Essentiality and Therapeutic Uses
THIS chapter is organized into four sections. The first section defines essentiality and specifies the criteria that have been used to determine whether a substance is an essential nutrient. The second section summarizes the evidence for or against the hypothesis that arsenic is an essential nutrient. The third section summarizes what is known about the therapeutic uses of arsenic.
Definition Of Essentiality
There is general agreement about the criteria needed to identify a substance as an essential nutrient (e.g., Cotzias 1967; Mertz 1970; Underwood and Mertz 1987; FAOIAEA-WHO 1996). The criteria are as follows:
1. The substance is present in all organisms for which it is essential.
2. Reduction of exposure to the substance below a certain limit results consistently and reproducibly in an impairment of physiologically important functions, and restitution of the substance under otherwise identical conditions prevents the impairment.
3. The severity of signs of deficiency increases in proportion to the reduction of exposure to the substance.
An additional criterion was proposed by Cotzias (1967)
4. The abnormalities produced by a substance's deficiency should always be accompanied by specific biochemical changesthat is, the biochemical mechanisms of action should be known.
Of the four criteria, the first two are crucial; if both are met, a substance under investigation is considered essential for that particular animal species. Satisfying the third and fourth criteria requires additional effort and time and is a goal of nutrition research. That goal is consistent with the continuous refinement of knowledge concerning the molecular mechanisms for nutrients that have produced substantial public-health benefits for decades (e.g., selenium and zinc) or even centuries (e.g., ascorbic acid).
Evidence For Essentiality
As stated above, two criteria must be met before a substance under investigation can be considered essential in a particular animal species: (1) that it is present in all organisms for which it is essential, and (2) that reduction of exposure below a certain limit results consistently and reproducibly in a reduction of physiologically important functions. It is universally accepted that arsenic is present in living matter. The remainder of this section, therefore, examines the evidence in support of the second criterion. That discussion is followed by consideration of information of secondary importance pertaining to the hypothesis of essentiality. This includes information on biochemical mechanism of action, similarities with selenium, and dose-response relationships.
A function is considered physiologically important when its impairment interferes with the normal development or survival of the individual (or the whole species). Data are presented below that examine the physiological importance of arsenic in goats, minipigs, rats, and chicks and in particular how semisynthetic diets with low arsenic concentrations affect reproduction and growth of young animals.
Essentiality in Humans
Arsenic has not been tested for essentiality in humans nor has it been found to be required for any essential biochemical processes.
Goats and Minipigs
Goats and minipigs fed semisynthetic diets low in arsenic (arsenic at less
than 35 ng/g of diet) produced offspring with an average depression of birth weight of 13% compared with the offspring of animals fed diets with arsenic at 350 ng/g of diet (p < 0.001 in both species) (Anke et al. 1976; Anke 1986, 1991; see Addendum to Chapter 9 for description of diets). Female goats fed the low-arsenic semisynthetic diet each produced on average 0.96 kids compared with 1.4 kids per arsenic-supplemented goat (p < 0.0010). The corresponding reproduction rate in minipigs was 6.3 compared with 12.8 (p < 0.001) for sows fed the low-arsenic semisynthetic or the arsenic-supplemented diet, respectively. An average of 0.8 kids delivered per goat fed the low-arsenic diet survived to the end of the suckling period of 91 days, compared with 1.4 kids per arsenic-supplemented goat (p < 0.05). The corresponding survival for minipigs was 2.5 compared with 5.9 (p < 0.01).
Most of the breeder goats fed the low-arsenic semisynthetic diet were reported to have died suddenly between the 17th and 35th day of their second lactation, and none of the low-arsenic-diet goats survived the second pregnancy. The low-arsenic-diet goats that did not get pregnant survived to more than 6 years of age (Anke 1991). Autopsies of the low-arsenic-diet goats revealed atrophy of cardiac and striated muscle fibers and distinct reduction of oxidative enzyme activity associated with rupture of liver, heart, and muscle mitochondrial membranes (Schmidt et al. 1984).
