniques. (Photo courtesy of Calgene, Inc.)
----------
Some university laboratories are seeking genes that will improve the drought and stress tolerance of crop plants. Given predictions of possible changing global weather patterns, the capabilities offered by genetic engineering may provide important tools in adapting agriculture to changing climatic conditions. Even aside from the prospect of change in climate, there are substantial benefits to be gained by the development of crops bearing improved tolerance to too little moisture, too much moisture or unfavorable soil conditions. Drought and flooding are frequent causes of crop losses today. If genetic improvements can be invented that would reduce the yield loss in the event of certain unfavorable weather conditions, farmers would have an important additional tool in dealing with these natural hazards.
Nitrogen Fixation
Improved plants that fix their own nitrogen, thus eliminating the need for application of some or all of t he synthetic fertilisers typically applied today by farmers, has long been considered by many to be one of the primary ways that biotechnology will bring about improved crops for agriculture.
Based on current levels of the scientific understanding of plant systems, this is a long range goal at best and perhaps one that will remain no more than conceptual for many years to come. The genetic basis for nitrogen fixation involves many genes, only some of which are characterised today. The nitrogen fixation process itself is not completely understood and, even if it were, our ability to insert into a recipient plant the necessary genetic components in a fashion that would make their operation likely appears remote. Unfortunately, as the scientific understanding of nitrogen fixation increases, so does the number of genes thought to be associated with it. This makes the task of the genetic engineer considerably more difficult. Beyond this, scientists continue to differ about whether the net effect of incorporated nitrogen fixation genes would ever result in plants with improved efficiency. It has been argued that the increased energy requirements of the nitrogen fixation process would detract from the plant's yield capability and hence offer no net benefit.
Depending on how these questions are resolved, initial nitrogen fixing plants, if they can be produced at all, are likely to have only a partial capability to fix nitrogen so that application of some synthetic fertilisers by the farmer would still be necessary. Some analysts have pointed out that because synthetic fertilisers are among the lowest cost inputs at current energy prices, the economic incentive to the farmer for such a partial reduction of fertiliser costs is likely to be marginal. In the absence of a clearly perceived strong market potential, commitment of funds to research in this area is going to lag behind relative to opportunities of more substantial potential.
Until research has clarified the technical issues and improved our understanding of plant energy utilisation, the incorporation of an ability to fix nitrogen into important crops is likely to remain a long range prospect. However, fertiliser prices are subject to the cost vagaries of petroleum and natural gas feedstocks so the long run prospects for continued low price fertilisers are open to question. Furthermore, the availability of synthetic fertilisers in many of the less developed countries is limited and the existence of self-fertilising crops would have considerable appeal to their governments. It is increasing environmental concerns about fertiliser contamination of surface and ground water that may ultimately serve as the major incentive for commercialisation of this technology.
The alternative to modifying plants to improve nitrogen fixation is to seek improvements in plant-microbe interactions which facilitate nitrogen uptake by the plant. Several companies have programs underway to develop improved strains of rhizobium inoculants to increase yields of legume crops. Rhizobium bacteria have been known for decades to be associated with symbiotic nitrogen fixation of such crops as soybean, alfalfa, peanut and dry beans. The world market for such microbial inoculants in the mid-1980s was about $20-25 million with the U.S. accounting for 1/2 to 2/3 of the total. Research is underway to use recombinant techniques to enhance the nitrogen fixation abilities of these bacteria for legumes and, in the longer run, to find microorganisms that can be used to facilitate nitrogen fixation by non-legume crops. Several rhizobium products have been on the market for many years but are used only infrequently by farmers because yield improvement results are seldom experienced and in any case are difficult to measure. It is widely thought that if rhizobium products were available that demonstrably improved the yield performance of the crop, the market would increase by as much as tenfold. In 1988, Agracetus developed a mutagenized form of rhizobium inoculant for use by U.S. soybean farmers in the mid-south. The company claims the product delivers reliable though modest yield improvements of 5-7% when used in soybean production where indigenous rhizobium is already present. Separately, also in 1988, Biotechnica International of Cambridge, Massachusetts, field-tested in Wisconsin the first recombinant rhizobium on alfalfa (see Figure 8). Biotechnica's rhizobium was genetically modified to increase expression of genes responsible for nitrogen fixation. Similar research has been undertaken at Biotechnica to develop genetically engineered rhizobium for use by soybean growers. Most research results, however, show that genetically engineered rhizobium only leads to significant yield increases under conditions where soil nitrogen is limiting and there is a low endogenous rhizobium population, such as in many developing countries. Chinese scientists have reported consistent yield increases of 5-10% using genetically engineered rhizobium strains.
