Toxicants Occurring Naturally in Foods. (1966)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

222 A. E. HARPER TABLE | Amino Acid Requirements of Adult Man and Approxi- mate Amounts of Individual Amino Acids Obtained from 2,500 Calories of Foodstuffs of Animal, Vegetable, or Mixed Origin AMOUNT IN 2,500 CALORIES OF REQUIREMENT ROUND PEAS AND WHOLE MIXED MALE® FEMALE? STEAK® BEANS?® WHEAT® U.S. DIET® (g/DAyY) (g) (g) (g) (g) Methionine 6.1 1.7 0.9 2.5 Cystine 101 055 3.4 1.8 2.6 1.5 Tryptophan 0.25 0.16 2.7 1.4 0.9 1.3 Threonine 0.51 0.31 10.6 6.1 3.1 4.6 Leucine 1.10 0.62 19.4 12.4 5.4 7.9 Isoleucine 0.70 0.45 10.7 8.1 3.4 5.5 Valine 0.80 0.65 13.3 8.1 3.5 5.8 Phenylalanine 10.5 7.9 3.4 4.6 Tyrosine 1.400 1120 95 7.0 2.8 4.4 Lysine 0.80 0.50 20.6 9.3 2.0 7.5 Histidine — — 6.8 4.1 1.4 2.2 Protein 30-70 242 152 78 106 Dry matter — — 426 643 623 470 * Report of the Amino Acid Committee of the Food and Nutrition Board’ in which the reasons for the rather large differences between the values for males and females are discussed. > Based on proximate analysis‘ and amino acid composition.® Some information about adverse effects of amino acids obtained from studies on rats is summarized in Table 3. The amounts of individual indispensable amino acids, expressed as percent of the diet, that cause retardation of the growth of the weanling rat fed a low protein diet (containing 6 percent of casein as the protein source) are listed in column 4. The numbers in column 5 indicate relative toxicity of the different amino acids. They are based on the severity of the growth depressions observed when rats are fed equal weights of individual amino acids in a low protein diet. From four to ten times the amounts required for normal growth cause growth depressions, yet a 70 percent protein (casein) diet, which causes only a transitory retardation of growth, contains amounts of seven of the indispensable amino acids that exceed those observed to depress quite severely the growth of rats fed a low protein diet.2-!! This gives some indication of the diffi- culties encountered in trying to establish definite toxic levels for amino acids. The difficulties are further emphasized by even a cursory review of the literature on the subject.

AMINO ACIDS 223 TABLE 2 Amino Acid Requirements of 10-kg Child and Approxi- mate Amounts of Individual Amino Acids Obtained from 1,100 Calories of Foodstuffs of Animal or Vegetable Origin AMOUNT IN 1,100 CALORIES OF REQUIREMENT® FOR 10-kg ROUND PEAS AND WHOLE CHILD STEAK? BEANS? WHEAT? (g/DAyY) (g) (g) (g) eonine | 0.85 2.8 0.8 0.4 Cystine ° 1.5 0.8 1.2 Tryptophan 0.22 1.2 0.6 0.4 Threonine 0.60 4.8 2.7 1.4 Leucine 1.50 8.8 5.6 2.4 Isoleucine 0.90 4.8 3.7 1.5 Valine 0.93 6.0 3.7 1.6 Phenylalanine 0.90 4.7 3.6 1.5 Tyrosine — 3.4 3.2 1.4 Lysine 1.05 9.3 4.2 0.9 Histidine 0.32 3.1 1.9 0.6 Protein 7.7-12.8 106 67 34 Dry matter — 188 286 274 «Report of the Amino Acid Committee of the Food and Nutrition Board.® + Based on proximate analysis‘ and amino acid composition. The amounts of amino acids listed in Table 3 as causing adverse effects in the rat were established in experiments on weanling animals fed a low protein diet.* Most adverse effects of indispensable amino acids have been demonstrated under such conditions. The young animal is much more susceptible to adverse effects from an amino acid load than is the mature animal. Also, the animal fed a low protein diet is more susceptible than one fed an adequate amount of protein; and the animal fed a diet deficient in certain vitamins, especially niacin, pyri- doxine, and vitamin By is more susceptible than one receiving adequate amounts of all vitamins. Thus, tolerance levels for individual indis- pensable amino acids differ with the nutritional and physiological state of the organism. Further, the adverse effects of diets containing ex- cesses of individual amino acids usually become less severe after 1 to 2 weeks, suggesting that animals undergo various types of adaptations that increase their ability to tolerate amino acids in excess. * Much of the literature on which this discussion is based is cited in references 8—11.