Growth retardation in rats fed low-arsenic semisynthetic diets has been reported independently by two groups of investigators. Addition of sodium arsenite at 0.5, 0.75, 1, and 2 µg/g of an amino-acid diet containing arsenic at 50 ng/g stimulated growth rates by 23%, 14%, 17% and 15%, respectively (Schwarz 1977). Those studies were not repeated because of the death of the investigator. Nielsen et al. (1978) reported strong growth depression and premature death in rats fed a milk-powder-based diet containing arsenic at 30 ng/g of diet compared with controls supplemented with arsenic at 4.5 µg/g of diet. Those studies also were not repeated because the supply of the low-arsenic milk powder ran out. Using a casein-based and ground-corn diet containing arsenic at 15 ng/g, Uthus et al. (1983) detected growth depression in a three-generation study; findings were more pronounced in males than in females. In addition, that study and a replicate study detected reduced fertility and litter size in low-arsenic-diet animals. Apparently, however, those results were not consistently obtained with the casein-based diet. In addition to the low-arsenic state, the investigators began imposing various forms of dietary stress, such as deficient or excessive supplies of certain nutrients or metabo-
lites. For example, supplementation with arsenic at 1 g/g of diet (see Addendum for description of diet) stimulated growth by approximately 40% in methionine-deficient (1 g/kg of diet) rats with reduced growth, but the resulting growth rates were still less than those fed a diet with adequate methionine content. In the latter, arsenic supplements had no effect (Uthus 1992).
In two initial experiments, chicks were fed a milk-powder-based diet with arsenic at 20 ng/g supplemented with arginine at 20 g/kg of diet. The addition of arsenic at 1 µg/g of diet produced a significant growth effect, as it did in a third experiment using a milk-powder-based diet containing arsenic at 35 ng/g of diet. A fourth experiment using a diet containing arsenic at 45 ng/g of diet, however, showed no stimulation of growth when arsenic was added (Nielsen 1980) (see Table 9-1). Those studies were not repeated because the supply of low-arsenic milk powder ran out. In subsequent studies using a casein-based diet with arginine and containing arsenic at 5 ng/g of diet, the addition of 2 µg of arsenic produced a significant (p < 0.004) growth stimulation (894 g compared with 747 g of body weight at 4 weeks; 24 chicks in each group). In a repeat experiment using different concentrations of zinc, arsenic stimulated growth to a lesser degree (Nielsen 1980). In another experiment, compounds that affect methyl-group availability were added to a diet containing arsenic at 15 ng/g of diet (Uthus 1992). The addition of guanidoacetate at 5 g/kg of feed strongly depressed growth rates in chicks. The addition of arsenic at 2 µg/g of diet stimulated growth significantly in the stressed chicks (p < 0.01). Arsenic was ineffective in stimulating growth in chicks that were not stressed.
TABLE 9-1 Weight of Chicks After 28 Days on Diet
Arsenic Content (ng/g)
(No. of Chicks)
521 ± 9
580 ± 17
586 ± 16
674 ± 25
699 ± 32
768 ± 15
684 ± 20
686 ± 19
Source: Adapted from Nielsen 1980.
Data Consistency and Reproducibility
In a series of experiments conducted since 1973, Anke (1991) evaluated the effects of a low-arsenic semisynthetic diet in goats. He repeated the experiments 12 times. The initial concentration of arsenic in the semisynthetic diets fed to goats was approximately 35 ng/g of diet. The concentration was gradually reduced to approximately 10 ng/g of diet. Anke (1991) reported that within the expected fluctuations, the results are considered consistent and reproducible. Confirmation by independent investigators is still lacking. Studies with minipigs, although not as extensive as those with goats, produced similar effects on growth and reproduction and can be considered consistent with the exception that sudden death of lactating animals was not observed in minipigs. Independent confirmation of those results is still lacking.
