Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
D Pesticide Innovation TRENDS IN INNOVATION EARL R. SWANSON The role of innovation in a pesticide regulatory action depends on the scope of the benefit analysis. Neither FIFRA nor the FDC Act (section 408) prescribes in detail the nature of the benefit analysis. For example, there is no legal requirement for a formal benefit-cost analysis and no specification of the future time period to be considered. Benefit assessments performed by the EPA usually focus on short-run economic impacts (three to five years) and consider only currently registered chemical and nonchemical controls as alternatives. There are cases, however, in which the EPA risk/benefit decision process has taken into account pending registrations. For example, in the Rebuttable Presumption Against Registration (RPAR) process on trifluralin (Treflan), the pending registrations of pendimethalin (Prowl) were considered.2 One of the reasons for the focus on short-run impacts in the EPA benefit analyses is the difficulty of forecasting the rates of innovation in pest control methods. Nevertheless, the committee believes that the EPA should give added emphasis in benefit analysis to alternative pest control technologies under development. The methodology for such evaluation, however, is not well developed at present. Ideally, information at each stage of the development of a pesticide would be useful. Although there is considerable firm-to-firm and compound-to-compound variation in the discovery and development process, the stages suggested by Gorings are informative. In terms of sequence, these five components may overlap and some are performed simultaneously: 1. Synthesis, screening, and preliminary field research; 2. Expanded field research, field development, and sales support; 226
PESTICIDE INNOVATION 227 3. Metabolism, environment, residues, and toxicology; 4. Formulation, process, and pilot plant; and 5. Registration. In this appendix, broad perspectives of the changes that are occurring in methods for control of insects and weeds are presented. The partial inventory of compounds undergoing testing presented in Chapter 6 illustrates one type of data that might be used in expanded nsklbenefit analyses. Other sources of information include examination of chemical patents and applications to the EPA for registration. Searches of the trade literature may also provide an indication of particular pest control innovations at venous stages of development. Certain limitations in the data sources should be noted. The field testing done under the auspices of public agencies and reported, for example, in the Fungicide and Nematicide Tests published by the American Phytopathological Society may underestimate the actual level of testing activity for new compounds. Universities and experiment stations are becoming less willing to perform tests on expenmen- tal pesticides, and an increasing amount of such testing is now conducted in the private sector and thus not reported. Nevertheless, the efficacy data on experimental compounds available in the reports of professional associations provide evidence of possible replacements for compounds presently used. Clearly, a systematic methodology needs to be developed for assessing the innovation process at its various stages and integrating such assessments into the benefit analysis. If the EPA were to emphasize the prospects for new pest control technologies in its benefit analyses, such a shift to a wider range of alternatives would decrease the long-run benefits of the pesticide under consideration, but not necessarily the more immediate impacts of its withdrawal. In principle, the broadened scope of benefit analysis would increase the risklbenefit ratio and the probability of cancellation of a registered pesticide or the rejection of the application of an unregistered pesticide. If this expanded benefit analysis by the EPA is perceived by industry to be reasonably stable, pesticide manufacturers may be ex- pected to respond by increasing production of registered substitutes and/or developing new pesticides for a changed market. NOTES 1. 7 USC § 136 (1978) and 21 USC § 346(b) (1984). 2. U.S. Environmental Protection Agency. 1982. Trifluralin (Treflan). Position Document 4. Office of Pesticides and Toxic Substances. Washington, D.C., pp. 59~0. 3. Goring, C.A.I. 1977. The costs of commercializing pesticides. Pp. 1-33 in Pesticide Management and Insect Resistance, David L. Watson and A.W.A. Brown, eds. New York: Academic Press.
228 APPENDIX D HERBICIDES WEED CONTROL FRED H. TSCHIRLEY During the past 15 years, the use of herbicides on crops in the United States has increased dramatically. Farm use of herbicides totaled 215 million pounds in 1971, 376 million pounds in 1976, 445 million pounds in 1982, and 435 million pounds in 1984.' Herbicides now account for about 65 percent of the total pesticide use on farms. This increase occurred because their use produces an economic benefit for growers. Although cultivation is still practiced for the control of weeds, and crop rotation provides some weed control, synthetic organic herbicides have become the predominant technology. Led by the discovery of the herbicidal properties of the phenoxy alkanoic acids in the early 1940s, chemistry soon followed that provided different mechanisms of action, a wider range of herbicidal activity on weeds, and differing selectivities to crops. Modern herbicides represent a large number of chemical classes, many of which have only one or a few herbicides in the entire class. Important classes include the phenoxy alkanoic acids, s-triazines, substituted amides, carbamates and thiocarbamates, substituted ureas, and nitroanilines. Herbicides in other classes are also important, including amitrole, paraquat and diquat, bensulide, chloramben, DCPA, endothall, picloram, and nitrogen. Herbicides used for weed control on corn and soybean crops, which represent 93 percent of the farm use of herbicides, are listed in Table D-1. Certainly, the past rate of increase of use will not continue. In fact, there are indications that use has already leveled off. Ninety-three percent or more of the acreage planted to corn, soybeans, cotton, peanuts, and rice was treated with herbicides in 1982. In addition, 71 percent of the tobacco acreage; 59 percent of grain sorghum; and 40 to 45 percent of the wheat, barley, and oat acreage was treated with herbicides. Although marked increases in herbicide usage are not expected in the foreseeable future, neither is a marked decrease expected, and herbicides will surely continue to be the predominant technology for weed control. NEW CHEMISTRY Manufacturers have become more sophisticated in designing new molecules with a reasonable expectation that they will have herbicidal activity. Researchers can target a specific enzyme system that is known to
PESTICIDE INNOVATION 229 TABLE D-1 Herbicidal Active Ingredients Used on Corn and Soybeans During 1982 Active Ingredient (million lbs.) Herbicide Corn Soybeans Single applications Acifluorfen 0.9 Alachlor 19.7 10.3 Atrazine 22.4 Bentazon 6.7 Butylate 22.4 Chloramben 2.7 Cyanazine 4.9 2,4-D 3.3 Dicamba 0.9 Fluchloralin 2.6 Glyphosate 2.2 Linuron 1.3 Metolachlor 3.2 6.9 Metribuzin 2.2 Trifluralin 20.4 Other 9.5 5.5 Total 86.3 61.7 Tank mixes Acifluorfen + bentazon 0.3 + 0.7 Alachlor + metribuzin 6.9 + 1.7 Alachlor + linuron 8.1 + 3.2 Alachlor + naptalam + dinoseb 1.5 + 1.3 + 0.6 Atrazine + alachlor 16.4 + 21.2 Atrazine + butylate 8.7 + 23.7 Atrazine + cyanazine 2.7 + 3.6 Atrazine + metolachlor 8.7 + 10.7 Atrazine + simazine 1.3 + 1.2 Bentazon + 2,4-D 0.4 + * Chloramben + alachlor 1.5 + 1.8 Chloramben + trifluralin 0.9 + 0.5 Cyanazine + alachlor 6.1 + 7.6 Cyanazine + butylate 2.7 + 4.9 Cyanazine + metolachlor 0.9 + 1.2 Dicamba + 2,4-D 1.0 + 1.6 Dinoseb + naptalam 1.2 + 2.4 Metolachlor + metribuzin 4.2 + 1.6 Metolachlor + atrazine + simazine 3.2 + 2.6 + 1.3 Metolachlor + cyanazine + atrazine 1.4 + 0.6 + 0.6 Oryzalin + linuron 0.4 + 0.3 Paraquat + others 0.4 + 1.7 Trifluralin + metribuzin 8.8 + 3.8 Other 8.9 11.1 Total 142.8 65.3 Total herbicides 229. 1 127.0 *Less than 100,000 pounds. SOURCE: Adapted from Delvo, H. W. November 1984. Inputs: Outlook and Situation Report. Washington, D.C.: U.S. Department of Agriculture, Economic Research Service.