224 A. E. HARPER TABLE 3 Amino Acid Requirements of the Rat, Dietary Levels of Individual Indispensable Amino Acids That Result in Adverse Effects and Amounts in a High Protein Diet DIETARY LEVEL AMOUNTS tcummener TOR ELT nN 10% IN PROTEIN (% OF DIET) ADVERSE (CASEIN) ROSE RAMA RAO EFFECTS RELATIVE DIET et al.6 et al.7 (%) Toxiciry*-> (%) Methionine 0.6 0.5 2.0 1 2.3 Cystine — — 3.0 5 0.3 Tryptophan 0.2 0.12 2.0 2 1.1 Threonine 0.6 0.5 5.0 8 3.2 Leucine 0.9 0.7 2.5 6 7.0 Isoleucine 0.5 0.55 5.0 6 4.7 Valine 0.7 0.55 5.0 6 5.2 Phenylalanine 0.9 0.9 4.0 4 4.1 Tyrosine — _ 3.0 3 4.5 Lysine 1.0 0.9 5.0 7 5.8 Histidine 0.4 0.26 2.0 3 2.1 * Based on literature cited in references 8, 9, and 10. ’ Low numbers indicate amino acids that are not well tolerated in excess. Methionine is the most toxic of the indispensable amino acids, but even the young rat fed a low protein diet will tolerate an excess of three to four times the requirement or about 2 percent of the diet as methionine. The tolerance increases as the rat matures and is greater when the diet provides an adequate level of protein. The adverse effects of excess methionine are alleviated when additional glycine is included in the diet. This probably accounts in part for the greater tolerance for excess methionine in animals with a high protein intake. In rats fed a low protein diet, adverse effects from excessive intakes of most amino acids are alleviated when the diet is supplemented with small amounts of the limiting amino acids. Evidently, a deficiency of one or more of the indispensable amino acids increases the susceptibility of the animal to an excess of the others. The reason for this is not clear but it seems probable that the adaptive phenomena mentioned above may be delayed or inhibited if one of the building blocks of proteins is in short supply.

AMINO ACIDS 225 It is evident from Table 1 that when adult man is existing on round steak (a diet that should result in close to maximal protein intake for this species), he is consuming some nine to eighteen times the amount of sulfur-containing amino acids that he requires. Nevertheless, the quantity of methionine ingested represents only about 1.5 percent of the total dry matter consumed, a figure below that leading to adverse effects in the young rat fed a low protein diet. Similar calculations for the other amino acids indicate that only leucine would be provided in such a diet in an amount (4.4 percent) that would exceed the value for the tolerance of the young rat (2.5 percent, Table 3). Again, the ad- verse effect of excess leucine is demonstrable primarily when protein intake is low. Additional isoleucine and valine alleviate the adverse effect of excess leucine, and supplements of other amino acids to- gether with isoleucine and valine enable the young rat to tolerate at least 5 percent of leucine in its diet. For the young child, who has higher amino acid requirements per unit of body weight than the adult, the degree of excess in relation to the requirement, even from a diet composed entirely of round steak, is less than that for the adult. Single oral doses of most of the indispens- able amino acids in amounts just slightly below those listed in column 2 of Table 2 have been administered to young children receiving an adequate diet without apparent ill effects (L. E. Holt, Jr., personal communication), indicating that the human body has a considerable tolerance for excesses of individual amino acids. The animal also tolerates the large quantities of amino acids in a high protein diet better than a large quantity of a single amino acid in a low protein diet. The latter results in a much greater disproportion in the dietary amino acid pattern. In all probability greater tolerance of a high protein diet is related to the metabolic response of the animal body to a high protein intake. Large excesses of individual amino acids, with the exceptions of tryptophan and tyrosine, do not cause rapid eleva- tions of the enzymes of amino acid catabolism. On the other hand, ingestion of a high protein diet results within a short time in sub- stantial increases in the activities of many of the enzymes of amino acid catabolism. A response of this type should facilitate the removal of excessive quantities of amino acids that are not needed for the synthesis of proteins and that can serve only as a source of calories. Effects of excessive amounts of tyrosine have been studied quite extensively in the rat. Animals fed a low protein diet containing just over 3 percent of L-tyrosine develop severe eye and paw lesions within

226 A. E. HARPER 1 to 2 weeks. Lesions do not develop in animals fed this amount of tyrosine with a diet that is adequate in protein presumably because the catabolic capacity of animals fed a high protein diet is greater. High intakes of threonine and certain other individual amino acids also alleviate the signs of tyrosine toxicity. Even in animals fed a low protein diet containing excess tyrosine, the lesions recede as the animal matures and usually disappear after 2 to 3 weeks. Tyrosine trans- aminase, one of the enzymes involved in tyrosine catabolism, is known to increase in activity in animals ingesting a diet that is high in tyrosine. Apparently animals fed a low protein diet also undergo adaptations that enable them to tolerate excesses of several of the amino acids, but the responses are slower than in those fed an adequate diet. Quite severe depressions have been observed in the food intake of animals fed on diets containing surpluses of individual indispensable _ amino acids and diets in which the amino acid pattern has been un- balanced by the addition of various mixtures of indispensable amino acids, particularly mixtures of all but one of the indispensable amino acids. The latter type of modification has been termed an amino acid imbalance.!2 In general, the effects of imbalances of amino acids are alleviated if the food intake of the animal can be stimulated and, as with animals fed surpluses of individual amino acids, those fed im- balanced diets appear to undergo adaptation, and the adverse effects are usually transitory. Animals receiving diets that meet their require- ments for all amino acids tolerate substantial amounts of unbalanced amino acid mixtures without showing ill effects. From the accumulated observations on animals, it seems unlikely that man would ingest excesses of individual amino acids in sufficient quantity to produce adverse effects when subsisting on a diet composed of natural foods. Quantities of individual amino acids sufficient to result in adverse effects would not be used as supplements to foodstuffs. It would require tenfold the normal supplementary level or more to approach the amounts shown to cause adverse effects in animals. Only if large doses of individual amino acids were administered regularly for some special reason would there be much likelihood of approaching toxic levels. Even then, unless the protein intake of the subject was low, it seems unlikely that quantities sufficient to produce more than mild adverse effects would be ingested. A few amino acids have been tested as possible therapeutic agents in man. Methionine has been tested in patients with liver disease,!3 as also have arginine and ornithine.'4:'5 Arginine HCI and lysine HCI have been tested as adjuvants to diuretics.!6-!7 Some others have been tested