Two groups of investigators have independently and almost simultaneously observed growth stimulation by arsenic in rats raised in a protected environment and fed low-arsenic diets (Schwarz 1977; Nielsen et al. 1978). Schwarz (1977) used a highly purified synthetic diet based on amino acids (arsenic at 50 ng/g), whereas Nielsen et al. (1978) fed rats a milk-powder-based semisynthetic ration (35 ng/g). Neither experiment was repeated under identical dietary conditions for reasons mentioned above. In subsequent studies with casein-based diets (15 ng/g), Nielsen (1980) found growth stimulation in some experiments, but not consistently so, unless some form of additional nutritional stress was superimposed. With additional stress, such as methionine deficiency or addition of compounds that affect methyl-group availability, consistent effects of arsenic on growth were reported. Providing supplemental arsenic to chicks on low-arsenic diets with superimposed nutritional stresses stimulated growth consistently (Uthus 1992).
Although the studies reviewed here have not been independently confirmed under identical experimental conditions, all replications by the authors have been consistent in goats and minipigs fed semisynthetic diets with low arsenic content, as well as in rats and chicks subjected to additional dietary stresses.
Biochemical Mechanism of Action
Although there is convincing evidence that the symptoms reported for the rats and chicks fed a low-arsenic semisynthetic diet can be modulated by substances that affect methionine metabolism (and perhaps, methyl groups in general), the mechanisms and sequence of events leading to the functional impairments described above are still unknown.
On the basis of the review of studies in four animal species, signs of depressed reproductive performance and growth were observed when the animals were raised in a controlled environment (see Addendum to Chapter 9) and fed semisynthetic diets casein-based, containing arsenic at 50 ng/g or less. The experimental diets met the nutritional requirements of the respective species; their protein sources ranged from purified amino acids to milk powder, casein, and urea. Although diets with arsenic as low as 5 and 10 ng/g were administered, a clear relationship between the arsenic content and the severity of the signs in goats, minipigs, and rats was not established. Chicks, on the other hand, fed diets containing arsenic at 20 ng/g exhibited the most marked impairment. Diets with arsenic content at 35 ng/g of diet still produced some impairment; at 45 ng/g, the chicks were no different from their controls (Nielsen 1980). It should be noted that the form of arsenic in the semisynthetic diets is not known. All these observations relate to the total arsenic naturally present in the diets. No speciation of dietary arsenic had been done and therefore the contribution of different arsenic species to the total diet is not known. The investigators also did not monitor concentrations of arsenic in tissues or urine, and thus, it is not known if dietary arsenic was absorbed and present in tissue. In addition, viral infections had not been checked for. In at least one standard experimental diet with a natural arsenic content of only 5 ng/g (Reeves et al. 1993), evidence of functional impairments in rats and mice was not observed.
As shown in Table 9-2, the concentrations of arsenic in the semisynthetic diets overlap with the concentrations of arsenic found in standard experimental diets, which range from 5 to 6,200 ng/g, with substantial variation between different batches of the same formulation (Reeves 1993). Probably, the diets with high concentrations of arsenic contain organic arsenic compounds. One study reported that mouse diets with an arsenic concentration of 500 ng/g contained fish meal, which is likely to contain arsenobetaine, as a source of protein (Vahter and Norin 1980). Also, the concentrations of arsenic in the semisynthetic diets are similar to those found in most human foods, except seafood (i.e., 10-200 ng/g) (see Chapter 3, Table 3-7). Thus, it is not clear whether the above mentioned effects were due to the low arsenic concentrations or to other factors in the prepared semisynthetic diets.
Inorganic arsenic supplements have ranged from 350 to 4,500 ng/g. Although few data establish a dose-response relationship for supplements of inorganic arsenic, a dose of 350 ng/g prevented impairment of reproductive functions and growth in goats and minipigs, as well as sudden cardiac death in lactating goats. Supplements ranging from 500 to 4,500 ng/g were reported
to stimulate growth in rats and chicks fed semisynthetic diets, often with additional dietary stress.
