The Control of Parasites: the Role of Drugs1

OF WASHINGTON, VOLUME 44, NUMBER 1, JANUARY 1977 17

Steinert, M., and A. B. Novikoff. 1960. The associated structures of Fasciola hepatica. existence of a cytostome and the occurrence Quart. J. Microsc. Sci. 104: 505-512. of pinocytosis in the trypanosome, Trypano- Thorsell, W., and N. Bjorkman. 1966. In soma mega. }. Biophys. Biochem. Cytol. 8: vitro studies on the effect of hexachlorophene 563-569. and its dimethylether on the liver fluke Fas- Sterling, C. R., and M. Aikawa. 1973. A ciola hepatica L. Z. Parasitenk. 28: 116- comparative study of gametocyte ultrastruc- 124. ture in avian Haemosporidia. J. Protozool. Van den Bossche, H., and P. A. J. Janssen. 20: 81-92. 1967. The biochemical mechanism of action Sterling, C. R., and D. L. DeGiusti. 1972. of the anthelmintic drug tetramisole. Life Ultrastructural aspects of schizogony, mature Sci. 6: 1781-1792. schizonts, and merozoites of Haemoproteus , and . 1969. Biochemical mech- metchnikovi. J. Parasit. 58: 641-652. anism of action of the antinematoclal drug Strufe, R., and R. Gonnert. 1967. Influence tetramisole. Biochem. Phannacol. 18: 35-42. of drugs on the metabolism of the tape- Vanderberg, J., J. Rhodin, and M. Yoeli. worm. Z. Tropenmed. Parasit. 18: 193-202. 1967. Electron microscopic and histochem- Terzakis, J. A., H. Sprinz, and R. A. Ward. ical studies of sporozoite formation in Plas- 1967. The transformation of the Plasmo- modium berghei. ]. Protozool. 14: 82-103. dium gallinaceum oocyst in Aedes aegypti Vickerman, K., and A. G. Luckins. 1969. Lo- mosquitoes. J. Cell Biol. 34: 311-326. calisation of variable antigens in the surface Theakston, R. D. G., and K. A. Fletcher. coat of Trypanosoma hrncei using ferritin- 197la. Electron cytochemical study of glu- conjugated antibody. Nature 224: 1125- cose-6-phosphate dehydrogenase activity in 1127. erythrocytes of malaria-infected mice, mon- Von Brand, T. 1973. Biochemistry of Para- keys and chickens. Life Sci. 10: 701-711. sites. 2nd ed. Academic Press, New York. •, and . 1971b. Ultrastructural lo- Yamao, Y. 1954. Histochemical studies on en- calization of NADH- and NADPH-dehydro- doparasites. V. distributions of the glycero- genases in the erythrocytic stages of the ro- monophosphates in the tissues of flukes, dent malaria parasite, Plasmodium berghei. Eurytrema coelomaticum, E. pancreaticum, Life Sci. 9: 421-428. Dicrocoeliiim lanceatum and Clonorchis sinen- Threadgold, L. T. 1963. The tegument and sis. J. Coll. Arts. Sci. Chiba Univ. 1: 9-13.

The Control of Parasites: The Role of Drugs1

W. C. CAMPBELL Merck Institute for Therapeutic Research, Rahway, New Jersey 07065

". . . the play is the tragedy 'Man', are the prevailing social and political forces in and its hero, the Conqueror Worm" the case of man, and the husbandry practices Edgar Allan Poe in the case of domestic animals. Of the mechanisms invoked intentionally to reduce para- Poe's words were allegorical rather than sitism, biological control and vaccination have parasitological; but parasite control might be played roles that have been overshadowed by regarded, in large measure, as an attempt to the role of chemotherapy. It seems likely that strip the worm of its role as conqueror. antiparasitic drugs will continue to be of im- Undoubtedly the major extrinsic factors in portance for some time. determining the extent and degree of parasitism Antiparasitic drugs actually play many roles —not just because there are many drugs, but because one drug in its time plays many parts. Some are general roles—the parts played by