AMINO ACIDS 227 in mental patients.!8 Although these substances have not proven to be of particular therapeutic value, the testing of them has provided some information about the effects of ingestion of large amounts of a few amino acids. Methionine has been administered in quantities of be- tween 10 and 20 g/day for 5 to 10 days; lysine HCl, 10 to 40 g/day; tyrosine and histidine, 20 g/day; tryptophan, 15 g/day; and arginine HCI at up to 40 g/day. In the main, even in patients receiving these large amounts of individual amino acids, the major response reported was mild gastric distress. Some patients with liver disease showed neurologic deterioration after treatment with methionine,!> and some mental patients showed behavioral changes.!8 There are, however, a number of genetic defects that alter the sus- ceptibility of individuals to excesses of amino acids;!9 also, when the liver is severely damaged, excesses of either protein or amino acids can cause hepatic coma.!3 Among the most extensively studied of the genetic defects are phenylketonuria, which results in intolerance of phenyl- alanine owing to a lack of one of the enzymes essential for the catabol- ism of this amino acid, and leucine hypoglycemia, in which there is intolerance of leucine either in free form or in proteins.2° Many other genetic defects of amino acid catabolism are known, and for many of these rigid control of amino acid or protein intake is an essential part of the treatment. In conclusion, animals (and presumably also man) fed a diet con- taining adequate amounts of all of the known essential nutrients have a surprising tolerance for surpluses of individual amino acids and un- balanced amino acid mixtures. Adverse effects of indispensable amino acids have been demonstrated most frequently in animals fed diets that are inadequate in some respect, particularly in protein. It 1s therefore of particular importance to ensure that amino acids used as supplements for low protein diets are those that are limiting in the diet and to ensure that supplements used under these conditions are provided only in amounts needed to maintain a satisfactory amino acid balance in the diet. REFERENCES 1. W. C. Rose, “The Amino Acid Requirements of Adult Man,” Nutr. Abstr. Rev., 27, 631 (1957). 2. E. S. Nasset, “Essential Amino Acids and Nitrogen Balance,” in Some Aspects of Amino Acid Supplementation, W. H. Cole, ed., Rutgers University Press, New Brunswick, N.J. (1956).