Table 9-2 Effects of Dietary Arsenic in Selected Species
Reduced growth and reproduction
5-50 ng/gc often with additional dietary stress
Goats, minipigs, rats,
bRange of concentrations (dry weight) in nonmarine foods (see Ch. 3, Table 3-7).
The highest supplementation dose (4,500 ng/g) corresponds to a daily dose of several milligrams of inorganic arsenic for an adult human (i.e., in the range of the therapeutic doses of arsenic used for Fowler's solution) (see the following section). The lowest supplemented dose (350 ng/g) would correspond to more than 100 µg of arsenic per day for an adult human. In the more recent studies on rats and chicks (Uthus 1992), arsenic had a growth-stimulating effect only in animals that were subject to additional dietary stress in the form of methionine deficiency or excess arginine or guanidoacetic acid. The supplemented concentrations (arsenic at 1-2 µg/g of diet) would correspond to 1 mg or more of arsenic per day for an adult human. The foregoing discussion of inorganic arsenic as an essential nutrient in animals should be distinguished from the use of aryl arsenic veterinary medications as additives to animal feed (Adams et al. 1994).
In conclusion, studies to date do not give evidence that arsenic is an essential element in humans or that it is required for any essential biochemical process. At very high doses (concentrations of 350-4,500 ng/g in the diet), arsenic supplementation seems to have a growth-stimulating effect in minipigs, chicks, goats, and rats.
Therapeutic Uses of Arsenic
The introduction of inorganic arsenic as a therapeutic agent in the modern medical era is generally attributed to Thomas Fowler, a British physician whose treatise ''Medical Reports of the Effects of Arsenic in the Cure of
Agues, Remitting Fevers, and Periodic Headaches" was published in 1786. After determining that arsenic was a key ingredient in a locally sold patent medicine, Fowler prepared and used a "mineral solution" containing approximately 1% arsenic trioxide to treat "agues," or malarial fevers. At an oral adult dose equivalent to approximately 11.4 mg of inorganic arsenite per day, Fowler reported therapeutic success in 242 of 247 patients. He did, however, note that "about one-third" of the patients thus treated experienced "operative effects'' (side effects) consisting of nausea, vomiting, or abdominal pain. Fowler's mineral solution quickly gained recognition as a therapeutic agent, and under the name Liquor Arsenicalis, it became officially listed in the London Pharmacopoeia beginning in 1809 and the U.S. Pharmacopoeia in 1820 (Langehan 1921). During the first half of the nineteenth century, "Fowler's solution," as Liquor Arsenicalis was commonly known, was advocated by many physicians often at widely variable doses for a broad spectrum of symptoms and illnesses (Haller 1975). By the latter half of the nineteenth century, Fowler's solution was recommended mainly for the treatment of skin diseases (particularly eczema, psoriasis, and pemphigus), asthma, chorea (probably in association with rheumatic fever), periodic fevers (e.g., from malaria), and pain. The noted physician Sir William Osler, writing in the first edition of his textbook Principles and Practice of Medicine, recommended inorganic arsenic in the treatment of pernicious anemia, chorea, leukemia, and Hodgkin's disease (Osler 1894). Prescribed doses commonly delivered approximately 5-10 mg of inorganic arsenite orally per day (Farquharson 1880; Stockman 1902; Langehan 1921). The chronic use of inorganic arsenic in this manner was sometimes associated with the development of cutaneous hyperpigmentation or, less commonly, peripheral neuropathy and other multisystemic signs of chronic arsenic poisoning (Osler 1894; Stockman 1902; Pope 1902; Silver and Wainman 1952).