230 APPENDIX D be affected by one or more functional groups. Unfortunately, mechanisms of action are not completely known for all active ingredients. For example, the mechanism of action of the phenoxy herbicides is still unknown, even after 40 years of use. Nevertheless, the discovery of new herbicides has a firmer scientific base today than 10 years ago. Several compounds representing new chemistry have been commer- cially introduced in the past several years, and others, now being tested under experimental permits, can be expected to reach commercial use in the next few years. An exciting aspect of this new chemistry is the remarkably low rates needed for weed control. Older herbicides were applied in pounds per acre; some of the new materials are effective at ounces per acre. For example, control of annual and perennial grass weeds is accomplished with 4 to 8 ounces of fluazifop per acre, 3 to 7.5 ounces of sethoxydim per acre, 1 to 5 ounces of sulfometuron methyl per acre, and 0.17 to 0.5 ounce of chlorsulfuron per acre. Such herbicidal activity is remarkable. One-sixth of an ounce per acre is only 0.09 mg per square foot. Ten or more other herbicides for which rates of fractions of an ounce or a few ounces per acre are needed are in various stages of development. Moreover, they are being developed by several manufacturers, and their chemistry varies, rather than being mere analogs of one basic molecule. An increase in the use of the potent (low-application-rate) herbicides would significantly decrease the quantity of herbicides being applied, and presumably, lower residues in raw agricultural commodities. At present, the crops on which these potent materials are registered is limited. Chlorsulfuron is registered only on wheat, spring oats, and barley; fluazifop on cotton and soybeans; and sethoxydim on soybeans, cotton, sugar beets, and nonbearing food crops. Sulfometuron methyl is not yet registered on any crops. Thus, registration of these herbicides is required on a far greater number of crops before herbicide use will significantly decrease. Herbicidal activity at such low rates requires cautious ap- praisal, however. If a material with high biological activity is resistant to degradation, its use would have to be limited to avoid carryover damage to other crops. In fact, carryover potential for the new classes of soybean herbicides is a matter of growing concern for weed scientists. BIOLOGICAL CONTROL Weed control by insects has been studied by a few scientists for a long time, and successful control has been accomplished for numerous weeds occurring in noncrop areas. However, it has not been successful in cropland, because crops are planted in fallowed land, which is ideal for the germination of weed seeds, phytophagous insects must have a specific host or a narrow host range so that weeds are destroyed without danger
PESTICIDE INNOVATION 23 ~ to crops, and there is usually a complex of weed species in cultivated crops, so controlling one will simply provide a,competitive edge to the remaining weeds. At best, the control of weeds by phytophagous insects might be feasible in perennial crops, such as orchards, or for particularly troublesome weeds, such as nutgrass (Cypercus sp.) or downy bromegrass (Bromus tectorum). Nutgrass and other sedges, however, are important beneficial plants in other habitats, and downy bromegrass provides forage for animals on rangelands.2 Recently, interest in plant pathogens to control weeds has increased, and several notable successes have occurred. Northern jointvetch in rice can be controlled with an endemic fungal disease,3 and milkweed vine in citrus is now controlled by a pathotype of Phytophthora.4 More recently, Walkers~8 reported the successful control by pathogens of spurred anode, prickly side, velvetleaf, and sicklepod. For this technology, spores are pro- duced in the laboratory, incorporated into an appropriate carrier, and then distributed in a selected area at the appropriate time. Combining spores of different fungi, Boyette and coworkers9 applied pathogens for the simulta- neous control of winged waterprimrose and northern jointvetch. The limited number of scientists pursuing research in this field may impede its rapid advance. Control by pathogens has the promise of contributing to the development of integrated weed control systems. Further success requires the discovery of more pathogens so that weed complexes can be controlled rather than just a single species. Moreover, for sustained success, farmers must be weaned away from the synthetic organic herbicides that ensure effective weed control. In a similar vein, increased emphasis has recently been given to natural phytotoxins from pathogens that might be formulated and applied to weeds. This bypasses the problem of introducing a living organism into the environment, which, through mutation, could persist and become destructive rather than beneficial. There is no assurance, however, that a natural phytotoxin would be any less hazardous to human health and the environment than the synthetic molecules now in use. ALLELOPATHY Allelopathy, coined by Hans Molisch in 1937, refers to the release of chemical inhibitors by certain plants, which adversely affect other plant species. Specific cases of allelopathy have been observed in crops, forests, grasslands, deserts, and even aquatic systems.~° The inhibiting chemicals may be released from living plants via exudation from roots, from litter on the soil surface, or from decomposing organic matter. Although, theoretically, allelopathy seems to offer a direct impact on weed control technology, the greatest benefits may come from spin-offs of allelopathic research. Although genotypes of some crops, such as cucum-