10. 11. 12. 13. 14. 15. 16. 17. 18. 19. A. BE. HARPER Food and Nutrition Board, National Academy of Sciences—National Research Council, Evaluation of Protein Nutrition, NAS-NRC Publication 711, NAS— NRC, Washington, D.C. (1959). H. A. Wooster, Jr., and F. C. Blanch, compilers, Nutritional Data, H. J. Heinz Co., Pittsburgh, Pa. (1949). R. J. Block and K. W. Weiss, Amino Acid Handbook: Methods and Results of Protein Analysis, Charles C homas, Springfield, Ill (1956). W. C. Rose, M. J. Oesterling, and M. Womack, “Comparative Growth on Diets Containing 10 and 19 Amino Acids, with Further Observations upon the Role of Glutamic and Aspartic Acids,” J. Biol. Chem., 176, 753 (1948). P. B. Rama Rao, V. C. Metta, H. W. Norton, and B. C. Johnson, ““The Amino Acid Composition and Nutritive Value of Proteins. III. The Total Protein and the Nonessential Amino Nitrogen Requirement,” J. Nutr., 71, 361 (1960). A. Sassen, “Revue de la toxicite des acides amines,”” Acta Gastroenterol. Belg., 18, 944 (1955). H. E. Sauberlich, ‘“‘Studies on the Toxicity and Antagonism of Amino Acids for Weanling Rats,” J. Nutr., 75, 61 (1961). A. E. Harper, “Amino Acid Toxicities and Imbalances,” in Mammalian Protein Metabolism, Vol. 2, H. N. Munro and J. B. Allison, eds., Academic Press, New York (1964), Chap. 13. A. E. Harper, “Effect of Variations in Protein Intake on Enzymes of Amino Acid Metabolism,” Can. J. Biochem., 43, 1589 (1965). A. E. Harper, P. Leung, A. Yoshida, and Q. R. Rogers, “Some New Thoughts on Amino Acid Imbalance,” Federation Proc., 23, 1087 (1964). E. A. Phear, B. Ruebner, S. Sherlock, and W. H. Summerskill, “Methionine Toxcity in Liver Disease and Its Prevention by Chlortetracycline,” Clin. Sci., 15,93 (1956). . J. S. Najarian and H. A. Harper, **A Clinical Study of the Effect of Arginine on Blood Ammonia,” Am. J. Med., 21, 832 (1956). T. B. Reynolds, A. G. Redeker, and P. Davis,“*A Controlled Study of the Effects of L-Aginine on Hepatic Encephalopathy,” Am. J. Med., 25, 359 (1958). D. A. Ogden, L. Scherr, N. Spritz, A. L. Rubin, and E. H. Luckey, **Manage- ment of Resistant Fluid Retention States with Intravenous L-Arginine Mono- hydrochloride in Combination with Mercurial Diuretics,”” Am. Heart J., 61, 16 (1961). R. P. Lasser, M. R. Schoenfeld, and C. K. Friedberg, “‘Lysine HCl: A Clinical Study of Its Action as a Chloruretic Acidifying Adjuvant to Mercurial Diuretics,” New Engl. J. Med., 263, 728 (1960). W. Pollin, P. V. Cardon, Jc., and S. S. Kety, “Effects of Amino Acid Feedings in Schizophrenic Patients Treated with Iproniazid,” Science, 133, 104 (1961). D. Y. Y. Hsia, Inborn Errors of Metabolism, Yearbook Publishers, Chicago, Il. (1960). C. C. Mabry, A. M. DiGeorge, and V. H. Auerbach, ‘‘Leucine-induced Hypo- glycemia. I. Clinical Observations and Diagnostic Considerations,” J. Pediat., 57, 526 (1960).

GEORGE K. DAVIS Toxicity of the Essential Minerals Since man is one of the end points in the food chain, he enjoys a certain degree of protection from deficiencies and toxicities that might result from variations in the levels of mineral elements in his food. Perhaps nowhere is this better exhibited than in the case of the mineral elements that we generally designate as the trace elements. The deficiencies of these elements that prevent normal development of plants and animals militate against use of these particular plants and animals by man as sources of food. Similarly, excessive amounts of these trace elements in the soil or in the feed of animals interfere with normal growth and development and result in a food selection that in a measure protects man from consuming toxic levels of the elements. The trace elements, which, as their name indicates, are required in normal nutrition in very small amounts in the diet, were in some cases recognized as toxic materials before their essentiality in nutrition was discovered. But in thinking of them as naturally occurring toxicants in foods we must examine those situations in which they may accumulate in abnormally high quantities in foods considered acceptable for human consumption. In examining copper, cobalt, iron, manganese, and zinc as dietary constituents it is well to evaluate the possibility of their interrelationships with other elements in a way that may interfere with best utilization for normal nutrition. Copper Copper requires special recognition as a potential toxicant since it is widely used in agricultural sprays with the resulting possibility of 229

230 GEORGE K. DAVIS the accumulation of copper residues in foodstuffs. Almost nothing is known, however, about the minimum level of copper required for the induction of chronic copper poisoning. Extrapolating from the results obtained with such monogastric species as rats and swine,!—3 one might assume that, to produce toxic reactions, it would require at least ten times the amount of copper normally occurring in the human diet. Such a projection suggests that levels of at least 200 to 500 ppm in the dry matter of the total diet might be necessary over considerable periods of time before chronic copper poisoning would be induced. : In practically all species, continued consumption of high levels of copper results in accumulation of copper in the tissues, and especially in the liver. Thus, although liver of sheep, swine, cattle, and poultry normally contains less than 300 ppm of copper in the dry matter, high levels of intake can result in liver contents in excess of 1,000 ppm. Such liver could be valuable as a supplement to a deficient diet, but, as the sole item of diet, such high levels of copper would need to be evaluated for possible toxic effects.2:45 The limited use of copper compounds in the preservation of food and the close regulation of copper compounds in agricultural use as sprays that might produce residues in products consumed by man have essen- tially eliminated the possibility of food products containing in excess of 50 ppm in the dry matter. It is known that soils that are high in copper content, either as a result of copper fertilization or as the result of the application of sprays containing copper, can result in some increase in the copper content of the plant materials growing in them. Heavy fertilization of soils with copper may result in a transient increase of the copper content of the plants to values of approximately 50 ppm in the dry matter. However, the copper content of food plants rarely has been found to exceed 25 ppm in the dry matter, and plant metabolism usually serves to regulate the copper level at between 10 and 15 ppm. Experimental work with animals suggests that chronic copper poisoning may require several weeks to develop to the point at which symptoms can be seen. Removal of excess copper from the diet results in rapid elimination of abnormal copper accumulation in the tissues. Iron The efficiency with which absorption of iron by man is normally regulated suggests that iron levels that could cause a toxic reaction probably are never derived from natural foodstuffs. There is always the