Table 1. Productivity of Merino sheep from age Despite the very evident immunogenicity of 6 months to age 18 months (50 sheep per group, the infection, vaccination has not become a all groups grazed together). Anthelmintic B has significant means of control; and it has been broader spectrum than A, and greater activity the role of the antiparasitic drugs to make against immature worms. Both drugs were ad- ministered monthly. From Gordon, 1963. feasible the modern intensive production of an important food animal. % sheep of In helminthiasis we have to a lesser, but 70 Ih or % fleeces over at 18 Wool showing still significant extent, moved from the use of months per head defective drugs for the salvage of sick animals to the Treatment old Ib. wool prevention of disease. This is accomplished Controls — n o not by continuous chemoprophylaxis but by anthelmintic treatment 16 6.7 32 tactical treatment (administration of drugs Anthelmintic A 35 7.4 14 whenever local conditions make a buildup of Anthelmintic B 62 7.9 4 parasitism seem likely) or strategic treatment (administration of drugs according to a schedule based on the epidemiological conditions drugs in the overall scheme of parasite control prevailing in a given region). Quite suddenly, —and I shall say a little bit about those. There as these things go, we no longer hear of dev- are innumerable specific roles: this drug kills astating outbreaks of acute fascioliasis in the Haemonchus in sheep, that drug kills Ancylo- United Kingdom, or of haemonchosis in Aus- fitoma in man, and so on. I am not going to tralia. More important, perhaps, than the pre- say anything about those roles. They have been vention of overt disaster is the prevention of recounted many times in recent years. There the more subtle deleterious effects of parasitism are specific roles of another kind—the biochem- on the quantity and quality of livestock prod- ical roles. I do want to say something about ucts. In either case, anthelmintics, too, are those, partly because they are not often pulled currently playing the role of enhancer of agri- together in one place, but mostly because I cultural production. Obviously the increase in want to end by discussing how they might productivity is easy to demonstrate under con- relate to the discovery of tomorrow's anti- ditions favorable to clinical parasitism, and parasitic drugs. becomes more difficult and more controversial under conditions where parasitism is very much subclinical. An example of enhanced produc- The Role of Drugs in tivity may be found in the work of Gordon Livestock Parasitism (1963); some of his data are given in Table 1. Although I will be focusing attention on I want, however, to go beyond such well- helminth infections, the general role of drugs documented effects and to consider a more in controlling parasitic diseases in domestic recent and more comprehensive examination of animals is perhaps best illustrated by the the subject. situation in coccidiosis. Since the introduction Only in recent years have attempts been of medicated diets for the control of coccidiosis made to determine whether the increased pro- in 1948, it has been virtually axiomatic that an ductivity associated with drugs necessarily re- effective coccidiostat is indispensable for suc- sults in economic benefit. Just last month a cessful intensive production of chickens—this, paper appeared that is, I think, a striking ex- mind you, in a parasitic situation in which im- ample of how parasitological and economic munity to reinfection is easily demonstrated studies can be combined and focused on this under both experimental and field conditions. point (Anderson et al, 1976). These workers Indeed the natural immunological protection determined the economic returns from 2 is exploited in conjunction with chemo- strategic schemes of helminth control in therapy, and the poultryman, in this day of weaned lambs in Western Victoria, Australia; sophisticated agriculture, takes a great interest and they compared these with returns from in the degree to which a given coccidiostat flocks without a treatment scheme and, in part, will permit the development of natural im- with flocks subjected to unusually intensive munity, especially in replacement chickens. treatment. The economic analysis included

Table 2. Maximum financial return (given favorable economic conditions) from flocks of lambs with no treatment schedule, a "traditional" strategic schedule, a "critical" strategic schedule, or an intense (biweekly) schedule. From Anderson, et al. 1976.

Maximum return ( Australian dollars ) None Traditional Critical Intensive Gross return per 100 sheep 1029 1115 1205 1441 Net return per 100 sheep* 1023f 1097 1186 1271 Percent return on cost of treatment:!: — 617 1254 151 * Gross return minus cost of treatment. t Lambs received one treatment to control clinical parasitism. :j: Cost in addition to cost of single treatment in flocks with no treatment schedule. extrapolation of data to years in which eco- The Role of Drugs in nomic conditions were more favorable or less Parasitism of Man favorable than the year in which the data were collected. The figures in Tables 2 and 3 have First we must remember that antiparasitic been extracted from the wealth of information drugs have played, and continue to play, a life- in this paper to illustrate the following points. saving role. I seem to recall from some intro- (1) All schemes of anthelmintic treatment gave ductory textbook that visceral leishmaniasis better livestock production than no treatment was about 95% fatal before the introduction scheme, and thus a higher gross financial re- of antimonial drugs, and about 95% nonfatal turn. (2) Under favorable economic condi- afterwards. The lifesaving role of the anti- tions, all treatment schemes also gave a higher malarial drugs needs no elaboration. Control net return. Thus treatment always made eco- of morbidity on a population basis has also nomic sense, although the intensive (biweekly) been of importance in protozoal and helminthic treatment was not nearly so profitable as the infections, but it must be admitted that the more judicious schemes. (3) Under unfavor- overall success rate has been disappointingly able economic conditions, both strategic treat- low, and the need for effective strategic control ment schedules were economically advantage- remains distressingly great. ous; but biweekly treatment resulted in The incentives of the marketplace ensure a financial loss in the face of production gains. steady search for new antiparasitic drugs for Thus it can be seen that anthelmintics can use in livestock, but the situation is less reassur- add plumpness not only to the carcass but also ing with respect to mankind. On the one hand to the purse; and that the financial rewards there is the lowly pinworm, parasitizing the are more or less proportional to the enlighten- posteriors of our progeny, doing it in all climes ment of the designer of the control program. and, most important, affecting the rich as well

Table 3. Minimum financial return (given unfavorable economic conditions) from flocks of lambs with no treatment schedule, a "traditional" strategic schedule, a "critical" strategic schedule, or an intense (biweekly) schedule. From Anderson, et al. 1976.

Minimum return (Australian dollars) None Traditional Critical Intensive Gross return per 100 sheep 333 368 398 451 Net return per 100 sheep* 327f 351 379 281 Percent return on cost of treatment:!: 200 400 -28 * Gross return minus cost of treatment. t Lambs received one treatment to control clinical parasitism. ± Cost in addition to cost of single treatment in flocks with no treatment schedule.