232 APPENDIX D her, inhibited some important weeds in the laboratory and greenhouse, the results were less dramatic and consistent in the field, perhaps because the concentrations of the allelopathic chemicals in the soil were too low to inhibit the weeds. Allelopathy could be effective in crops such as turfgrass, cereal grains, and forage legumes, because of a higher concen- tration and more even distribution of the inhibitory chemicals. Develop- ing the technique requires the identification of allelopathic properties and their incorporation into crops. Once allelopathic chemicals are identified, they might be synthesized as herbicides. That route engenders the same problems that now beset organic herbicides synthesized de nova. As with phytotoxins, natural products may be no less hazardous to humans and the environment than ones first synthesized by man. GENETIC ENGINEERING Conceivably, crop varieties could be developed for allelopathic control of weeds. For example, Putnam~° reported that some wild progenitors of modern crop varieties demonstrate greater allelopathy than the varieties now in use. Attention has also been given to breeding varieties that have greater tolerance for herbicides, so that rates to control weeds can be used without endangering the crop. Incorporation of herbicide resistance in the crop has been achieved in three ways:'2 1. By the transfer of a metabolic detoxification mechanism (in which an enzyme inactivates the herbicide) from a resistant plant to a susceptible one. A good example is the herbicide atrazine, which is used widely in corn. Weeds lack the rapid detoxification pathway of corn that replaces the chlorine atom with a peptide via a conjugation reaction. Several laboratories have shown that the enzyme is glutathione-S-transferase. In principle, it should be possible to transfer the glutathione-S-transferase gene into, for example, soybeans, to make it herbicide resistant. Research is still needed, however, before application is practical. 2. By the transfer of a restricted uptake or translocation trait. A new plant variety has emerged in Egypt that is resistant to paraquat. The phytotoxicity of paraquat results from its chemical reductions in the chloroplasts, which generates free radicals that destroy the plant. In the Egyptian biotype, an unknown process restricts the paraquat to the veins of the leaves, preventing it from entering the cells that contain the chloroplasts. Today, however, the probability of genetically transferring this sort of trait from one crop to another is low. 3. By modification of the target of the herbicide. In the short term, target site modification looks promising. A herbicide translocated to a
PESTICIDE INNOVATION 233 specific target in the plant interacts with that target, blocking some metabolic event and killing the plant. If, through genetic manipulation, the target site could be altered so that it no longer recognizes the herbicide, the plant would be resistant. An example is pigweed, which is resistant to atrazine because a natural mutation occurred that changed the protein that normally binds the atrazine. The protein in susceptible plants contains the amino acid serine, which is required for hydrogen bonding of the triazine molecule to the protein. In resistant plants, this amino acid had been replaced by glycine, with which triazine cannot bond. This mechanism of resistance has been exploited to develop a triazine-resistant tobacco plant. Another example comes from scientists of Calgene, Inc., who incorpo- rated a mutant EPSP synthase gene, isolated from glyphosate-resistant Salmonella, into tobacco. Other scientists from Monsanto Chemical Company achieved greater glyphosate resistance in petunia plants by inducing them to make 20 to 40 times the usual amount of normal petunia EPSP synthase. DuPont researchers used both chemical and random mutations to isolate mutant plants that varied in response to chlorosulfuron and sulfometuron methyl. Various tests and correlations established the site of action as acetolacetate synthase. Production of an insensitive form of the enzyme is the basis for resistance. CONCLUSIONS Since their introduction in the early 1940s, synthetic organic herbicides have dominated weed control. Although alternative weed control tech- nologies hold some promise and may become more important, synthetic organic herbicides seem certain to be the preferred technology until the end of the century. Development of alternative technologies will require not only time and research, but also practical demonstrations to convince farmers that the alternatives will be as economical and dependable as synthetic organic herbicides. NOTES 1. Delvo, H. W. November 1984. Inputs: Outlook and Situation Report. Washington, D.C.: Department of Agriculture, Economic Research Service. 2. Morrow, L. A., and P. W. Stahlman. 1984. The history and distribution of downy brome (Bromus tectorum) in North America. Weed Sci. 32(Suppl. 1):2- 6. 3. Daniel, J. T., G. E. Templeton, R. J. Smith, Jr., and W. T. Fox. 1973. Biological control of northern jointvetch in rice with an endemic fungal disease. Weed Sci. 2 1 :303-307.
234 APPENDIX D 4. Ridings, W. H., D. J. Mitchell, C. J. Shoulties, and N. E. El-Ghell. 1976. Biological control of milkweed vine in Florida citrus groves with a pathotype of Phytophthora citrophthora. Pp. 22~240 in T. E. Freeman, ea., Proceedings of the IV International Symposium on Biological Control of Weeds. 5. Walker, H. L. 1980. Alternaria macrospora as a potential biocontrol agent for spurred anode: Production of spores for field studies. U.S. Science Education Administration, Advanced Agricultural Technology, South. Ser. (ISSN 0193-3728), No. 12, 5 pp. 6. Walker, H. L. 1981. Granular formulation of Alternaria macrospora for control of spurred anode (Anoda cristata). Weed Sci. 29:342-345. 7. Walker, H. L. 1981. Fusarium lateritium: A pathogen of spurred anode (Anoda cristata), prickly side (Sida spinosa), and velvetleaf (Abutilon theophrasti). Weed Sci. 29:629~31. 8. Walker, H. L., and J. A. Riley. 1982. Evaluation of Alternaria cassiae for the biocontrol of sicklepod (Cassia obtusifolia). Weed Sci. 30(6):651~54. 9. Boyette, C. D., G. E. Templeton, and R. J. Smith, Jr. 1979. Control of winged waterprimrose (Jussiaea decurrens) and northern jointvetch (Aeschynomene virginica) with fungal pathogens. Weed Sci. 27:497-501. 10. Putnam, A. 1983. Allelopathic chemicals: Nature's herbicides in action. Chem. & Eng. News April 4, 1983, pp. 34 ~ 5. . Marx, J. L. 1985. Plant gene transfer becomes a fertile field. Science 230(4730): 1148-1150. 12. Chemical and Engineering News, October 29, 1984, p. 16. INSECT CONTROL T. ROY FUKUTO INTRODUCTION Because they are effective, economical, and fast-acting, insecticides and acaricides are unique tools for relegating damaging insect and mite populations to subeconomic levels. Thus, despite problems such as the development of insecticide-resistant pest populations and undesirable nontarget effects, they will remain one of the basic tools for managing insects and mites in crops. Virtually all major insecticides that are widely used on crops, except organochlorines, are acute neurotoxins and fall into the chemical classes of organophosphates, carbamates, and pyrethroids. Owing to their per- sistence in the environment and unfavorable toxicological properties, most of the organochlorine insecticides either have been banned or are used only in special situations. Pyrethroids are now receiving the greatest attention from industry. These broad-spectrum insecticides are highly effective at application rates measured in ounces and fractions of an ounce instead of the 0.5 to 2 pounds applied per acre of most compounds in the other classes.