Inorganic arsenic continued to be used as a therapeutic agent through the mid-twentieth century, by which time its recognized uses were confined predominantly to leukemia, psoriasis, and chronic bronchial asthma (Goodman and Gilman 1955). In the 1950s, the chronic, often unsupervised use of Gay's solution containing potassium arsenite, digitalis, potassium iodide, and phenobarbital for asthma created controversy when reports of success were countered by reports of overt arsenic toxicity (Silver and Wainman 1952; Pascher and Wolf 1952; Gay 1954). In 1967, Harvard investigators Harter and Novitch (1967) reported the results of a controlled trial of Gay's solution in patients whose asthma was "intractable" to treatment with either bronchodilators alone or bronchodilators plus corticosteroids. The patients' pre-enrollment regimen was supplemented, in a double-blind manner, with variants of Gay's solution containing or lacking inorganic arsenic. "Definite" clinical
improvement was found within 10 days in 7 of 18 patient trials that included arsenic administration (5-6.7 mg of arsenite per day) compared with only 1 of 11 patient trials lacking arsenic (p = 0.007). Approximately one-fourth of the patients receiving arsenic manifested gastrointestinal toxicity. Although inorganic arsenic might still occasionally be encountered in non-Western traditional medicines or folk remedies (Kew et al. 1993; Espinoza et al. 1995), its availability in medications listed in official Western formularies ended in the 1970s.
Organic arsenic antibiotics were used extensively in the first half of the twentieth century, principally in the treatment of spirochetal and protozoal diseases (Goodman and Gilman 1955). The first such official agent, Salvarsan, or arsphenamine, was introduced by Ehrlich in 1907 for the treatment of syphilis. Arsphenamine and other trivalent derivatives, such as neoarsphenamine, were widely used as antisyphilitics in the first two decades of the century. They were replaced by more stable trivalent arsenoxides, such as oxophenarsine and dichlorophenarsine, in the 1930s and 1940s. The availability of penicillin in the 1940s and 1950s largely supplanted the use of anti-syphilitic arsenicals. Pentavalent arsonic acid derivatives such as tryparsamide, used for trypanosomiasis, and carbarsone, used for amebiasis, were used in the 1930s through the 1960s. By the 1980s, the only remaining organic arsenical was melarsoprol, available through the U.S. Centers for Disease Control and Prevention for treatment of the meningoencephalitic stage of African trypanosomiasis.
The precise mechanisms by which inorganic arsenic exerted salutary effects in treatment have not been elucidated, but it is of interest that its reported benefit in psoriasis, eczema, and bronchial asthma and its antipyretic effect in certain febrile diseases suggest that it might have exerted suppressive effects on immune-mediated inflammation. Recently, intravenous administration of arsenic trioxide (10 mg per day or 0.5 mg/kg per day) was reported to induce remission in acute promyelocytic leukemia (Shen et al. 1997; Soignet et al. 1998). Preliminary investigations suggest that the mechanism might involve induction of apoptosis (Look 1998).
Summary And Conclusions
In this chapter, the subcommittee reviewed the evidence for the beneficial effects or essentiality for arsenic from experimental studies. Data on the physiological importance of arsenic in goats, minipigs, rats, and chicks were considered. There are no comparable data for humans. Reported therapeutic effects of arsenic are also reviewed.
On the basis of its review of the data relating to the essentiality and therapeutic effects of arsenic, the subcommittee concludes the following:
· Arsenic has not been tested for essentiality in humans.
· Data from four species indicate that semisynthetic diets with arsenic concentrations in the range of 35 to 50 ng/g or less in combination with dietary or reproductive stress result in functional impairments. Such concentrations might occur naturally in some experimental diets and are similar to those found in most human foods except seafood. The mechanisms and sequence of events leading to functional impairments are not known.
· Studies show that arsenic supplementation of low-arsenic semisynthetic diets prevents the occurrence of abnormal reproductive performance in goats and minipigs (350 ng/g) and reduced growth in chicks, and rats (500 to 4,500 ng/g). Although the studies have had no independent confirmation under identical experimental conditions, replications by the original investigators have been consistent with goats and minipigs fed semisynthetic diets, as well as with rats and chicks subjected to additional dietary stress. Toxic effects of the supplementation have not been studied.
· Studies to date do not provide evidence that arsenic is an essential element in humans or that it is required for any essential biochemical process. Arsenic supplementation seems to have a growth-stimulating effect at very high doses in minipigs, chicks, goats, and rats. This conclusion is consistent with NRC (1989).