MINERALS 231 possibility that certain pathological conditions may occur in which tissue accumulation of iron might be augmented by the intake of food- stuffs containing high levels of this element. Since human diets are not low in phosphate if cereal grains or animal products are consumed, the interference of iron with phosphorus metabolism, which may occur in animals with low phosphorus intakes, is not a likely occurrence in man. In ruminant animals, diets low in copper may result in abnormal accumulations of iron that is not utilized for the formation of hemo- globin,® but similar observations in the human, or in monogastric species generally, have not been reported. The levels of iron that occur in dietary items vary widely, and iron deficiency, not toxicity, is the most likely dietary situation involving iron that may affect man. Values in foods of 16 ppm and over have been classified as excellent.’ Liver may contain several thousand parts per million,’ and certain legumes may contain considerably more than 100 ppm.?.10 Manganese The well-known antiseptic properties of such compounds as potassium permanganate, and consequently their toxic action, merit consideration of manganese as a naturally occurring toxic material. Manganese is an essential trace element, but a deficiency of manganese in man has not been observed as a result of a dietary lack. Manganese toxicity that has occurred in man has usually stemmed from the inhalation of manganese ore dust rather than as the result of abnormal dietary intakes. Studies of the interrelationship between manganese and iron have indicated that high levels of manganese intake, of the order of 1,000 to 5,000 ppm in the diet, have resulted in reduced accumulation and utilization of iron.!!-13 Such high levels would be extremely difficult to attain in human foods. Cereals and their products, which are considered rich in manganese, contain only slightly more than 100 ppm on a dry matter basis, and other foodstuffs range very much lower, mostly below 30 ppm.!4.15 Cobalt Since 1929, it has been recognized that cobalt can produce a poly- cythemia in animals when fed or when injected parenterally.!6 This effect of cobalt has been confirmed many times since, and these obser- vations justify inclusion of cobalt as a naturally occurring toxic material.

232 GEORGE K. DAVIS However, it should be immediately pointed out that the amount of cobalt present in foods is extremely small, and the element appears to be utilized nutritionally only as a component of vitamin Biz. The wide variation that occurs in cobalt levels in foodstuffs is illustrated by analyses showing values as low as 0.01 ppm in corn compared with values of approximately 1.0 ppm in some leguminous plants.!7 Since even the highest values that occur in foods are less than 1 percent of the levels required to produce a polycythemia, the possibility that the cobalt occurring naturally in foods might cause toxic reactions may be considered nonexistent.!8 Zinc Most experimental animals have shown a relatively high tolerance to zinc in the diet. Nevertheless, zinc poisoning has been reported in animals and in man, and under experimental conditions high levels ot zinc in the diet will interfere with copper nutrition and utilization, which in turn interferes with iron metabolism.!9 Zinc is an essential element. Its accumulation in tissues of the body varies widely, ranging from less than 20 ppm in such organs as the brain, lung, and adrenal gland to several thousand parts per million in some parts of the eye.2° The natural occurrence of zinc in foods and in feeds is only about | percent of the several thousand parts per million required to produce toxicity. Cereal grains contain on the average between 30 and 40 ppm of zinc, and protein concentrates contain from less than 20 to more than 100 ppm.2! Some seafoods have been reported to be quite rich in zinc, with shellfish yielding some of the highest values—in excess of 400 ppm. Calcium The two mineral nutrients, calcium and magnesium, occur in natural foods and feedstuffs at widely varying levels. The well-recognized need for calcium for the formation of osseous tissues has led to the addition of sources of calcium, such as calcium carbonate, to the diets of animals, and both prophylactic and therapeutic supplements of calctum compounds are frequently recommended for man. The extremely high intakes of calcium that may occur in some indi- viduals have led to a consideration of the effect both of such high levels of the calcium compounds, per se, and of their potential influence on the utilization of other nutrients.

MINERALS 233 There is considerable evidence that man and other animals readily adjust to varying levels of calcium intake with consequent variation in the efficiency of calcium absorption and utilization.43—28 Since there is rapid adjustment to variations in level of calcium intake, adverse effects of excessive intakes of calcium appear to be restricted to the influence of calcium upon the utilization of other nutrients. There are reports in the literature that high levels of calcium in the diet, due to the addition of calcium compounds as supplements to the diet, may interfere with the utilization of phosphorus if the ratio of calcium to phosphorus is wide. This apparent antagonistic effect of calcium is overcome by adequate vitamin D in the diet.29 High levels of calcium intake by animals have been associated with interference with zinc utilization, and either a reduction in the intake of calcium or an in- crease in the intake of zinc has been necessary to prevent the develop- ment in swine of parakeratosis, which is a result of zinc deprivation. A parallel situation has not been observed in man.30—32 Although there has not been demonstrated as clear-cut or as intense a relationship with other elements as with zinc, calcium at high levels in the diet interferes with utilization of magnesium, iron, iodine, and manganese. This interaction may be nutritionally important if the latter elements are at marginal levels in the diet. Slight increases in the level of these elements with a consequent narrowing of the ratio with cal- cium has overcome the reported interference.33 It should be pointed out that the calcium occurring naturally in foodstuffs is unlikely greatly to exceed 1 percent except in the case of supplemental foods such as bonemeal. The exceptionally high intakes of calcium that have caused a need for increasing intakes of trace elements will occur because of the addition of supplemental calcium compounds to the diet. Magnesium Although the pattern of adjustment appears to be somewhat different from that which occurs with calcium, there is nonetheless a rapid adjustment to high levels of magnesium in the diet.34 There is very little evidence of any adverse effect from any levels of magnesium in human nutrition, but attention of investigators has been primarily directed to the possibility of a magnesium deficiency rather than an excess. There is some evidence that very high levels of mag- nesium may increase the excretion of phosphorus and calcium although the significance of this effect is uncertain.35