PESTICIDE INNOVATION 235 Advances in insect physiology, toxicology, and analytical chemistry are responsible for discoveries of new compounds with novel modes of action that disrupt the normal growth of insects. The juvenile hormone analogs, for example, prevent the insect from molting to the adult stage. Unfortunately, because the larval stages typically are most damaging to crops, these compounds appear to have limited use in crop protection. They will, however, control such insects as fleas and biting flies, which are pests in the adult stage. Antijuvenile hormones causing insects to molt prematurely to adults have been discovered and offer more promise for managing agricultural insect pests. The relatively recent discovery of compounds that disrupt the molting process of insects by interfering with the synthesis and deposition of chitin (a principal component of the exoskeleton of insects) also holds promise. One such chitin inhibitor, Diflubenzuron (Dimilin) is registered for control of cotton boll weevils and gypsy moths. Similarly, advances in natural products chemistry and the study of plant defenses against insects are leading to the identification of numer- ous, naturally occurring, insecticidal and acaricidal compounds with novel modes of action. To date, biologically active, natural products have been looked to by the agrochemical industry as leads for the chemical synthesis of structurally related compounds with improved biological and physical properties that are amenable to large-scale chemical synthesis. This latter requirement may ultimately become less important with advances in genetic engineering, since even complex molecules can be produced on a large scale, using fermentation processes with genetically . . . englneerec . microorganisms. NEW CHEMISTRY Motivation for the discovery of new pest control agents by the chemical industry originates from the ongoing desire to develop a proprietary agent with superior pesticidal activity and favorable environmental and toxico- logical properties. Although a significant amount of effort is still being devoted to research on organophosphates, carbamates, and pyrethroids, the chemical industry is turning to other classes of compounds in seeking new control agents. Increased attention to unconventional chemicals has been hastened by the prospect of the development of insect resistance to present-day insecticides. During the past decade, novel insecticides with different modes of action have been discovered. With the elucidation of their modes of action, the way has been paved for further search within these classes for new insect control agents. Areas that have been or are currently being explored for new insect control agents are described below.
236 APPENDIX D Octopamine Agonists Chlordimeform (Fundal, Galecron) and amitraz are formamidine deriv- atives that effectively control phytophagus mites, ticks, and a limited range of insects, for example, many Lepidoptera, Hemiptera, and some Homoptera.2 The formamidine derivatives are most effective as ovicides although they are also toxic to nymphs and adults. In addition to mortality, the formamidines also cause unusual behavioral effects, for example, on locomotion, flight, dispersal, and oviposition. Due to adverse human health effects, chlordimeform is registered for use only on cotton. Evidence accumulated over the past decade supports an octopamino- mimetic mechanism of action for chlordimeform and related compounds. The elucidation of the mechanism of action of this compound has stimulated work on the design and synthesis of compounds with octopaminomimetic activity. Octopamine, a biogenic amine that serves as neurotransmitter and neuromodulator, is found primarily in invertebrates and, therefore, compounds mimicking its action are expected to be selectively toxic to insects and acarines. Avermectins and Milbemycines The avermectins and milbemycins are natural products obtained by fermentation of the soil fungus species Streptomyces, which have dem- onstrated potent anthelmintic, acaricidal, and insecticidal activities.34 For example, the avermectins are highly effective against common veterinary ectoparasites, phytophagus mites, nematodes, and various insect species of Lepidoptera, Coleoptera, and Homoptera. They are highly complex molecules consisting of eight major components. Ivermectin, a commercial product currently under development, is a hydrogenated derivative of avermectin B~, the most active of the eight components. The avermectins behave as agonists or cause the release of the inhibitory neurotransmitter~y-aminobutyric acid (GABA). Avermectins and milbemycins exhibit unusually potent insecticidal and acaricidal activities, but have highly complicated structures. Therefore, work has been started on the synthesis and evaluation of analogs of less structural complexity.S 6 This work is expected to result in new analogs with similar modes of action. A new class of insecticide, the 1,4-disubstituted-2,6,7-trioxabicyclo- t2,2,2]octanes, has recently been discovered.7 These compounds appear to act at the neuromuscular junction by inhibiting GABAergic synaptic transmission, possibly by closing off chloride channels. The high insecti- cidal potency of the avermectins, milbemycins, and trioxabicyclof2,2,21-
PESTICIDE INNOVATION 237 octanes warrants further exploitation of the GABA system for new insecticides. Am id i n o by d raz o n es New amidinohydrazone insecticides have recently been registered under the trade names Amdro for control of imported red fire ants and Combat and Maxforce for use against cockroaches. The compound is a slow-acting stomach insecticide whose mode of action appears to be inhibition of electron transport and oxidative phosphorylation.8 The amidinohydrazones represent a novel structure for an insecticide, and other compounds of similar mode of action are presently being sought. However, Amdro is unstable in light and therefore is not useful for agricultural purposes. Benzoylphenylureas The benzoylphenylureas represent a new class of insecticides that are effective as larvicides and ovicides by either contact or as a stomach poison.9 The most prominent of the benzoylphenylureas is diflubenzuron (Dimilin) and BAY SIR 8514. As indicated earlier, diflubenzuron inhibits chitin synthesis and is widely regarded as an insect growth regulator.