Table 4. Probable mode of action of common Today—The Challenge and the Opportunity" anthelmintics (from the literature). Compiled in (Anonymous, 1975a). It discusses three tropi- collaboration with Dr. M. H. Fisher. cal diseases with a combined prevalence of 600 million cases, and three other very serious 1. Neuromuscular anthelmintics a. Acetylcholinesterase inhibitors but less common diseases of man. Five of dichlorvos these six diseases are caused by protozoa or trichlorfon b. Choline mimetics helminths. The report stresses the urgent need bephenium thenium for a coordinated research program aimed at hycanthone the development of better control measures for c. Neuromuscular depolarizers methyridine these diseases. In the program proposed by pyrantel morantel WHO, "task forces" of internationally reputed d. Neuromuscular hyperpolarizers scientists will define the needs, and plan and piperazine e. Ganglion stimulants direct scientific operations to meet those needs. levamisole The operations will be carried out by a "net- 2. Inhibitors of fumarate reductase work" of medical school and institutional lab- thiabendazole cambendazole oratories in the tropics, mostly in Africa. The 3. Inhibitors of glucose uptake objective is to discover, develop and make mebendazole available on a large scale, new and better pyrvinium dithiazanine remedies for these diseases before the end of styrylpyridinium the century. This strategy is indeed admirable, 4. Inhibitors of glucose metabolism but are the tactics sound? antimonials In commenting on the WHO report, Smith 5. Disruptors of glycogen metabolism (1976) has pointed out that even if new drugs niridazole and vaccines are developed, they will have 6. Uncouplers of oxidative phosphorylation rafoxanide little impact in disease control in developing clioxanide countries unless major efforts are made to oxyclosanide niclosamide solve the practical problems associated with bithionol hexachlorophene their use under field conditions. That is true, nitroxynil but unfortunately there is a more immediate 2,4-dinitrophenol aspidixim oleoresin cause for concern. It seems to me that the 7. Inhibitors of dihydrofolate reductase WHO proposal does not offer the best oppor- suramin tunity for discovering the needed new drugs; 8. Others and that is a subject to which I shall return. bitoscanate tetrachlorethylene phenothiazine organo-arsenic Biochemical Roles of Antiparasitic Drugs To review very briefly the biochemical roles as the poor. Sad to relate, the amount of of antiparasitic drugs, I have put them in cate- money spent for the treatment of pinworm gories, in collaboration with Dr. M. H. Fisher, probably is more or less inversely proportional as shown in Table 4. It must be emphasized to its importance as a pathogen. On the other that the actual mode of action is rarely if ever hand there are the many serious parasitic in- known with certainty. Of the many effects re- fections that plague the underdeveloped na- ported in the literature, those listed in the table tions, especially those in tropical regions. Good seem at this date to be the most likely mech- drugs are available for some of these diseases anisms of anthelmintic action. but not for others. The commercial incentives In the first category there is a large group for the development of new drugs for these of anthelmintics that interfere with nervous diseases are, at best, uncertain. What can be control of muscle function, either at the neuro- done about them? muscular junction or at the nerve ganglion. The World Health Organization recently Some, such as the organophosphates, inhibit published an important, and already widely the deactivation of acetylcholine by acetyl- quoted, document entitled "Tropical Diseases cholinesterase. Others, such as bephenium and

Copyright © 2011, The Helminthological Society of Washington OF WASHINGTON, VOLUME 44, NUMBER 1, JANUARY 1977 thenium, mimic this effect by usurping the the dihydrofolate reductase of the worm is acetylcholine receptor site, and thus foiling the more sensitive to the drug than is the isoen- esterase. Others interfere with neuromuscular zyme of man. Mel W has an antimitotic effect transmission by altering the electric charge of on the embryogenesis of filarial worms, and the nerve cell membrane—either depolarizing this apparently causes the death of adult fe- it, in the case of methyridine, pyrantel, and males. Hycanthone is an intercalator of DNA morantel, or hyperpolarizing it, in the case of of bacterial origin, but there is no informa- piperazine. The former action makes the tion as to whether it has a similar effect on worms uptight, while the latter makes them schistosome DNA. It has recently been pro- floppy. The final item in this category is posed that it is an acetylcholine blocker with levamisole, which probably acts directly on the a special affinity for the acetylcholine of schis- nerve ganglion—producing a paralysis that is tosomes. For some drugs, such as the classic self-limiting in vitro. Levamisole also inhibits tetrachlorethylene and phenothiazine, there is fumarate reductase (but it paralyzes worms no convincing evidence of a biochemical mode that lack this enzyme) and inhibits alkaline of action. phosphatase and acetylcholinesterase in mam- malian test systems. What Roles Do We Want Most of the benzimidazole anthelmintics Drugs to Play? probably kill nematodes by inhibiting their Foremost on the long list of desirable proper- fumarate reductase, thereby blocking carbo- ties of an anthelmintic is efficacy against the hydrate metabolism. When the enzyme is ex- target helminth or group of helminths. By tracted from benzimidazole-sensitive Haemon- this we usually mean that we want the drug chus, it is inhibited by thiabendazole in vitro; to kill the worms. I believe we should think but this is not the case when the enzyme is less about that. Worms that are merely para- obtained from benzimidazole-resistant strains lyzed reversibly in the intestinal tract may be of the same worm. passed from the host before they can recover, Mebendazole apparently blocks carbohydrate so that a direct lethal action is not necessary. metabolism in a way that is quite different It is less often appreciated that such sublethal from that of other benzimidazoles. Through effects are probably sufficient in some extra- inhibition of glucose uptake, or interference intestinal parasites. Schistosomes which are with some later step, there is a marked deple- merely dislodged from the mesenteric vessels tion of glycogen. There is also a rapid disrup- and swept into the liver will in many cases fail tion of secretory organelles in the tegument to recover, and their gradual demise in the liver of cestodes and in the intestinal cells of nema- sinuses may be even more desirable than their todes and these changes could well account sudden destruction in the vessels. But I am for the death of the worms. The cyanine dye thinking of something other than slow versus anthelmintics also inhibit the uptake of ex- fast destruction of worms. Some diseases of ogenous glucose, and probably act differently medical and veterinary importance are due al- in aerobic and in anaerobic worm species. most entirely to the reproductive products of The antimonial compounds interfere with the worms. Prominent examples are schistoso- glucose metabolism in schistosomes by inhibit- miasis and trichinosis. In both cases the adults ing phosphofructokinase. The enzyme of the are known to be relatively harmless (except schistosome is more sensitive to such inhibition when, present in very large numbers) and in- than is the isoenzyme of the host. Niridazole deed they may confer on the host some degree disrupts the glycogen metabolism of susceptible of immunity against reinfection. We would schistosomes by inhibition of two essential en- almost entirely suppress the pathogenicity of zymes. these worms if we were to suppress their A large group of anthelmintics, consisting reproduction. As it happens, in both cases we mainly of salicylanilides, bis-phenols and nitro- know of chemicals that will do precisely that— phenols, probably act as uncouplers of oxida- nicarbazin, thiosinamine, dapsone and others in tive phosphorylation. Thus they block energy the case of schistosomiasis (Campbell and production. Cuckler, 1967; Machado et al., 1970; Davies Suramin probably kills Onchocerca because and Jackson, 1970; Pellegrino and Katz, 1975)