"Pharming"
The production of pharmaceuticals from plants, or ‘pharming’, is one of the most intriguing applications of plant biotechnology to have been accomplished during the last decade. In 1989, the sales of non-food products from transgenic plants was only $50 million. Consulting Resources Corporation has predicted that sales of such products, a major component of which is likely to consist of pharmaceuticals, will increase at the rate of about $15 million per year to over $320 million by the year 2005. Several genes for human and animal products have been transferred and expressed in plants, including mouse and human interferon, hemoglobin, aequorin (the luminescent protein from jellyfish), metal-binding metallothionein proteins from several mammals, human serum albumin, human growth hormone and human epidermal growth factor.
The most novel application in this area is the production of vaccines and antibodies. Mass immunization is difficult in areas of the world where vaccine production costs are high and where the medical and technical hurdles for administering and distributing currently available vaccines cannot be met. Vaccines produced in local crops would be cheaper, would not need a ‘cold-chain’ of refrigerated transport and storage and would be safer than conventional vaccines. Vaccines can be produced in two main ways. Firstly, the viral genes coding for specific antigens are spliced into the genome of a plant pathogenic virus. Suitable host plants are then infected with the engineered virus which multiply in the plant tissues. The diseased plant tissues can then be extracted to yield large amounts of the vaccine. In this way, several European research teams, notably Agricultural Genetics (Cambs., UK) have used cowpea mosaic virus, in combination with cowpeas (black-eyed peas), to produce effective animal vaccines against foot and mouth disease, HIV-1 and mink enteritis virus. Up to 200 doses of a vaccine can be produced in a single leaf using this method. Further work by various companies on vaccines for hepatitis A and B, Plasmodium sp., the causative agent of malaria, and cold and wart viruses is ongoing using this technique. A second technique for producing vaccines from plants is to splice the genes for the epitope directly into the genome of the plant. Hepatitis B vaccines were first produced by plants in 1992, and since then the technique has been used for producing vaccines against cholera, rabies and Norwalk virus. Feeding the plants to suitable animals has led to effective immunisation, so the technique has tremendous possibilities for the cheap, safe and effective production of vaccines for animals and humans against important diseases. Current interest lies in transforming suitable crops for delivery of the vaccines. Crops such as potatoes, which are unsuitable for use in tropical climates and need to be cooked prior to consumption, thus destroying the vaccine, are being replaced by ones which can be eaten fresh, such as the tomato and banana.
Monoclonal antibodies are currently used as diagnostics and therapeutic agents to fight viral disease and cancer. The production of monoclonal antibodies is currently carried out using cell cultures and is an expensive process, often with unpredictable yields, which limits the availability of the product. Research by the Scripp’s Institute and by Agracetus has led to the production of monoclonal antibodies (‘plantibodies’) in tobacco, soybean and maize. Given the specificity of antibodies, such transgenic plants have other potential properties and uses, such as resistance to disease agents, including viruses and nematodes, the cleanup of contaminated soils and the production of passive immunotherapy in humans and animals. ‘Plantibodies’ produced against Streptococcus mutans, the main cause of dental caries in humans, have also been shown to protect against tooth decay.
Genetically Improved Animals
Most research directed at improving the genetic make-up of animals has been focused on altering livestock, poultry or fish to give better reproductive performance and weight gain. As techniques are perfected, a wide variety of new genetic components may be incorporated into economic animals for production. For example, some research has been done in the area of livestock coat characteristics. This could have important benefits in animal health as well as in improving the value of wool produced by the animals. Improved aquaculture has also received some attention with reports of successful expression of growth hormone genes in commercial species such as trout and catfish. These genes have the effect of speeding the growth rate of the fish to improve production economics. Transfer of growth hormone genes has been accomplished by direct transfer of DNA using a heat shock treatment where fish embryos are exposed to abnormally high temperatures to change their chromosome number.
---------
Figure 8. The first Midwest application of genetically engineered microorganisms was made by Biotechnica Agriculture, Inc., at the Chippewa Agricultural Station, located near Arkansaw, Wisconsin. The company is testing genetically engineered Rhizobium meliloti strains for increased nitrogen fixation in alfalfa.
---------
One non-recombinant DNA approach to animal genetic improvement is the embryonic cloning of animals. Figure 9 illustrates the results of two such cloning experiments. From a business standpoint this method is an extension of the techniques of artificial insemination: exceptionally fast growing or highly productive cows are selected and mated with similarly selected bulls. The embryo derived from such in vitro fertilization is cultured until it reaches the 16 cell stage. At that time it is dissected into single cells and their nuclei removed and transplanted into the egg cells of a second group of cows from which the nuclei had been removed. These re-nucleated cells can be multiplied in the test tube to create whatever number of identical cells are desired. At an appropriate time the cultured cells can be transplanted into the uterus of a third group of cows and grown to full term calves. This third group need not carry any particularly valuable genetic feature. This technique works but involves a lengthy and complex process in order to bring it to completion. As the method is refined it promises to yield virtually unlimited numbers of highly valued, genetically identical animals via the reproductive systems of ordinary cows.
