Banned Read online

  When the PHS announced that it was withdrawing support of all clinical investigations of malaria in the U.S. because malaria no longer constituted a public health threat, the army assumed financial support of the Stateville antimalarial study on January 1, 1950. Chen recommended against myopically limiting the study to primaquine, but in the course of subsequent studies, primaquine proved so effective against Korean malaria that development of new antimalarial drugs virtually ceased.20

  Before the incorporation of volunteers at Stateville Penitentiary, four of the Tox Lab scientists served as subjects for a study, a clear indication of their dedication to the research. Kelsey, Oldham, Dearborn, M. Silverman, and E. W. Lewis monitored the excretion of atabrine in urine. All of the “subjects” developed mild toxic symptoms in response to a daily dose for forty-five days, and the excreted amount of atabrine never amounted to more than 11 percent of a daily dose. Even fifty-five days after the last dose was administered, there were appreciable amounts of atabrine in the subjects’ urine.21 Studies involving self-experimentation suggest that the Tox Lab scientists were passionately committed to their research. At the same time, willingness to participate on the part of scientists may also indicate that they were not concerned about the toxicity of antimalarial drugs.

  One of the most important studies that Chen and Geiling carried out under the OSRD grant assessed the joint toxicity of several antimalarial drugs.22 With the considerable development of antimalarial drugs during and immediately following World War II, some physicians began to experiment with combinations of drugs with the expectation that they might be more effective than an individual drug in curing of disease. Chen and Geiling sought to determine the joint toxicity in the host as well as the efficacy of various combinations of several antimalarial drugs in mice. For guidance regarding dosage-mortality relationships, they turned, once again, to the research of Chester I. Bliss, who, as we have seen, had developed rigorous biostatistical approaches to dose-mortality curves and the LD50 for individual drugs. In addition, Bliss devised statistical methods for the evaluation of joint toxicities. The subject was of interest to Bliss on theoretical grounds: “What was the impact of multiple chemicals on a toxicity curve?” But Bliss also found joint toxicity interesting for practical reasons, particularly as it related to the search for new insecticides: “In the search for new insecticides combined poisons offer many possibilities, but criteria are needed for separating mixtures in which the combined ingredients possess an enhanced toxicity from others in which they act independently since the former group provides the more promising field of investigation.” Bliss cited a study of the toxicity of two pesticides (rotenone and pyrethrin) used in combination in which the authors did not find evidence of synergism while another researcher utilized the same original data and discovered definite evidence of synergism.23

  In order to resolve such confusion, Bliss provided the definition and quantitative analysis of three kinds of joint toxic action in which the percentage mortality was employed as the measure of response. In the case of “independent joint action,” the poisons or drugs acted independently and had different modes of mortality. Susceptibility of an organism to one component might or might not be correlated with susceptibility to the other. Quantitatively, the toxicity of the mixture could be predicted from the dosage-mortality curve for each constituent applied alone and the correlation in susceptibility to the two poisons. Bliss employed the term “similar joint action” for poisons or drugs that produced similar but independent effects, such that one component could be substituted at a constant proportion for the other. Individual susceptibility would be completely correlated or parallel. Quantitative calculation of the toxicity of compounds with similar joint action could be predicted directly from the toxicities of the constituents as long as their relative proportion was known. Finally, and perhaps most significant, Bliss delineated “synergistic action,” in which the effectiveness or toxicity of a chemical mixture could not be assessed from that of the individual components but rather depended on knowledge of the chemicals’ joint toxicity when used in different proportions. Synergistic action had the most serious implications for pharmacology and toxicology because one component exacerbated or diminished the effect of the other.24 As Bliss predicted, his research and methodologies had wide application in the development and applications of drugs, insecticides, and other chemical mixtures.