Copyright © 2011, The Helminthological Society of Washington 22 PROCEEDINGS OF THE HELMINTHOLOGICAL SOCIETY and thiabendazole in the case of trichinosis "There is little doubt—given time, sufficient (Campbell and Guelder, 1964). In the case resources and the harnessing of modern molec- of schistosomiasis the chemosterilizing drugs ular biology, chemistry and immunology into have not been put to practical use, but the well balanced laboratory teams—that the de- potential for this general approach in schisto- sirable new drugs and vaccines can be discov- somiasis seems to warrant much more attention ered." Two thoughts come to mind. First, we than it has received. In the case of trichinosis have not in fact been given much time, if we opportunities for use of such a drug would be are to meet the WHO deadline of the end of very limited because the disease is usually the century. Second, I share Smith's confi- diagnosed after the progeny have been pro- dence that the harnessing of molecular biology, duced. Yet the principle has actually been biochemistry and immunology will eventually put into practice with apparent success (Ger- give us the vaccines we need, but I do not wel et al., 1974). Among several people who think it will give us the drugs we need in the were known to have eaten heavily contam- meantime. How then should we look for new inated pork, there was one who had eaten it antiparasitic drugs? completely raw and who had eaten a whole This is the part that will make you wince. pound of it (containing an estimated 11,500 I'll come right out with it. I am a great be- larvae). That person was in grave danger— liever in massive empirical screening. I am yet treatment prevented any serious illness in especially delighted this afternoon to have the that person or his less indulgent colleagues. opportunity to speak to so reputable an audi- The point I wish to make is not that chemo- ence on a subject thought by so many to be so sterilization of Trichinella is of great conse- disreputable. quence (except in special cases) but that we It seems to me that those who oppose empiri- should think carefully of the type of efficacy cal screening and favor the so-called rational that could be most useful for whatever para- approach do so largely because they find the sitic diseases we have in our sights. There idea of screening intellectually humiliating. seems to be no general agreement on the type This attitude may arise in part from a lack of drug (if any) that would be most useful in of incentive to evaluate the two approaches dis- the strategic control of hydatid disease or passionately. The incentive in terms of finan- cysticercosis. We take for granted that it cial grants (to say nothing of professional es- would be desirable to kill paramphistosomes in teem and advancement) is to focus on the sheep, preferably the immature fluke in the rational approach and indeed to exclude any small intestine. Do we stop to wonder whether notion of empiricism. we could block its pathogenicity by inducing There is one category of anthelmintic re- the immature fluke to migrate prematurely to searcher that we can readily dismiss. They are the rumen? Suppose that grazing cattle had a those academicians who (taking an attitude rumen "bullet" (of the type used for bloat attributed to E. Y. Davis [Leake, 1976]) ap- control) that released a nonabsorbed com- parently regard an anthelmintic as something pound that either caused Fasciola cysts to ex- that, when put into a parasitized animal, pro- cyst in the rumen or prevented their subsequent duces a scientific paper. Not surprisingly, the excystment in the duodenum. We might then chemotherapeutic papers of such workers are prevent all clinical phases of fascioliasis and often rubbish—the scientific acumen of the be free of drug-related tissue residue and milk authors having been betrayed by their lack of residues into the bargain. For all of our serious interest in the subject matter. I was recently verminous diseases we should earnestly seek intrigued to discover a similar exasperation on some subtle alternative to the direct vermicidal the part of the editor of a journal devoted not approach. to livestock health, but to human LeaJth. It How Should We Look for New appeared to him that "one of the more popular Antiparasitic Drugs? ways to achieve a research publication is to persuade a pharmaceutical company to supply, Let us return for a moment to Smith's com- free of charge, a quantity of a new—or a fre- ments on the WHO proposal for disease control quently not so new—preparation, then give in the tropics (Smith, loc. cit.). Smith says, this to a number of patients who may or may