234 GEORGE K. DAVIS Although magnesium compounds are frequently used therapeutically for man, a situation comparable to that in animals, where dolomite may be used as a replacement for limestone as a supplement in the diet, has not been reported. Since magnesium levels in the naturally occurring foodstuffs are quite low, it 1s difficult to conceive of a diet of natural foodstuffs that would contain harmful levels of magnesium. REFERENCES 1. R. Boyden, V. R. Potter, and C. A. Elvehjem, “Effect of Feeding High Levels of Copper to Albino Rats,” J. Nutr., 15, 397 (1938). 2. J.T. McCall, “The Interaction of Copper and Other Dietary Factors in Animal Metabolism,” Doctoral Dissertation, University of Florida (1958). 3. H. D. Wallace, J. T. McCall, B. Bass, and G. E. Combs, “High Level Copper for Growing-Finishing Swine,” J. Animal Sci., 19, 1153 (1960). 4. L.B. Bull, “The Occurrence of Chronic Copper Poisoning in Grazing Sheep in Australia,” Special Conference in Agricultural, Proceedings (Australia), 1949, Publ. 1971, HMSO, London (1951), pp. 300-310. 5. E. M. Hall and E. M. MacKay “Experimental Hepatic Pigmentation and Cirrhosis. 1. Does Copper Poisoning Produce Pigmentation and Cirrhosis of the Liver,’ Am. J. Pathol. , 7, 327 (1931). 6. H.R. Marston, ‘‘Cobalt, Copper and Molybdenum in the Nutrition of Animals and Plants,”’ Physiol. Rev., 32, 66 (1952). 7. H. K. Stiebeling, ““The Iron Content of Vegetables and Fruits,” U.S. Dept. Agr. Circ. No. 205, 205 (1932). 8. A. Shoden, B. W. Gabrio, and C. A. Finch, “The Relationship between Ferritin and Hemosiderin in Rabbits and Man,” J. Biol. Chem., 204, 823 (1953). 9. K. C. Beeson, “The Mineral Composition of Crops with Particular Reference to the Soils in which They Were Grown,” U.S. Dept. Agr. Misc. Publ. No. 369 (1941). 10. NAS-NRC Committee on Animal Nutrition and the National Committee on Animal Nutrition, Canada, Joint United States-Canadian Tables of Feed Com- position, National Academy of Sciences—National Research Council Publ. 659, NAS-NRC, Washington, D.C. (1959). 11. R. H. Hartman, G. Matrone, and G. H. Wise, “Effect of High Dietary Manganese on Hemoglobin Formation,” J. Nutr., 57, 429 (1955). 12. G. Matrone, R. H. Hartman, and A. J. Clawson, “Studies of a Manganese-Iron Antagonism in the Nutrition of Rabbits and Baby Pigs,” J. Nutr., 67, 309 (1959). 13. N. W. Robinson, S. L. Hansard, D. M. Johns, and G. L. Robertson, “‘Excess Dietary Manganese and Feed Lot Performance of Beef Cattle,” J. Animal Sci., 19, 1290 (1960). 14. W. H. Peterson and J. T. Skinner, ‘“‘Distribution of Manganese in Foods,” J. Nutr., 4, 419 (1931). 15. P. J. Schaible, S. L. Bandemer, and J. A. Davidson, ‘“‘Manganese Content of Feedstuffs and Its Relation to Poultry Nutrition,” Mich. State Univ. Agr. Expt. Sta. Tech. Bull., 159 (1938).