~° Because of chitin-inhibiting action, diflubenzuron and related analogs should affect insects in all cuticle-forming stages. In view of the favorable selectivity and effectiveness of diflubenzuron, a great deal of interest has developed in the search for other insect growth regulators. Pyrethroids Owing to the outstanding insecticidal properties of the synthetic pyrethroids, much effort is still being devoted to the synthesis and evaluation of new compounds. Since the stereochemical structure of the pyrethroid molecule has such a profound effect on insecticidal activity, work is being conducted on the stereospecific synthesis of the most active enantiomers. Research on new pyrethroids is expected to continue, and new compounds are likely to be developed in the future. Proinsecticides During the past decade, new, less toxic derivatives of methylcarbamate insecticides have been discovered, and several are now close to commercial use.~2 These carbamate derivatives are called procarbamates; that is, they are carbamate derivatives that must be transformed back to the original
238 APPENDIX D carbamate by a plant, animal, or microorganism in order for intoxication to occur. Thiodicarb (a derivative of the carbamate methomyl) and carbosulfan (a derivative of the carbamate carbofuran) are examples of procarbamate insecticides that have attained commercial importance. Both are highly effective insecticides and are substantially less acutely toxic to mammals than the parent carbamates. Several other procarbamate insecticides are currently undergoing commercial development. Nereistoxin, a substance found in a poisonous marine annelid, has been derivatized to form another type of proinsecticide. Nereistoxin paralyzes insects by a blocking action on the central nervous system. Examples of nereistoxin proinsecticides are cartap and bensultap. Bensultap, a more recent discovery, has shown excellent effectiveness against the Colorado potato beetle and different lepidopterous larvae.'3 Natural Products Much effort is being devoted to the study of various plant products that could be used for protection against plant-feeding insects. ~4 For example, pellitorine, a potent insecticidal amide recently isolated from the root of a compositae, has stimulated the synthesis and examination of structural analogs. ~5 Pellitorine, although highly insecticidal, unfortunately is unsta- ble in a field environment. Other types of plants being sought as control agents are insect growth and behavior regulators, morphogenetic agents, insect juvenile hormones and phytochemical analogs, antijuvenile hormones, sex and alarm pheromones, and antifeedants.'4 The examination of plant products for antifeedant com- pounds has recently attracted much attention.'6 A number of plants are recognized for their elaborate chemical defense systems against phytophag- ous insects, and various naturally occurring compounds are being discovered that permanently impede feeding by specific insects. In general, natural products occurring in plants, animals, and microorganisms provide a rich source for new types of insect control agents. Although synthetic organic chemicals remain the principal pest control materials, other types of control agents or methods are currently in use or have the potential to provide effective pest control. They may be divided into three major categories biological control, natural products ap- proach, and plant modification. These are briefly described below. GENETICALLY ENGINEERED MICROORGANISMS Strategies and methods have been proposed for the use of microorga- nisms for pest control. Among the microorganisms showing promise are bacteria, viruses, and fungi. The potential for microorganisms as pesti-
PESTICIDE INNOVATION 239 cides has been increased by progress in genetic engineering, which is expected be important in the development of bacterial, viral, and fungal pesticides effective enough to displace the synthetic chemicals, which dominate the market today.~7 Bacterial Insecticides The sporeforming bacteria Bacillus thuringiensus kurstaki has been developed commercially and is registered by the EPA as a bacterial insecticide for use on field and vegetable crops, trees, ornamentals, and stored products (primarily grain and grain products) to control lepidopter- ous larvae. However, the bacteria's effectiveness is limited to certain species of Lepidoptera. Monsanto is attempting to engineer a microbial pesticide by taking the b-endotoxin gene from B. thuringiensis kurstaki and placing this toxin gene in another kind of bacteria, for example, Pseudomonas puorescens, that can colonize the roots of plants such as corn.~9 When root-eating pests ingest the genetically engineered bacteria on the plant roots, the toxin in the bacteria will get into the gut of the pests where it will be activated and will intoxicate them. Unfortunately, agricultural pests that are vulnerable to this microbial pesticide are still mainly lepidopterous species (tobacco hornworms, black cutworms, cabbage and soybean loopers, and corn earworms) that do not attack plant roots. Discovery of other B. thuringiensis isolates producing proteins toxic to root-feeding species would be required for this particular strategy to work. However, the same general strategy might work using genes from presently available B. thuringiensis strains and bacteria that colonize plant foliage. Monsanto reasons that since these engineered strains are not toxic to beneficial insects such as honeybees, their genetically engineered bacterial pesticides will have the same attributes. There is under way considerable research directed toward identifying strains of B. thuringiensis that produce more virulent toxins and are effective against a wider diversity of insect pests. Research of this type has already led to the commercial development of B. thuringiensis var. israelensis, which is an effective control agent for mosquito larvae and will very likely expand the spectrum of crop pests that can be controlled by bacterial insecticides. RESISTANCE Resistance is a preadaptive phenomenon, and since insects and bacteria have been together in nature for ages, it is conceivable that low levels of resistance to the bacterial toxins already exist.