  For their part, Chen and Geiling directly applied Bliss’s definitions and methods to antimalarial drugs. Atabrine and quinine, for example, acted in an independent and similar manner, as did quinine and hydroxyethylapocupreine. However, the combinations of quinine and pamaquine, as well as quinine and pentaquine, were much more toxic than predicted from their individual toxicities, clear cases of synergism in the two combinations. The dosage-mortality curves looked like those for different drugs rather than the summation of the curves for the individual drugs. Chen and Geiling explained the joint toxicity of atabrine and quinine by suggesting a common site of action, but they were at a loss to explain the synergism between quinine and pamaquine: “Since only a very small amount of quinine, 1/30 of the minimal lethal dose, is sufficient to reveal its synergistic action with a minimal lethal dose of pamaquine in acute mortality, the site and the mechanism of this action of quinine are evidently different from those causing sudden death of animals with a lethal dose of quinine.”25 Speculatively, Chen and Geiling suggested that the joint toxicity might result from an effect on an enzymatic process essential for life. Emphasizing acute toxicity, their paper mentioned chronic toxicity only in passing, cautioning that it may (or may not) correlate to acute toxicity.26 This important distinction was often overlooked in the early toxicology and pharmacological literature. As we will see in chapter 4 and remaining chapters, joint toxicity became an issue of central importance in the study and legislation of pesticides.

  Chen and Geiling also attempted to reveal the nature of drug resistance in a continuing study of drugs to treat trypanosomes (parasitic protozoans that cause trypanosomiasis), specifically “trypanocidal activity,” in lab mice. Their first report on this research described a simple, quantitative method of assay of the therapeutic activity of antimonials and provided a comparison of the trypanocidal potency and the toxicity of well-known organic antimony preparations. They based the assay for the potency of trypanocidal substances on the suppression of infection and the cure of the disease. In a theoretical sense, the most important finding of Chen and Geiling’s initial study of trypanosomes was the quantitative evaluation of the “therapeutic index,” or the ratio of maximal tolerated dose to minimal curative dose. Earlier researchers developed the therapeutic index in a qualitative sense without regard to biological variation. Citing recent advances in quantitative pharmacology, Chen and Geiling adopted the 50 percent level (i.e., the ratio of maximal tolerated dose to minimal curative dose was two to one). They reasoned that the weight of an observation was greatest when the effect was 50 percent, which was highly desirable in an assay like the one they devised in which the number of animals was small.27

  To address the problem of drug resistance, Chen and Geiling first developed an in vitro procedure for determining the antitrypanosome effect of antimonials.28 With this procedure, a doctoral candidate (F. W. Schueler) joined Geiling and Chen, to complete the research necessary for his dissertation, and they developed a criterion of resistance based upon the inhibiting power of mapharsen (a standard drug of reference) on the glucose utilization by trypanosomes. The resistance factor was equal to the ratio of the 50 percent suppressive dose of mapharsen on glucose utilization for the resistant mouse strain to the 50 percent suppressive dose of mapharsen for the normal mouse strain. Schueler, Chen, and Geiling suggested that the determination of parasite resistance could be based on a criterion of 15 percent suppression for toxicity studies or 90 percent suppression for investigations of lethal dosages rather than on the 50 percent suppression level they used.29 Thus the methodology could cover the full range of
studies in toxicity.

  The study of antimalarial drug therapies began as an attempt to replace and improve upon existing antimalarial drugs for the war effort. The war’s end brought about changes in emphasis and funding, but Chen, Geiling, and others continued and expanded their research on antimalarials. In addition to expanding the constellation of drugs that controlled malaria, Chen and Geiling addressed more subtle aspects of the drugs, such as joint toxicity and resistance. Joint toxicity or potentiation and resistance became significant problems as toxicologists examined the toxicity of synthetic insecticides. Researchers drew upon techniques developed for antimalarial drug therapies in their subsequent research with pesticides.