One problem with the use of cultured embryonic cells as a source of improvement genes is that the final characteristics of the adult animal are unknown. A major milestone in the production of cloned animals occurred with the birth of "Dolly" the sheep at the Roslin Institute in the UK, reported in 1997. Dolly was the first animal to be cloned by transferring the nucleus directly from tissues of adult animals to enucleated embryonic cells. As a result, a duplicated adult animal of known characteristics was produced. This technique should prove useful for the replication of a limited number of animals from an elite herd. Caution has to be exercised with this technique, however, as it could lead to a narrowing of the genetic base of farm animals, leading to disease susceptibility.
Further intensive research is currently being conducted into using transgenic animals to produce tissues which may be used in human transplantation, such as heart valves in pigs. In addition, they can be used to manufacture high value therapeutics, such as blood clotting factors required by haemophiliacs or tissue plasminogen activator used in the treatment of heart attack victims, secreted in the milk of transgenic cows, pigs and rabbits.
Genetic engineering also holds promise for another activity of economic importance, the modification of insect pests and their enemies for use in integrated pest management programs. In this application, biological as well as chemical control strategies are employed to prevent the economic loss of crops to the ravages of insects. For example, by suitably changing the appropriate genes of the insects and then releasing large numbers of altered individuals to mate with wild pests, the natural population might effectively be wiped out by the genetic sterilization or poor fitness of the offspring of such matings. Another promising approach is the use of baculoviruses engineered to carry toxin genes to insects. Baculoviruses are very efficient dose delivery system because they carry the toxin gene not the toxin themselves. The toxin gene is expressed only in suitable insect hosts and are non-toxic to mammals. American Cyanamid Co. has recently field tested the baculovirus Autographa californica, which was transformed with genes for an insect-specific toxin protein from the venom of the scorpion Androctonus australis. This transgenic virus can be used to control lepidopteran pests in tobacco, cotton and leafy vegetables. Baculoviruses are also now being engineered to deliver genes for toxins from mites and wasps, as well as diuretic hormone, juvenile hormone and Bt toxin.
CONCLUSION
This chapter has sought to convey a broad overview of how biotechnology is being applied to agriculture. The first part was devoted to a description of the context in which agricultural biotechnology will evolve. This includes the scientifically unique aspects of applying biotechnology to agriculture as well as an overview of the agriculture sector, agribusiness and public policy issues as they affect the farmer and developer of biotechnological products. The section on the content of agricultural biotechnology then described the applications of the new technology to agriculture. Near term products encompass insect, disease and herbicide resistant plant varieties in crop agriculture while in animal agriculture, near term developments include new, more effective vaccines and growth or productivity-enhancing products such as bovine somatotropin. Longer term applications are considerably more diverse and range from crops with improved stress tolerance and new quality and processing traits to genetically engineered animals.
---------------
Figure 9. Two calves, "Fusion" (back) and "Copy" (front) which were the result of two independent experiments designed to demonstrate the techniques of nuclear transplantation or "cloning". This research was performed by Dr N.L. First and colleagues at the University of Wisconsin Madison in association with American Breeders Service (now called ABS Global Inc)(Photo courtesy of American Breeders Service, Inc.)
------------
We certainly cannot today have an accurate view of the long term potential of this technology. Much of the development likely to be really significant will become evident to us only gradually in the next few years as scientific efforts progress and our understanding of plant and animal systems increases in tandem with our ability selectively to incorporate specific genes into plants and animals. The pace of the change is quickening as the seeds of change germinate.
FURTHER READING
Arntzen, C. (1997). High-tech Herbal Medicine: Plant-based Vaccines. Nature
Biotechnology, vol. 15, pp. 221-222.
Atkinson, H.J., Blundy, K.S., Clarke, M.C., Hansen, E., Harper, G., Koritsas, V.,
McPherson, M.J., O’Reilly, D., Scollan, C., Turnbull, S.R., Urwin, P.E. (1994). Novel plant defences against nematodes. In F. Lambertini et al. (eds.). Advances in Molecular Plant Nematology. Plenum Press, New York, pp. 197-210.
Barrett, J.A. (1996). Biotechnology and the Production of Resistant Crops. Sci Total
Environ. vol. 188, suppl. 1, pp. S106-S111.