240 APPENDIX D Evidence has not been provided for the development of insect resis- tance to B. thuringiensis in the field although a recent report has demonstrated that resistance to B. thuringiensis could be selected for in the laboratory. In a microbial control program, development of pest resistance can be averted by the following strategies: · Use of multifunctional agents (B. thuringiensis produces several toxins), because the greater the number of targets in the insect the less likely it is that mutations will lead to increased resistance; · Simultaneous use of chemical and microbial agents for the same reason more than one target is involved; and · Use of an agentmicrobiological or chemical with a rapid toxic action, to avoid a lasting selection pressure for resistance since the number of mutants produced will be proportional to exposure time. However, in a stable environment such as in stored grains where the bacterial toxin is stable, the insect can breed for successive generations in contact with B. thuringiensis. In this situation, resistance is very likely to develop. This scenario has recently been observed with the Indian meal moth Plodia interpunctella, which developed a 100-fold increase in resistance after 15 generations on diets treated with bacterial toxin.20 In this case, the resistance was inherited as a recessive trait. Fortunately, in field crop situations, the instability of foliarly applied B. thuringiensis and the transitory nature of plant pest interactions decrease the possibility of resistance. The use of B. thuringiensis over a wide geographic area for several years would be required to expose the pests for many successive generations. PRODUCT NAMES AND USES A number of biological insecticides exist on the market that have B. thuringiensis as their active ingredient. These include · Thuricide, having B. thuringiensis Berliner as the active species; · Thuricide-HP, also derived from B. thuringiensis Berliner. However, unlike Thuricide, it is twice as concentrated; and · Bactospeine, Javelin, and Dipel all contain B. thuringiensis Berliner var. kurstaki as their active ingredient. However, with regard to concen- tration, the ratio of active ingredients among the three is 1:2:4, re- spectively. These formulations of B . thuringiensis are active over a broad range of lepidopterous pests in a vast array of crops, including vegetables, cotton, and various fruits. Among the disadvantages, however, is the slow killing
PES TI CIDE INNO VA TI ON 24 ~ action that allows more damage before death. These materials are also less toxic to large worms. MUTATIONS Dangerous mutations may be of two types: mutation to infect a mammal and mutation to produce a toxin harmful to mammals. The most useful test for detecting the ability of B. thuringiensis to mutate is serial passage of the agent in an environment in which the mutants in question would have a selective advantage over the parent agents and so reveal their presence in the mammalian body. TOXICITY In Europe and North America, new B. thuringiensis products have been subjected to extensive toxicological tests, which confirmed their innocuity. However, regulatory agencies have not specified what tests should be required for new B. thuringiensis products. Recently, the toxin of B. thuringiensis israelensis, when dissolved and injected intravenously into mice, was found to be highly toxic (LDso 1.3 mg/kg), being more so to the mouse than to the American cockroach and cabbage looper.22 FUTURE OF SAFETY TESTING Work has been started on the improvement of industrial strains of B. thuringiensis. It is still mainly at the stage of selecting from existing strains, with a start being made toward utilization of genetic engineering to transfer and to amplify characteristics for example, the possible use of B. subtilis to mediate change in B. thuringiensis. Unique codes of safety are being formulated worldwide for genetic engineering. Safety problems are not expected during manipulation of pest-control pathogens, because factors harmful to pests rather than to humans are being manipulated. From this viewpoint, it has been postulated that bacterial insect pathogens are ideal systems for basic work on genetic engineering. However, mediator organisms must be selected with care and a watch kept to avoid undesirable contaminants entering the systems. B. THURINGIENSIS USAGE Since bacterial control agents are not restricted-use materials, quanti- tative information on the usage of B. thuringiensis in agriculture is difficult to obtain. An annual report on pesticide use is published by the California Department of Food and Agriculture (CDFA). The most recent report
242 APPENDIX D (1983) provides quantitative data for virtually all pesticides used in California, giving amounts used on different crops, but figures were not available for bacterial agents. Therefore, it was necessary to approach manufacturers of B. thuringiensis (for example, Abbott Laboratories and Sandoz-Zoecon) for this information. T. Hsieh of Sandoz-Zoecon indicated that Javelin, a recent B. thuringi- ensis isolate developed by his company, is used primarily for the control of forest insects (gypsy moth, spruce budworm) and vegetable and alfalfa insects. He estimates that in the first two-and-one-half months on the market, 70,000 to 90,000 gallons of Javelin was sold in the United States alone. One to two quarts of Javelin are required per acre. Hsieh admitted that growth in the use of B. thuringiensis has been very slow, attributable mainly to the relatively low cost and effectiveness of conventional insecticides. Further, as a stomach poison, Javelin is restricted primarily to lepidopteran larvae that chew. However, recently a B. thuringiensis isolate has been discovered in Germany which is highly active against the Colorado potato beetle. Hsieh's estimate of the total amount of conventional B. thuringiensis (not including Javelin) sold by Sandoz-Zoecon last year is around 2 million gallons. This material is sold in many developing countries to control vegetable crop pests that can no longer be controlled by conven- tional insecticides. Phillip Grau of Abbott Laboratories estimates worldwide sales of Abbott's B. thuringiensis (Dipel) to be in the neighborhood of 3.5~.0 million pounds. It has been sold mainly for use on vegetables (lettuce, cole crops, tomatoes, mixed vegetables) and mosquito control. More recently, it is finding increasing use against forest insects. However, use on vegetable crops is being supplanted to some degree by the pyrethroids since they are registered for use on vegetables. B. thuringiensis is also used effectively to control mosquito larvae. According to recent annual reports of the California Mosquito and Vector Control Association, the following amounts of B. thuringiensis were used for mosquito control in California: 1983- 5,547 x 109 biological units (approximately 20,350 pounds); 198~18,630 x 109 biological units (approximately 68,370 pounds). For 1985, usage is expected to have doubled over that of 1984. According to M. Mulla (University of California, Riverside) and Hsieh, approximately 1 million gallons per year of B. thuringiensis are being used in West Africa against black flies (vector of onchocerciasis) by the World Health Organization. Mulla stated that a new bacterium (B. sphaericus) is being developed for use specifically on mosquitoes. It is more persistent than B. thuringiensis and will be used to complement B. thuringiensis. From discussion with a number of individuals, including those men-