· Arsenic might have therapeutic properties for certain disorders.
Validated analytical data on arsenic concentrations in preparations for total parenteral nutrition (TPN) should be obtained and related to the health status of patients on long-term TPN.
Future studies on the beneficial effects of arsenic in experimental animals should carefully monitor the amount and speciation of arsenic in diets and water (as fed), use biomarkers to assess arsenic exposure and bioavailability, and use techniques that assess toxicity to and benefits from arsenic in a more specific manner than is possible through measurement of growth and reproductive success.
Adams, M.A., P.M. Bolger, and E.L. Gunderson. 1994. Dietary intake and
hazards of arsenic. Pp. 41-49 in Arsenic: Exposure and Health, W.R. Chappell, C.O. Abernathy, and C.R. Cother, eds. Northwood, U.K.: Science and Technology Letters.
Anke, M. 1986. Arsenic. Pp. 347-372 in Trace Elements in Human and Animal Nutrition, Vol. 2, 5th Ed., W. Mertz, ed. Orlando, Fla.: Academic.
Anke, M. 1991. The essentiality of ultra trace elements for reproduction and pre-and postnatal development. Pp. 119-144 in Trace Elements in Nutrition of ChildrenII, R.K. Chandra, ed. New York: Raven.
Anke, M., M. Grün, and M. Partschefeld. 1976. The essentiality of arsenic for animals. Pp. 403-409 in Trace Substances in Environmental HealthX, Proceedings of the University of Missouri's Tenth Annual Conference on Trace Substances in Environmental Health, D.D. Hemphill, ed. Columbia, Mo.: University of Missouri Press.
Cotzias, G.C. 1967. Importance of trace substances in environmental health as exemplified by manganese. Pp. 5-19 in Proceedings of the University of Missouri's First Annual Conference on Trace Substances in Environmental Health, D.D. Hemphill, ed. Columbia, Mo.: University of Missouri Press.
Espinoza, E.O., M.J. Mann, and B. Bleasdell. 1995. Arsenic and mercury in traditional Chinese herbal balls. N. Engl. J. Med. 333:803-804.
FAO-IAEA-WHO (Food and Agriculture Organization of the United Nations-International Atomic Energy Agency-World Health Organization). 1996. Trace Elerments in Human Nutrition and Health. Geneva: World Health Organization. 343 pp.
Farquharson, R. 1880. On the use of arsenic in skin-diseases. Br. Med. J. (May 29):802-804.
Fowler, T. 1786. Medical Reports of the Effects of Arsenic, in the Cure of Agues, Remitting Fevers, and Periodic Headaches. London: Johnson and Brown.
Gay, E.D. 1954. Asthma and arsenic. JAMA 156 (Dec. 25):1628.
Goodman, L.S., and A. Gilman. 1955. The Pharmacological Basis of Therapeutics. New York: Macmillan.
Haller. J.S. 1975. Therapeutic mule: The use of arsenic in the nineteenth century materia medica. Pharm. History 17(3):87-100.
Harter, J.G., and A.M. Novitch. 1967. An evaluation of Gay's solution in the treatment of asthma. J. Allergy 40:327-336.
Kew, J., C. Morris, A. Aihie, R. Fysh, S. Jones, D. Brooks. 1993. Arsenic and mercury intoxication due to Indian ethnic remedies. Br. Med. J. 306:506-507.
Langehan, H.A. 1921. A Century of the United States Pharmacopoeia: 1820-1920. Liquor Potassii Arsenitis. Bull. Univ. Wisc. Ser. No. 1153,
Gen. Ser. No. 936.
Look, AT. 1998. Arsenic and aproptosis in the treatment of acute promyelocytic leukemia. J. Natl. Cancer Inst. 90:86-88.
Mertz, W. 1970. Some aspects of nutritional trace element research. Fed. Proc. Fed. Am. Soc. Exp. Biol. 29:1482-1488.