Copyright © 2011, The Helminthological Society of Washington OF WASHINGTON, VOLUME 44, NUMBER 1, JANUARY 1977 23 not be suitably chosen, make up some tables of and devised a drug to exploit that difference results, hazard a chi square confirmation of to the detriment of the worm. Would we be scrappy results and, finally, proclaim in print heroes? Probably not, because the probability is a wonder drug with a most unlikely cure rate" high that the drug would do something else (Anonymous, 1975b). besides damaging the worm; and the probabil- But my quarrel is with the serious and dedi- ity is high that that something else would be cated chemotherapists who are committed to a Bad Thing. It seems to me that we would be the "rational approach" to the discoveiy of guilty of arrogance were we to assume that useful new anthelmintics. They are by no means we could predict the total impact of any novel limited to academic and governmental labs, biodynamic substance on the host. Inevitably, but are even now hard at work in industrial the more novel the prospective drug, the more labs (surrounded by empiricists of all sorts helpless we are in predicting all its effects. and conditions). While the greatest danger is the rational selec- The rationalist hopes that by studying the tion of a chemical structure that turns out to basic physiology and biochemistry of helminths, have unsuspected host toxicity, the converse he will discover a means to disrupt those pro- also imposes a limit on the potential success of cesses selectively and thereby provide a novel the rationalist. The knowledge that dietary chemotherapeutic agent. He may find that a thiamine is required for the health of poultry worm has a high requirement for some sub- might well have dissuaded one from proposing stance that the host does not require at all. a thiamine antagonist as a continuous dietary He would then select known antagonists or medication for poultry coccidiosis. Fortunately, competitors of that substance for therapeutic neither the anti-thiamine action of amprolium trials, or he might design entirely new mole- nor the dependence of coccidia on exogenous cules for the same purpose. He may find in a thiamine was known when amprolium was dis- helminth a metabolic step that is not known to covered as a coccidiostat! Thus, because of our occur in the host species—or, better yet, that present and future inability to predict all the is known not to occur in the host species. The effects, good and bad, of a new chemical struc- enzyme responsible for a particular metabolic ture, the rationalist takes a big chance. step in the helminth may be qualitatively or But chance is supposed to be what empirical quantitatively different from the enzyme con- screening is all about! Chance, of course, plays trolling the same step in the host. Whatever a part in all scientific research (and has been the difference, the objective of the therapist is particularly prominent in most of the particu- a chemical that will disrupt the metabolism of larly important scientific advances). It plays the parasite but not that of the host. a larger than usual role in experimental chemo- Now what is wrong with all of that? Some therapy, but on the whole I think that the role say that what was wrong in the past was our it plays in empirical screening is no whit infinitesimally small knowledge of helminth greater than the role ascribed to it above in the biochemistry; that the situation is improving rational approach. (The glaringly obvious ele- constantly; and that we must strive for greater ment of chance in blind screening in vivo is and greater knowledge to give us more and offset to a slight extent by the fact that one more chemotherapeutic targets to shoot at. I'll gets at least a crude indication of differential go along with the idea of greater basic informa- toxicity right off the bat.) tion giving more targets, but I think this is The other thing that gives screening a bad not the important point. No, the trouble with name is its routine aspect. What is often for- the whole scheme is not just our inadequate gotten is that a bioassay is simply a sorting knowledge of parasite biochemistry or host bio- device, as is a fractionation column, an ammo chemistry. The big problem is our lack of acid analyzer, or a computer program for any knowledge of the hypothetical drug. Suppose kind of data analysis. A scientist who spends we discovered almost everything about the all of his time cranking the handle of a chemo- metabolism of Trichostrongylus axei, and al- therapeutic assay is mistaking the tool for the most everything about the metabolism of Bos trade. So is any other scientist who is too busy taurus (we can never discover everything); collecting data to have time to do anything and suppose we selected a metabolic difference with them. Nor does the problem lie in the

Copyright © 2011, The Helminthological Society of Washington 24 PROCEEDINGS OF THE HELMINTHOLOGICAL SOCIETY fact that the chemotherapist appears to use so employ empiricism, not because future suc- little of the data that goes through his sorting cesses are guaranteed but because they are machine. After all, the work that lead to the probable and because we can influence their discovery of quasars involved in the collection probability. of astronomical numbers of (astronomically) It seems to me that the WHO report to uninteresting astronomical facts. The chemo- which I have referred fails to take into account therapeutic empiricist must concentrate on the these aspects of drug discovery. There are, of nature of the bioassay to be used and on the course, strong arguments for placing the effort results that come out of it. As in other fields, in non-industrial laboratories in non-industrial he may use an assay that has previously been nations. The WHO is undoubtedly right in established; he may wish to devise a novel saying that solutions developed in the tropics assay; or he may have to devise a novel assay. would be more acceptable in the tropics. The decision is of crucial importance; and if Further, commercial organizations find little or new methodology is to be involved, the exer- no commercial incentive for tackling the dis- cise may be scientifically demanding. eases in question. Yet, because of the inherent A good example of the rational exploitation frailties of the proposed "task force" and "net- of empirical screening is the tetramisole story. work" system, we must continue the search for Anthelmintic screening has traditionally been an acceptable alternative approach to the im- carried out in mice or rats, but the original mense problem of tropical disease. predecessor of tetramisole (the initial lead) It is a question of probabilities; and I believe was inactive against helminths in these animals. the probability of finding the drugs needed by Fortunately, the screening in this case had the developing nations could be increased by actually been carried out in chickens in which bringing the industrial-type of drug screening this predecessor was apparently active (Raey- into the overall picture. Under some circum- maekers et al., 1966). As it turned out, it was stances non-commercial agencies can mount a the chicken, not the scientist, that synthesized prodigious screening effort, such as the anti- the anthelmintic. But it was the scientist who malarial program of the United States govern- discovered it, who postulated, sought and iden- ment (Kinnamon and Rothe, 1975). Alterna- tified in the feces of treated chickens a metab- tively it should be possible to develop a system olite active in mammals; and it was the scientist in which private industry would do the screen- who made a more active analog of the metab- ing and primary evaluation of drugs, while the olite, and who found a less toxic isomer of the further evaluation and development of prom- analog. Another important example is the ex- ising candidates would be assigned to the "task ploitation of the early thioxanthone lead to forces" and "networks" of the nonindustrial yield hycanthone—a development that made sector. In this way the balance between invest- use of the chemical prowess of a mold (Archer ment risk and potential profit might be tipped and Yarinsky, 1972). Although these two ex- in favor of a really intensive industrial screen- amples involved nonhuman chemistry in un- ing program. It would not be easy to put usual ways, it must be remembered that it is such a dual (industrial and nonindustrial) ap- the work of the human chemist that is of proach into practice but it seems to me that paramount importance in these and all other such an approach offers the only hope for examples of the exploitation of empirical leads. achieving reasonable control reasonably soon. Virtually all modern anthelmintics represent I hope no one will come to the conclusion vast improvement over initial discoveries—im- that I am disparaging the value of biochemi- provement achieved by an elaborate interplay cal research with respect to the chemotherapy of molecule manipulation and biological evalua- of parasites. I would emphasize its value. In tion . the first place, the high odds against the ratio- Perhaps the biggest thing that empiricism nal approach do not mean that the approach has going for it is its past success. So far as I should not be tried. (In any case the urge to know, all antiparasitic and antibacterial drugs try it is irresistible.) But the greater value of had their origin either in empirical screening biochemical research in chemotherapy lies else- or in a chance observation made in the clinic where. or in the laboratory. We should continue to One of the roles played by antiparasitic