MINERALS 235 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. K. Waltner and K. Waltner, ““Cobalt and Blood,” Klin. Wochschr., 8, 313 (1929). C. Hurwitz and K. C. Beeson, “Cobalt Content of Some Food Plants,” Food Res., 9, 348 (1944). W. C. Grant and W. S. Root, “Fundamental Stimulus for Erythropoiesis,” Physiol. Rev., 32, 449 (1952). G. W. Monier-Willams, Trace Elements in Food, Chapman and Hall, London (1949). E. J. Underwood, Trace Elements in Human and Animal Nutrition, Academic Press, New York (1962). G. K. Davis, R. L. Shirley, and J. F. Easley, unpublished data. J. B. Lackey, personal communication. S. N. Gershoff, M. A. Legg, and D. M. Hegsted, “Adaptation to Different Calcium Intakes in Dogs,” J. Nutr., 64, 303 (1958). H. Z. Hollinger and C. J. Pattee, ““A Review of Abnormal Calcium and Phos- phorus Metabolism,”’ Can. Med. Assoc. J., 74, 912 (1956). B. B. Migicovsky and J. W. S. Jamieson, “Calcium Absorption and Vitamin D,”’ Can. J. Biochem. Biophys., 33, 202 (1955). H. H. Mitchell, ““The Need for Calcium is Flexible,”” Mod. Med., 23, 85 (1955). R. Nicolaysen, N. Eeg-Larsen, and O. J. Malm,““Physiology of Calcium Metab- olism,” Physiol. Rev., 33, 424 (1953). G. K. Davis, Transfer of Calcium and Strontium Across Biological Membranes, Academic Press, New York (1963), p. 129. A. T. Schohl and S. Farber, “Effect of A.T. 10 (Dihydrotachysterol) on Rickets in Rats Produced by High-Calcium Low-Phosphorus Diets,” J. Nutr., 21, 147 (1941). H. F. Tucker and W. D. Salmon, “‘Parakeratosis or Zinc-Deficiency Disease in the Pig,” Proc. Soc. Exptl. Biol. Med., 88, 613 (1955). R. M. Forbes, “‘Excretory Patterns and Bone Deposition of Zinc, Calcium and Magnesium in the Rat as Influenced by Zinc Deficiency, EDTA and Lactose,” J. Nutr., 74, 194(1961). G. K. Davis, Transfer of Calcium and Strontium Across Biological Membranes, Academic Press, New York (1963), pp. 130-132. Ibid., pp. 129-142. L. A. Graham, J. J. Caesar, and A. S. V. Burgen,““Gastrointestinal Absorption and Excretion of Mg 23 in Man,” Metab. Clin. Exptl., 9, 646 (1960). Y. Kamitani, ‘“‘Magnesium Metabolism. 1. Magnesium Metabolism by Adult Men with the Usual Diet,” Eiyo To Shokuryo, 9, 50, 113 (1956-57); cited in Chem. Abstr. 51, 9823f (1957).

JAMES F. MEAD* and ROSLYN B. ALFIN-SLATER Toxic Substances Present in Food Fats For the purpose of this review, the toxic substances occurring in food fats are divided into categories as follows: (1) toxic substances that are integral components of the fat (substances present in the natural fat, artifacts of processing) and (2) toxic substances soluble in the fat but not essentially lipids themselves (substances present in the natural fat, artifacts of processing). No attempt is made to present an exhaustive survey of all possible toxic substances in these categories. Rather, certain important examples are used to illustrate the classes of com- pounds other examples of which may suggest themselves because of chemical relationships. INTEGRAL COMPONENTS OF FATS Fluorine-Containing Fatty Acids One of the most interesting examples is that of the w-fluoro fatty acids occurring in the seeds of Dichapetalum toxicarium, a glabrous shrub from Africa known also as “rat bane” or “broke back.” Efforts by Peters and his collaborators!-3 resulted in isolation, from the seeds, of a fluorinated unsaturated fatty acid, w-fluoro-cis-9-octadecenoic acid accompanied by small amounts of an w-fluoro-saturated acid. The structure of the major component has been confirmed by synthesis. * Supported in part by Public Health Service Research Career Award No. GM- K6-19, 177 from the Division of General Medical Sciences, National Institutes of Health. 236

FOOD FATS 237 The toxicity of this type of fatty acid stems from its metabolic con- version into 8-fluoroacetic acid and to fluorocitrate which then inhibits the reactions of the tricarboxylic acid cycle, resulting in the accumula- tion of citric acid in the tissues. Although this substance is not normally present in the human diet, it has been responsible for considerable damage to livestock. Higher Wax Esters When laboratory animals are fed diets containing 15 percent of wax esters such as cetyl oleate or sperm oil, they develop a severe seborrhea and die within a month. This effect is shown only by those esters con- taining about 33 or more carbon atoms.’ A possible explanation ap- pears in a report by Hansen and Mead,‘ who found that the oily fur of rats on a diet containing 15 percent of oleyl palmitate derived from undigested wax ester eliminated in the feces. The purgative effect of the oleyl palmitate, however, was not solely responsible for the early death of animals reported by Kaneda and Sakurai5 and others and may have simply contributed to the effect of a low-protein diet. Saturated Fat Recently, Tove’ reported a toxic effect in both weanling and adult mice after feeding high levels (20 to 40 percent) of glycerylmonopalmitate and glycerylmonostearate for a period of 3 weeks. Alleviation of the symptoms, i.e., poor growth and high mortality, with oleic acid and glycerylmonooleate as well as with safflower oil or the fatty acids from safflower oil rule out the possibility of an enhanced essential fatty acid deficiency as the sole reason for the toxic symptoms. Digestibility studies showed that the monoglycerides were digested and that the calories were being used although no data on food consumption were presented. Tove suggests that an acute metabolic defect was produced possibly involving uncoupling of oxidative phosphorylation, which has been observed also in essential fatty acid deficiency.’ It is also suggested that the saturated fat toxicity reported here may be related to the lipogranuloma observed by Herting ef al.9 in rats fed diets rich in saturated fats. These results agree with those reported by Coleman et a/.!° in feeding studies using the acetoglyceride, acetostearin. Acetoglycerides are modified fats in which acetyl groups are substituted for one or two of the long-chain fatty acids of the triglycerides of natural fats and oils.