Nielsen, F.H. 1980. Evidence of the essentiality of arsenic, nickel, and vanadium and their possible nutritional significance. Pp. 157-172 in Advances in Nutritional Research, Vol. 3, H.H. Draper, ed. New York: Plenum.
Nielsen, F.H., D.R. Myron, and E. O. Uthus. 1978. Newer trace elements-Vanadium (V) and arsenic (As) deficiency signs and possible metabolic roles. Pp. 244-247 in Trace Element Metabolism in Man and Animals, Vol. 3, M. Kirchgessner, ed. Freising-Weihenstephan, Germany: Technische Universitat Munchen.
NRC (National Research Council). 1989. Recommended Dietary Allowances, 10th Ed. Washington, D.C.: National Academy Press.
Osler, W. 1894. Principles and Practice of Medicine. New York: Appleton.
Pascher, F. and J. Wolf. 1952. Cutaneous sequelae following treatment of bronchial asthma with inorganic arsenic: Report of two cases. JAMA 148:734-736.
Pope, F.M. 1902. Arsenic in the treatment of chorea. Br. Med. J. (Oct.18): 1229-1230.
Reeves, P.G., F.H. Nielsen, G.C. Fahey, Jr. 1993. AIN-93 purified diets for laboratory rodents: Final report of the American Institute of Nutrition ad hoc writing committee on the reformulation of the AIN-76A rodent diet. J. Nutr. 123:1939-1951.
Schmidt, A., M. Anke, B. Groppel, and H. Kronemann. 1984. Effects of As deficiency on skeletal muscle, myocardium and liver: A histochemical and ultrastructural study. Exp. Pathol. 25:195-197.
Schwarz, K. 1977. Essentiality versus toxicity of metals. Pp. 3-22 in Clinical Chemistry and Chemical Toxicology of Metals, S.S. Brown, ed. Amsterdam: Elsevier.
Shen, Z.X., G.Q. Chen, J.H. Ni, X.S. Li, S.M. Xiong, Q.Y. Qiu, J. Zhu, W. Tang, G.L. Sun, K.Q. Yang, Y. Chen, L. Zhou, Z.W. Fang, Y.T. Wang, J. Ma, P. Zhang, T.D. Zhang, S.J. Chen, Z. Chen, and Z.Y. Wang. 1997. Use of arsenic trioxide (As2O3) in the treatment of acute promyelocytic leukemia (APL): II. Clinical efficacy and pharmacokinetics in relapsed patients. Blood 89:3354-3360.
Silver, A.S., and P.L. Wainman. 1952. Chronic arsenic poisoning following use of an asthma remedy. JAMA 150:584-585.
Soignet, S.L., P. Maslak, Z.G. Wang, S. Jhanwar, E. Calleja, L.J. Dardashti, D. Corso, A. DeBlasio, J. Gabrilove, D.A. Scheinberg, P.P. Pandolfi, and R.P. Warrell, Jr. 1998. Complete remission after treatment of acute promyelocytic leukemia with arsenic trioxide. N. Engl. J. Med. 339:1341-1348.
Stockman, R. 1902. The therapeutic value of arsenic and the justification of its continued use in the light of recent observations concerning its toxic action. Br. Med. J. (Oct. 18):1227-1229.
Underwood, E.J. and W. Mertz. 1987. Introduction. Pp. 1-19 in Trace Elements in Human and Animal Nutrition, Vol. 1, 5th Ed., W. Mertz, ed. San Diego, Calif.: Academic.
Uthus, E.O. 1992. Evidence for arsenic essentiality. Environ. Geochem. Health 14:55-58.
Uthus, E.O., W.E. Cornatzer, and F.H. Nielsen. 1983. Consequences of arsenic deprivation in laboratory animals. Pp. 173-189 in Arsenic: Industrial, Biomedical, Environmental Perspectives, W.H. Lederer and R.J. Fensterheim, eds. New York: Van Nostrand Reinhold.
Vahter M., and H. Norin. 1980. Metabolism of 74 As-labeled trivalent and pentavalent inorganic arsenic in mice. Environ. Res. 21:446-457.