PESTICIDE INNOVATION 243 tioned above, it is clear that the use of bacterial agents for insect control will increase substantially in the immediate future. However, it must be pointed out that the total amount of these materials used compared to synthetic organic chemicals is still extremely small, probably less than 0.5 percent. Viral Insecticides A typical nucleopolyhedrosis virus (NPV) is Baculovirus heliothis which produces crystal-like, irregular, proteinaceous polyhedral inclusion bodies (PIB) in nuclei of infected cells.23 Development of the NPV of Heliothis sp. began in 1961, progressed through various research and development phases, and attained technical realization as the first com- mercial viral pesticide. An exemption from the requirement of a tolerance was granted in May 1973 by the EPA and a label was approved in December 1975. Currently, B. heliothis is marketed as safe and effective for use on cotton against Heliothis sea under the name Elcar (Sandoz, Inc.~; Nutrilite products, Inc., has an equivalent experimental product called Biotrol-VHZ. RESISTANCE Selection pressures of LCs~7O maintained for 20 and 25 generations did not yield resistance in H. sea. Similar results were obtained with laboratory populations of H. armigera selected for resistance over 22 generations. There is no record of indisputable resistance of insects to viral agents in field trials or control programs. However, these agents have not been used for long and it is possible that low levels of resistance are present but are not readily detectable. In one case, NPV collected from distant plantations was more effective against the wattle bagworm, Kotochalia junodi, than virus collected from the local plantation in which tests were performed. Resistance might have been acquired by the local insects to the local virus or the observation might reflect differing levels of virulence among virus isolates. STABILITY, SENSITIVITY, AND PERSISTENCE Natural sunlight-ultraviolet radiation (> 290 nm) is the major environ- mental factor inactivating B. heliothis and probably most insect viruses. Although field temperatures of 15°-45°C had no effect on the stability of FIB, viral replication was inhibited at 40°C. In general, high temperatures (70°-80°C) and the presence of water completely inactivate FIB. Acids
244 APPENDIX D and alkalis disrupt FIB and thus presumably destroy viral activity. Early field and lab studies have indicated that most insecticides or insecticidal adjuvants are compatible with B. heliothis. TOXICOEOGICAE STUDIES The baculovirus of H. zea and Lymantria dispar exhibited no harm to mammals, fish, birds, or beneficial insects (including parasitic insects), had no relationship to arboviruses, and had no effect on aquatic invertebrates.24 Extensive safety testing of Neodiprion lecontei NPV and N. sertifer NPV was undertaken; carcinogenicity tests on newborn hamsters and a two-year carcinogenicity test on rats were among the battery of tests. There was no evidence that the viral preparation has any harmful effects. PRODUCTION B. heliothis can be produced only in Heliothis larvae, although several sophisticated processes have been suggested. Production of sawfly (N. sertifer) NPV is complicated by the fact that there are no synthetic diets or established cell cultures for sawhies.25 Hence, for virus production, larvae must be reared on their host food plant, infected with virus, and then harvested and processed. FIEED TRIAES Control by B. heliothis generally was as effective on cotton as with standard insecticides. Control was less effective on corn than with standard insecticides. Although all spray treatments were effective on soybeans, poorer results were obtained by releasing virus-infected larvae. Desired levels of control were not obtained on tobacco and tomatoes. Results on sorghum were comparable to those obtained with carbaryl and endosulfan. With the notable exception of sawflies, little is known of viruses pathogenic for nonlepidopteran insects, and continuous cell cultures from such important groups as the Hymenoptera, Coleoptera, and Orthoptera are lacking. FUTURE DEVELOPMENTS Several procedures for producing baculoviruses in either homologous or nonhomologous hosts have been proposed. Although B. heliothis produced in cell lines was as effective as that from larvae and it will replicate in cell lines, this technology will not be commercially practical for a long time. Significant development in production of B. heliothis will
PESTICIDE INNOVATION 245 probably come from new techniques such as production in nonhomolo- gous hosts coupled with recombinant DNA techniques. A key to the development of viral products is whether they can be safely released into the environment.26 Lois Miller, University of Idaho, is studying the replication of the baculovirus DNA, which could lead to a better understanding of how to enhance the virus' pathogenicity and expand its host range as well. EPA scientist Daphne Kamely observed that the fate of the baculovirus and retroviruses in the environment is not well understood. Studies are currently going on at Harvard University and at the National Institutes of Health to develop risk assessment models that may help the EPA to evaluate the consequences of the release of viruses into the environment. Fungal Insecticides Although mycoses caused by the entomopathogenic Fungi Imperfecti, Beauveria bassiana, Metarhizium anisopliae var. anisopliae and var. major (the color of the spores) have been studied for about a century, it is principally during the past 15-20 years that special attention has been focused on them to develop new methods of microbial control of insects.27 For many years they were regarded as biological agents of secondary interest, due to pessimistic conclusions from the first field trials in several countries at the end of the last century. However, in the 1950s East European countries started investigations, particularly with B. bassiana, as part of a general strategy to control the Colorado potato beetle, Leptinotarsa decemlineata. PRODUCTION A new stage technique for mass production of B. bassiana conidiospores is used in the USSR. First the biomass is produced as mycelium in a fermenter, and subsequently surface-cultured in trays of nutrient medium for sporulation. A similar technique is used for the mass production of M. anisopliae. The preparations have a limited viability of 2 or 3 months, a serious failure that considerably limits their industrial potential. In addition, production costs are high. TOXICITY In numerous safety tests, no infections have been induced in mammals with the common microbiological control fungi. These include short-term tests (feeding, inhalation, and intravenous and subcutaneous injection) and 90-day subacute inhalation and feeding studies of B. bassiana and M.
246 APPENDIX D anisopoliae in rodents; two-year intraperitoneal and subcutaneous injec- tion, lactatation and fertility tests in rats; and three-month and one-and- a-half-year studies of dusting effects on rats and mice. RESISTANCE Since these pathogens have been in nature for a long time, it is reasonable to presume that insects have developed low levels of resis- tance to them. With regard to infectivity, fungi reach their sites of action in the haemocoel through the cuticle, or possibly through the mouth parts and not via the gut wall as do bacteria and viruses. They become established only when the infective phase interacts with a susceptible host (that is, one that does not produce local reactions to ward off penetration). These factors are sobering reminders that fungi and their products as pest control agents do not have an unlimited potential. PRODUCTS AND USES Boverin, the trade name under which B. bassiana is marketed, is not used in the United States possibly because of economic considerations, which may not have been taken into account in the USSR in evaluating effectiveness, or because of climatic considerations. The climate in the Ukraine has justified the combined use of Boverin and reduced dosages of chemical insecticides such as chlorophos or malathion. However, when summers are particularly dry and hot, results are poor. Mycar, a preparation of Hirsutella thompsonii, was marketed by Abbott Laboratories until recently. The product was discontinued be- cause it was expensive to produce and large quantities were needed for each application. It is well established now that entomopathogenic fungi have a certain specificity. In the same species of fungus, different strains can have very different activity. FUTURE DEVELOPMENTS IN MICROBIOLOGICAL AGENTS The success of B. thuringiensis as an insecticide has initiated research to incorporate its toxin-coding genes into plant genomes in a manner that will allow them to be expressed and the toxin to occur either symplasti- cally or apoplastically within plants. One idea is to nick the circular B. thuringiensis plasmid and join it directly to a plant plasmid in vitro. Upon reintroduction of the now extended plasmid into a plant cell, it is conceivable that the B. thuringi- ensis toxin could be one of the translation products of its expression.