  In addition to studying antimalarial drug therapies, scientists at the Toxicity Laboratory at the University of Chicago devoted considerable effort to the analysis of nitrogen mustards.30 In 1942, C. C. Lushbaugh, a pathologist in the laboratory, noted that mice gassed with nitrogen mustards had many fewer white blood cells than normal, and that the bone marrow and lymph nodes of the animals no longer formed blood cells. This discovery prompted Leon Jacobsen and Charles Spurr in the Department of Medicine at the University of Chicago to test the effectiveness of the nitrogen mustards against certain diseases such as leukemia, lymphosarcoma, and Hodgkin’s disease.31 This research resulted in some of the first chemotherapeutic agents against cancers.32 One symptom associated with all these diseases is the presence of abnormally large numbers of white blood cells. The mustard compounds affected the blood cells much like X-rays. The research at the Tox Lab complemented research conducted in the Department of Pharmacology at Yale University (also sponsored by the OSRD), which revealed the potential of nitrogen mustards for use in chemotherapy.33

  During the war, the Toxicity Laboratory was committed to the discovery and analysis of the most toxic compounds, either for potential use in combat or in anticipation of what the enemy might do. At the close of the war, however, members of the Tox Lab returned to some of the nitrogen mustard compounds they had deemed “relatively nontoxic” during the early part of the World War II. Some of these compounds affected white blood cells despite their lesser toxicity.34 After the war, Tox Lab researchers discovered that certain nitrogen mustards attacked all proliferating normal tissues, and they belonged to a group of compounds known as “mitotic arrestors.” Although the nitrogen mustards fell short of a cure for cancer, clinical trials revealed that they promoted extended remissions of the disease and reactivated the sensitivity of tumors to X-ray therapy. Both of these factors encouraged further research.35

  Even as physicists from the University of Chicago and elsewhere raced to complete research on the first atomic bomb in 1944, scientists had begun to explore the potential of nuclear research to benefit society. The result of these musings was the development of the Jeffries Committee, chaired by Zay Jeffries. Other members of the committee included R. S. Mulliken (secretary), Enrico Fermi, James Franck, T. R. Harness, R. S. Stone, and C. A. Thomas. The Jeffries Committee convened to determine insofar as possible the future of a new field, which they called “nucleonics.” In biology and medicine, the most promising avenue of research seemed to be the use of radioisotopes to examine basic problems in animal and plant metabolism, such as respiration, photosynthesis, fat and protein metabolism, and minor problems, for example, the role of micro-nutrients.36

  In 1945, the CWS assumed the funding for the Tox Lab, but by the close of the war, its parent organization, the OSRD, had disbanded. The lab changed hands once again in 1947 and entered into a contract with the Atomic Energy Commission (AEC). The new source of funding attracted the interest of Toxicity Laboratory researchers to the new technology of radioisotope markers. Before the war, pharmacologists had to rely on chemical agents as tracers in pharmacological research. Such tracers included certain dyes or chemicals with specific properties.

  Radioisotopes promised to greatly enhance and improve the ability of pharmacologists to trace the pathways of a given drug. Unlike chemical tracers, which diluted or otherwise affected the makeup of a drug, radioisotopes led to “natural tracers,” suggesting that they shared identical atomic structure with nonradioactive drugs. In fact, radioactive drugs were almost exactly like their nonradioactive counterparts except that some of the atoms in the labeled drug were radioactive and emitted radiation which scientists could track with the use of highly sensitive instruments, such as the Geiger counter, the scintillation counter, or the ionization chamber. With these devices Geiling and his colleagues could detect the presence of the smallest amounts of a drug and all its metabolized products in all organs and tissues of the body.

  Geiling enumerated the numerous challenges posed by certain drugs that the use of radioisotopes in pharmacology could address: “A number of our most useful drugs have (1) complex chemical structures, (2) cannot as yet be readily prepared in the chemical laboratory, (3) are administered to patients in such small doses that, when distributed in the tissues and body fluids, the conventional biological and chemical methods are inadequate, (4) some of the available methods may be able to detect the unchanged drugs, but not the metabolites, (5) another important advantage of using labeled drugs is that they can be studied at a therapeutic or sub-toxic level.”37 Earlier drug distribution studies received the criticism that the doses used were well above the therapeutic level and at times even in the toxic range. The lack of sensitivity of tracer chemicals had required experimenters to use large doses.