similar PABA analogs were tested in coccidiosis (Rogert et al., loc. cit.). In contrast to the situation in bacteria, they not only worked but some of them were many times more potent than the sulfonamides (and a derivative of one of them, ethopabate, became a highly success- ful commercial coccidiostat). Here, knowledge of the mode of action of empirically discovered coccidiostats led to the testing (and thence to the discovery) of a drug that was therapeu- tically more effective. Note that the rationale Figure 1. Major metabolic steps in the bio- rested on the mere presumption that folic acid genesis of i'olic acid in bacteria and probably in synthesis in coccidia was the same as in bac- coccidia. teria. Actually, the mode of action of the newer drug is probably not quite the same as that of the original. It is true that the activity of etho- drugs is that of educator. The study of empir- pabate is reversed by the addition of PABA to ically discovered drugs has taught us far more the diet of infected chickens; and it is possi- about basic chemotherapeutic mechanisms than ble that (in chickens on a normal diet) etho- has basic study of the parasites themselves. pabate simply blocks the reaction between This is not only important in terms of the pteridine and PABA (Step A in Figure 1). progress of biological science (a worthy end in However, it is considered more likely that etho- its own right) but opens the way for the semi- pabate blocks Step B, i.e., that, in competi- rational approach to chemotherapy. While tion with PABA, it reacts with pteridine to form the triumphs of this method have been few the ethopabate analog of dihydropteroic acid; and far between, the principle is clear and and that this analog (2'-ethoxydihydropteroic attractive: empirical discovery of a drug is acid) is sterically incapable of being coupled followed by elucidation of its mode of action, to glutamic acid to form dihydrofolic acid which in turn provides a basis for looking for (Rogers et al., loc. cit). other compounds that will achieve the same The next step in the folic acid pathway therapeutic effect while yielding benefits in (Step C in Figure 1) may be the Achilles' heel terms of improved spectrum, efficacy, patent- for many parasites, yet so far it has apparently ability or whatever. Often this approach leads been struck only by the arrows of the empir- to the serendipitous discovery of compounds icists. Jaffe (1972) has recently reviewed the having radically different therapeutic effect, presence of dihydrofolate reductase in para- but sometimes it works as intended. A particu- sites and their differential sensitivity to folate larly pertinent type of semirational approach is analogs and "antifol" drugs such as pyrimeth- the antimetabolite attack exemplified by the amine and trimethoprim. He cited the de- work of Hitchings (1972) and Rogers et al. tection of dihydrofolate reductase in plasmodia, (1964) in bacterial and protozoal infections but trypanosomes, schistosomes, Nippostrongylus apparently not yet used successfully against brasiliensis, and four genera of filarial worms. helminths. The reductase of Onchocerca volvulus turns out Once it became known that the antibacterial to be highly sensitive to inhibition by suramiii— activity of sulfonamides was due to antagonism offering a possible retrospective explanation of of the bacterial metabolite p-aminobenzoic acid the clinical efficacy of suramin in the treatment (PABA), attempts were made to achieve sim- of onchocerciasis in man. Even more recently, ilar antagonism and similar therapeutic efficacy the enzyme has been detected in coccidia thus by using analogs of PABA, such as 4-amino-2- offering a retrospective biochemical explana- chloro beiizoic acid (Wyss et al., 1943). In- tion of the action of pyrimethamine in avian deed such analogs were effective to some ex- coccidiosis (Wang et al., 1975). tent against some bacteria. Because of this, Obviously there is a continuous gradation and because sulfur and sulfonamides were between empirical and rational methods, and known to be active against coccidia in chickens, the so-called "semirational" approach can in