PESTICIDE INNOVATION 247 Since gene-incorporated traits are generally transmissible to future prog- enies, the B. thuringiensis gene might well end up in new seeds on the market. This was accomplished in tobacco by the Rohm and Haas Co. The plant did not produce enough insecticide, however, to kill insects. Another idea is to transfer the B. thuringiensis plasmid through the intermediacy of gall-producing bacteria to plants. A possible plasmid carrier could be attenuated Agrobacterium tumefasciens or its antagonist A. radiobacter. Plasmid transfer to the nitrogen-fixing bacteria in the Rhizobium genus and eventual expression inside the root system of the plant is also a possibility. Since these bacteria are already used commercially to inocu- late legumes, it is only one step to incorporate an extra gene into them for B. thuringiensis toxin production. One cautionary note in these ideas is whether the B. thuringiensis toxin-producing genes can or will be transferred through the plant plasmids to weeds, thereby having an adverse affect on the beneficial insects that suppress the proliferation of weeds. SUMMARY The potential of microbiological insecticides has barely been tapped. The advent of new viral, bacterial, and fungal insecticides with remark- able insect toxicity to selected target pests and negligible mammalian toxicity is possible. The pragmatic view of biological insecticides taken by regulatory agencies is likely to continue and anecdotal reports of toxicity such as that of B. thuringiensis israelensis to mice by intravenous administration will be placed in their proper perspective. Viral insecticides are particularly promising for the future. Safety prospects are also good and the chances of mutation to forms that are virulent to mammals and other vertebrates are practically nonexistent. The future is also likely to see structure-activity studies on the microbial toxins to determine if any underlying common molecular rationale exists to explain their mode of toxic action. These studies and the topographic details of the toxins derivable from them could also form the basis for a new generation of highly selective chemical insectides with high toxicity to only a very narrow spectrum of pests. These new chemicals are expected to better withstand scrutiny by the Delaney Clause. NOTES 1. National Research Council. 1986. Pesticide Resistance: Strategies and Tactics for Management. Washington, D.C.: National Academy Press.
248 APPENDIX D 2. Hollingworth, R. M. and A. E. Luna. 1982. Pesticidal Mode of Action, J. R. Coats, ea., New York: Academic Press. 3. Albers-Schonberg, G., B. H. Arison, J. C. Chabala, A. W. Douglas, P. Eskola, M. H. Fisher, A. Lusi, H. Mrozik, J. L. Smith, and H. L. Tolman. 1981. J. Amer. Chem. Soc. 103:4216. Fisher, M. H. 1985. Pp. 53-72 in Recent Advances in the Chemistry of Insect Control, N. F. Janes, ed. London: Burlington House. 5. Kay, I. T., and M. D. Turnbull. 1985. Pp. 229-244 in Recent Advances in the Chemistry of Insect Control. 6. Baker, R., C. J. Swain, and J. Head. 1985. Pp. 245-256 in Recent Advances in the Chemistry of Insect Control. 7. Palmer, C. J., and J. E. Casida, 1984. J. Agr. Food Chem. 33: 976. 8. Hollingshaus, J. G., and R. J. Little. 1985. Abstract. Agrochemicals Division National Meeting, American Chemical Society, Miami Beach, Fla., April 28-May 3. 9. Grosscurt, A. C. 1978. Pestic. Sci. 9:373. 10. Cohen, E., and J. E. Casida. 1982. Pestic. Biochem. Physiol. 17: 301. 11. Janes, N. F., ed. 1985. Recent Advances in the Chemistry of Insect Control. London: Burlington House. 12. Fukuto, T. R. 1984. ACS Symposium Series, No. 255. Washington, D.C.: American Chemical Society, pp. 87-101. 13. Sakai, M. 1983. Abstracts. 5th International Congress on Pesticide Chemicals, IIa-2, Kyoto, Japan. 14. Bowers, W. S. 1985. Pp. 53-72 in Recent Advances in the Chemistry of Insect Control. 15. Miyakado, M., I. Nakayama, A. Inoue, M. Hatakoshi and N. Ohno. 1985. J. Pestic. Sci. 10:11. 16. Lay, S. V. 1985. Pp.305-322 in Recent Advances in the Chemistry of Insect Control. 17. Agrios, G. N. 1978. Plant Pathology, 2nd ed. New York: Academic Press. 18. Dulmage, H. T. 1981. Insecticidal activity of isolates of Bacillus thuringiensis and their potential for pest control. P. 193 in Microbial Control of Pests and Plant Diseases (1970-1980), H. D. Burges, ed. London: Academic Press. 19. Kolata, G. 1985. Genetically engineered organisms in agriculture. Science 229 (5 July):34. 20. McGaughey, W. H. 1985. Insect resistance to the biological insecticide Bacillus thuringiensis. Science 229(12 July): 193-195. 21. Burges, H. D. 1981. Safety, safety testing and quality control of microbial pesticides. P. 737 in Microbial Control of Pests and Plant Diseases (1970-1980). 22. Roe, M. R., P. Y. K. Cheung, B. D. Hammock, D. Buster, and R. A. Alford. 1985. Endotoxin of Bacillus thuringiensis var. israelensis broad spectrum toxicity and neural response elicited in mice and insects. Pp. 279-292 in Bioregulators for Pest Control, P. A. Hedin, ea., ACS Symposium Series 276. 23. Ignoffo, C. M., and T. L. Couch. 1981. The nucleopolyhedrosis virus of Heliothis species as a microbial insecticide. P. 330 in Microbial Control of Pests and Plant Diseases (1970-1980). 24. Lewis, F. B. 1981. Control of the gypsy moth by a Baculovirus. P. 363 in Microbial Control of Pests and Plant Diseases (1970-1980). 25. Cunningham, T. C., and P. F. Entwistle. 1981. Control of Sawflies by Baculovirus. 379 in Microbial Control of Pests and Plant Diseases (1970-1980). P. 26. Sun, M. 1985. Biotechnology focus on viruses. Science 228 (14June):129~1295. 27. Ferron, P. 1981. Pest control by the Fungi Beauveria and Metarhizium. P. 465 in Microbial Control of Pests and Plant Diseases (1970-1980).