  With the cooperation of botanists, zoologists, organic chemists, and individuals trained in radioisotope techniques, Geiling and his colleagues developed methods to produce radioactive drugs using carbon-14. To do this, they grew plants like digitalis (Digitalis purpurea) and nicotine (Nicotiana rustica) in a closed system into which they introduced radioactive carbon dioxide (which the plants absorbed during the process of respiration). By drying and processing the digitalis one could obtain radioactive digitoxin; running nicotine through the same process produced radioactive nicotine. In both cases, Geiling was able to demonstrate a high degree of purity for the drugs. Even in the earliest, exploratory experiments with carbon-14, Geiling believed that the technique was far superior to other methods of marking: “This may be an advantage in the use of such materials in biological problems, since it permits the tracing of all carbon-containing metabolic fragments of the drug rather than only the single atom usually labeled in synthetic drugs.”38 Further experimentation with radioactive digitoxin revealed that the effect of the drug was not cumulative: “Our preliminary experiments indicate clearly that digitoxin cannot be regarded as a cumulative drug since it is largely metabolized and excreted in a relatively short time. The mechanism of digitoxin action thus needs to be reappraised with the aid of these new techniques.”39 This further research raised Geiling’s expectations for the new technique in human experiments: “The use of radioactive digitoxin in suitable patients should throw considerable light on the metabolic pathway of this important agent.”40

  E. M. K. Geiling and a colleague preparing radioactive digitoxin. Courtesy of the University of Chicago Photographic Archives apf 1–06306, Special Collections Research Center, University of Chicago Library.

  The use of radioisotopes as tracers was not limited to drugs synthesized from plants like digitalis and belladonna. A young doctoral student named John Doull (b. 1922) developed a method for the biosynthesis of radioactive bufagin as the basis for his doctoral thesis. Under the supervision of Geiling and Kenneth DuBois (a faculty member recruited during the war), Doull fed tropical toads (Bufo marinus) with radioactive algae supplied by W. F. Libby of the Institute for Nuclear Studies at the University of Chicago. Radioactive carbon-14 was distributed throughout the algae, which had been grown in a closed atmosphere containing radioactive carbon dioxide. As a preliminary test, Doull fed slugs on radioactive lettuce prepared according to Geiling’s methods. Toads consumed the radioactive carbon-14 either in liver mixed with the algae or in the slugs. Doull went on to extract radioactiv
e venom from the parotid glands of the toads, from which he then tried to isolate the bufagin by fractional crystallization. He acknowledged, however, that the effort required to extract venom of high radioactivity was much greater than to incorporate radioactive carbon dioxide into plant principles through photosynthesis.41

  Having worked out the biosynthesis of radioactive digitoxin and other drugs, Geiling and the Tox Lab researchers employed the new technology to advance their knowledge of drug action. For example, radioactive digitoxin showed that the drug crossed the placental membrane in rats and guinea pigs. This finding had several grave implications for the use of drug therapies in pregnancy. First, embryonic tissue showed a marked ability to catabolize digitoxin (break it down into metabolites), or there may have been a selective penetration of the digitoxin metabolites across the placenta (indicated by the high metabolite-digitoxin ratio). Second, on a tissue-to-weight ratio, digitoxin and its metabolites appeared more concentrated in the embryonic heart than in the maternal heart.42 The latter finding suggested that a given drug therapy for a mother would subject the developing fetus to a much higher dose of the drug. As we have seen in the case of joint toxicity and resistance, the differential effect of drugs and other chemicals on mothers and fetuses would become very important in the study of insecticides.