might then have designed or selected chemical structures with this property, or have screened randomly for structures with this property. In doing so one might well have discovered some of the more potent benzimidazoles that have in fact appeared. The trouble with that kind of approach is that one tends to find drugs that are basically similar to the original in terms of C3H70 efficacy, though they are likely to be different (for better or worse) in other respects. They are, therefore, likely to share weaknesses as well as strengths in therapeutic spectrum, and to share problems of drug resistance. But the most instructive aspect of this hypothetical example of semirational drug development, is that one would not have discovered the one benzimidazole anthelmintic that does differ radically from the original. Mebendazole apparently does not inhibit the fumarate reductase of Haemonchus (Prichard, 1973). Consider three examples of benzimidazole anthelmintics (Figure 2). All share the Figure 2. Chemical structure of thiabendazole basic benzimidazole structure, but it is at once (top), oxibendazole (middle), and mebendazole apparent that thiabendazole is very different (bottom). from oxibendazole and from mebendazole, while the latter two compounds are very similar to each other. Yet in terms of anthelmintic practice have either a low or a high ratio of spectrum I believe it is fair to say that thia- empirical to rational content. The antimetab- bendazole and oxibendazole are more similar olite method has been developed by Hitchings than oxibendazole and mebendazole. It is re- (1972) and by Rogers (1976) into a general ported that mebendazole acts by inhibiting approach to therapeutic problems. Once a glucose uptake (Van den Bossche, 1972). Dr. parasite metabolite is known (whether identi- R. O. McCracken, working in my laboratory fied as a result of drug studies or not), known with Hymenolepis diminuta, observed a drastic or suspected antagonists may be tested as depletion in glycogen content following me- chemotherapeutic agents. The degree of ratio- bendazole treatment, but was unable to nality in a particular case depends on the demonstrate an inhibition of glucose uptake specificity of the available information. Com- (suggesting an impaired glycogenesis or gly- pounds may even be screened randomly for cogenolysis). We are thus confronted with the antagonism to a given metabolite. Further, possibility that a group of chemically related compounds known to be antimetabolites in one compounds might contain anthelmintics that biological system may be tested "empirically" in operate in two entirely distinct ways; and that a system in which a role for the metabolite in a given drug might affect different helminths question has not been demonstrated. Thus in entirely different ways. empirical and rational components can be over- We need to know how the existing drugs lapping and mutually reinforcing. work—not to get more of the same, but to get The semirational approach has its own more understanding. We need insight into the hazards. It has been proposed (as already in- host parasite relationship so that we can select dicated) that the anthelmintic activity of thia- radically different and exciting targets for the bendazole derives in some way from its inhibi- empiricists to shoot at. The practical utility tion of fumarate reductase in the parasite of discovering the site-location mechanism of (Prichard, 1973). The effect of the drug on a helminth within the host would probably not this enzyme can be demonstrated in vitro. One lie in the design of a molecule that would block

Copyright © 2011, The Helminthological Society of Washington OF WASHINGTON, VOLUME 44, NUMBER 1, JANUARY 1977 27 that mechanism—but in the creation of a Literature Cited screening assay to detect compounds that Anderson, N., R. S. Morris, and I. K. McTag- would block that mechanism. We need to gart. 1976. An economic analysis of two know why some members of a worm popula- schemes for the antlielmintic control of hel- tion are not removed by a treatment that re- minthiasis in weaned lambs. Aust. Vet. J. moves most of them. Many years ago Stoll 52: 174-180. (1962) cautioned us against the presumption Anonymous. 1975a. Tropical diseases today— that it is desirable to remove 100% of the the challenge and the opportunity. World worms present. We have not yet come to grips Health Organization. Geneva, 1-15. with that problem (perhaps the drug-resistance Anonymous. 1975b. Editorial. J. Trop. Med. and immunity components of the problem are and Hyg. 78: 193. in conflict). We need to know about extrinsic Archer, S., and A. Yarinsky. 1972. Recent developments in the chemotherapy of schis- factors that affect the success of treatment. tosomiasis. Progress in Drug Research 16: About 100 years ago, the renowned Kuchen- 11-66. meister examined his own feces for pinworm for Campbell, W. C., and A. C. Cuckler. 1964. 329 successive days, and correlated his findings Effect of thiabendazole upon the enteral and with the phases of the moon in order to de- parenteral phases of trichinosis in mice. J. termine the best time for treatment (Kuchen- Parasit. 50: 481-488. meister, 1857). We need to follow his example and . 1967. Inhibition of egg —in principle. We need biochemical exploita- production of Schistosoma mansoni in mice tion of the innumerable mysterious o b s e r v a t i o n s treated with nicarbazin. J. Parasit. 53: 977- that fly from the empiricist's benchlike sparks 980. from a blacksmith's forge. Usually they fade Davies, P., and H. Jackson. 1970. Experimen- tal studies on the chemosterilization of Schis- away without igniting any interests on the part tosoma. mansoni. Parasitol. 61: 167-176. of those who could best make use of them. Gerwel, C., Z. Pawlowski, W. Kociecka, and L. Most new observations of antiparasitic efficacy Chodera. 1974. Probable sterilization of do not lead to useful new drugs—but they Trichinella spiralis by thiabendazole: fur- might teach us a great deal. This is especially ther clinical observation of human infections. so when high therapeutic efficacy is exhibited In: Kim, C . W. (editor) Trichinellosis. In- by a chemical that is relatively simple and bio- text Educational Publishers. New York. 471- chemically well characterized. Similarly we 475. might learn much from a chemical that kills Gordon, H. McL. 1963. The efficiency of new one species but not another of the same genus, anthelmintics for sheep. Proceedings World or a chemical that destroys the vitellaria of one Veterinary Congress, Hanover, Germany, trematode but not another. Such observations, Volume 1, 717-724. Hkchings, G. H. 1972. Inhibitors of folate which must abound wherever screening is con- biosynthesis and utilization—evolutionary ducted, are not to be regarded as crumbs changes as a basis for chemotherapy. Ann. for the jaws of Lazarus b u t as grain for the grist Repts. Med. C h e m . 7: 1-5. mills of the biochemists and physiologists. Jal'l'e, J. J. 1972. Dihydrofolate reductases in If we are really to resolve the three-ring parasitic protozoa and helminths. In: Van conundrum that we discuss today, we must den Bossche, H. (editor). Comparative Bio- make use of all kinds of research—all disci- chemistry of Parasites. Academic Press, New plines and all approaches. It is from such wide- York and London. 219-233. ranging research that the big revolutions will Kinnamon, K. E., and W. E. Rothe. 1975. emerge—not just some new drug. It is in the Biological screening in the U.S. Army anti- nature of things that we cannot imagine the malarial drug development program. Amer. nature of those revolutions any more than Gal- J. Trop. Med. H y g . 2 4 : 174-178. vani could imagine the lights of Times Square. Kuchenmeister, F. 1857. Animal and vegetable parasites of the human body. Sydenham So- We have come a long way since the time ciety, London. Vol. I , 452 pp. when a person with worms might have a ram's Leake, C. D. 1975. An historical account of tail inserted in his rectum so that the worms pharmacology to the 20th century. Charles could gnaw at that, and so be easily removed C Thomas, Springfield, Illinois, pp. 210. —but we still have a long way to go. Machado, A. B. M., M. L. H. da Silva, and