Bramley, P.M. (1997). Enhancing levels of carotenoids by genetic manipulation. The
Biochemist vol. 19, pp. 16-20.
Chevre, A-M. et al. (1997). Gene flow from transgenic crops. Nature vol. 389, p. 924.
Christou, P. (1996). Transformation Technology. Trends in Plant Sci. vol. 1, pp. 423-
431.
Habben, J.E. and Larkins, B.A. (1995). Genetic modification of seed proteins. Current
Opinion in Biotechnology vol. 6, pp. 171-174.
Hancock, J.F., Grumet, R., Hokanson, S.C. (1996). The Opportunity for Escape of
Engineered Genes from Transgenic Crops. HortScience vol. 31, pp.1080-1085.
Herrera-Estrella, L., Simpson, J. (1995). Genetically engineered resistance to
bacterial and fungal pathogens. World Journal of Microbiology and
Biotechnology vol. 11, 383-392.
Holton, T.A. and Tanaka, Y. (1994). Blue roses - a pigment of our imagination?
TibTech vol. 12, pp. 40-42.
Jorgensen, R.B., Hauser, T., Mikkelsen, T.R., Ostergard, H. (1996). Transfer of
engineered genes from crop to wild plants. Trends Plant Sci, vol. 1, pp. 356-358.
Knauf, V.C. (1995). Transgenic approaches for obtaining new products from plants.
Current opinion in Biotechnology vol. 6, pp. 165-170.
Maeda, S. (1995). Further development of recombinant baculovirus insecticides.
Current Opinion in Biotechnology, vol. 6, pp. 313-319.
Moffat, A.S. (1995). Plants as chemical factories. Science, vol. 268, p. 659.
Moffat, A.S. (1995). Exploring transgenic plants as a new vaccine source. Science,
vol. 268, p. 658.
Paleyanda, R.K. et al. (1997). Transgenic pigs produce functional factor VIII in milk.
Nature Biotechnology vol. 15, pp. 971-975.
Paoletti, M.G. and Pimental, D. (1996). Genetic engineering in agriculture and the
environment. BioScience vol. 46, pp. 665-673.
Pappu, H.R., Niblett, C.L. and Lee, R.F. (1995). Application of recombinant DNA
technology to plant protection: molecular approaches to engineering virus resistance in crop plants. World Journal of Microbiology and Biotechnology Vol. 11, 426-437.
Peferoen, M. (1997). Progress and prospects for field use of Bt genes in crops.
Trends Biotechnol. vol. 15, pp. 173-177.
Postgate, J. (1990). Fixing the nitrogen fixers. New Scientist, Feb. 3, pp. 57-61.
Poirier, Y., Nawrath, C. and Somerville, C. (1995). Production of
polyhydroxyalkanoates, a family of biodegradable plastics and elastomers, in bacteria and plants. Bio/Technology vol. 13, pp. 142-150.
Rotino, G.L., Perri, E., Zottini, M., Sommer, H. and Spena, A. (1997). Genetic
engineering of parthenocarpic plants. Nature Biotech. vol. 15, pp. 1398-1401.
Schertz, L.P. (1979). Farming in the United States. Structure Issues of American
Agriculture, USDA Economic Research Service Agricultural Economic Report No. 438.
Siemens, J. and Schieder, O. (1996). Transgenic plants: genetic transformation -
recent developments and the state of the art. Plant Tissue Culture and Biotechnology vol. 2, pp. 66-75.
Stilwell, D. (1997). Public acceptance of genetically modified foods. The Biochemist
vol. 19, pp. 21-23.
Theologis, A. (1994). Control of ripening. Current Opinion in Biotechnology. vol. 5,
pp. 152-157.
Topfer, R., Martini, N. and Schell, J. (1995). Modification of plant lipid synthesis.
Science vol. 268, pp. 681-686.
USDA (1984) Handbook of Agricultural Charts.
USDA (April 1982). Will midsize farms fade as small and big farms multiply?
Farmline, pp. 4-6.
Whitelam, G.C., Cockburn, W. (1996). Antibody expression in transgenic plants.
Trends in Plant Sci., vol. 1, pp. 268-272.
Wilmut, I., Schnieke, A.E., McWhir, J., Kind, A.J. and Campbell, K.H.S. (1997).
Viable offspring derived from fetal and adult mammalian cells. Nature vol. 385, pp. 810-813.
Yilma, T., Giav Edoni, L., Seliki, J., brown, C., Melius, C. and Jones, L. (1992).
Protection of goats against pest de petites ruminants by a vaccinia virus double recombinant expressing the F and M genes of rinderpest virus. Proc. US Animal Health Assoc. 96:184-189.