J. Pellegrino. 1970. Mechanism of action Antiparasitic drugs. V. Anticoccidial activity of thiosinamine, a new egg suppressive agent of 4-amino-2-ethoxybenzoic acid and related in schistosomiasis. J. Parasit. 56: 392-393. compounds. Proc. Soc. Exper. Biol. Med. Pellegrino, J., and N. Katz. 1975. [Experi- H7: 488-492. mental chemotherapy of schistosomiasis. X. Smith, C. E. G. 1976. Medicine in a develop- Antischistosomal activity of diaminodiphenyl- ing tropical environment. Trans. Roy. Soc. sulphone (DDS) in laboratory animals and Trop. Med. and Hyg. 7 0 : 1-9. in humans]. In Portuguese. Revista Inst. Stoll, N. R. 1962. Chairman's introduction. In: Med. Trop. S. Paulo, 17: 199-205. Goodwin, L. G. and R. H. Nimmo-Smith Prichard, R. K. 1973. The fumarate reductase (editors) Drugs, Parasites and Hosts. Little, reaction of Haemonchus contortus and the Brown and Company, Boston 1962. 3-14. mode of action of some anthelmintics. Intl. Van den Bossche, H. 1972. Biochemical ef- J. Parasit. 3: 409-417. fects of the antihelmintic drug mebendazole. Raeymaekers, A. H. M., F. T. N. Allewijn, J. In: Van den Bossche, H. (editor) Com- Vandenberk, P. J. A. Demden, T. T. T. parative Biochemistry of Parasites. Academic Van Oifemvert, and P. A. J. Janssen. Press, New York and London. 139-157. 1966. Novel broad-spectrum anthelmintics. Wang, C. C., R. L. Stotish, and M. Poe. 1975. Tetramisole and related derivatives of 6-aryl- Dihydrofolate reductase from Eimeria te- imidazo [2,1-b] thiazole. J. Med. C h e m . 9: nella: rationalization of chemotherapeutic 545-549. efficacy of pyrimethamine. J. Protozool. 22: Rogers, E. F. 1976. The antimetabolite concept 564-568. in drug design. Ann. Repts. Med. Chem. In Wyss, O., M. Rubin, and F. B. Strandskov. press. 1943. Mechanism of sulfonamide action. Rogers, E. F., R. L. Clark, H. J. Becker, A. A. III. Biological action of substituted p-amino- Pessolano, W. J. Leanza, E. C. McManus, benzoic acids. Proc. Soc. Exper. Biol. M e d . F. J. Andriuli, and A. C. Cuckler. 1964. 52: 155-158.

The Control of Parasites: The Role of the Host1

E. J. L. SOULSBY Department of Pathobiology, School of Veterinary Medicine, University of Pennsylvania, Philadelphia, Pennsylvania 19174

"It takes all the running you can do to keep in tence to take a holistic view of parasitism and the same place. to formulate concepts on the control of para- "If you want to get somewhere else, you must sitic infections. Nevertheless we are doing that run at least twice as fast as that. today and my particular task is to examine the Red Queen to Alice. role of the host in the control of parasitic infections. In doing so I intend to consider, The complex relationship between parasite briefly though it may be: some of the host and host poses problems of comprehension mechanisms which permit parasitism to occur which become more acute every year. It is in the first place and which, if manipulated, with an increasing degree of trepidation that might provide the basis for control; the host one gets out of the rut of ones supposed compe- mechanisms which may be abrogated or avoided by parasites to permit their survival 1 An address presented at the 500th meeting of the Helminthological Society of Washington, held at New and finally the application of the host re- Dolton Center, School of Veterinary Medicine, University of Pennsylvania on May 15, 1976. sponses to the control of infections.