238 JAMES F. MEAD AND ROSLYN B. ALFIN°SLATER Methods for the preparation of acetoglycerides involve the inter- esterification of the triglyceride with triacetin or the acetylation of the monoglyceride using acetic anhydride.'! The introduction of an acetyl group into the glyceride containing stearic acid results in a reduction in melting point and a change in crystal structure from a hard brittle form to a waxy, plastic, flexible material.!2 Acetostearins can act as barriers to water vapor and atmospheric gases,!3 and these properties have sug- gested their use as coatings for processed meats, cheese, nuts, and confections, and in spreads.!2 However, in studies on rats, growth was markedly retarded and survival was poor in animals fed acetostearin at relatively high levels (30 percent). Both growth and survival were con- siderably improved in animals supplemented with linoleate, and markedly improved when cottonseed oil was used as a supplement. Although the digestibility of acetostearin when ingested as the sole dietary fat was poor (53 percent digestibility in weanling animals 5 weeks on diet decreasing to 35 percent digestibility after 12 weeks), the digestibility was improved by the addition of either linoleate or cotton- seed oil to the diet. The authors suggest that both essential fatty acid deficiency and caloric restriction are responsible for the poor per- formance of rats fed acetostearins, since food consumption studies revealed a decreased food intake, but it is possible that some metabolic derangement may also be involved. Although it may seem strange that so common a foodstuff as satu- rated fats should be listed under toxic materials, it should be recognized that the toxic symptoms occur only with relatively large percentages and in the absence of unsaturated fats. These conditions are not likely to be met in the usual human diet. Peroxidized and Polymeric Products It has been known for a considerable time that unsaturated lipids may become toxic with long exposure to air. Barnes et al/.!4 found that part of this effect was due to oxidative destruction of micronutrients in the food. However, it became apparent that even when vitamin supple- ments were given separately, extensively oxidized fats were toxic. Kaneda and his co-workers!5 and Matsuo!® demonstrated that the highly unsaturated fish oils, though not toxic when freshly prepared, quickly become so upon exposure to air. This toxicity was shown to be largely the result of peroxide accumulation, since it could be pre- vented by chemical reducing agents. Some idea of the concentration of peroxide necessary for the toxic effect was obtained by Andrews et al.,!7 who found that when oxidized vegetable oil was fed as 15 percent of the

FOOD FATS 239 diet, a peroxide number (PN) of 100 (milliequivalents of peroxide oxygen per kilogram of oil) appeared to have no adverse effect on growth rate and appearance of rats raised from weaning on the diet. A PN of 400, however, resulted in decreased growth and a PN of 1,200 or above was quickly fatal. It was ascertained in these studies that the effect must take place largely in the gut since peroxides are destroyed during absorption. As unsaturated oils are heated at high temperatures either in the presence or absence of air, any peroxides formed (in the former case) are destroyed and a new type of toxicity appears. This has been shown by Crampton ef al.!819 to be due to the accumulation of cyclized products of the polyunsaturated acids, which appear to be very toxic in fairly low concentrations. The question arises as to whether either of these products can be formed in harmful amounts in dietary fats outside the laboratory. Most experiments designed to answer this question have definitely shown that they are not,2° although related changes may produce polymeric substances that are not absorbed and consequently decrease the nutritive value of the fat. Nevertheless, caution should certainly be exercised in the use of oils containing large amounts of polyunsaturated acids to provide adequate antioxidants for prevention of autoxidation, as has been demonstrated by Horwitt and his co-workers.?! 22 A recent example has been reported by Witting,“ who found that the creatinuria produced in rats by a low-tocopherol high-polyunsaturated-fat diet could be partially prevented by selenium. Although the adverse effects of insufficient antioxidant have generally been demonstrated by chick encephalomalacia in the laboratory, at least one report has indicated that a cooking oil used for human consumption has shown toxicity for rats. SUBSTANCES SOLUBLE IN FOOD FATS Many of the toxic substances discussed in this volume are fat soluble and occur in the natural fat of foods. Of particular and current interest are the mycotoxins that are discussed in the chapter on fungal toxins (page 126) and in the chapter on tumorigens and carcinogens (page 24). Hydropericardium-Producing Factors The death of millions of broilers fed on certain types of chicken feed compounded with the use of specific lots of feed-grade animal fats

240 JAMES F. MEAD AND ROSLYN B. ALFIN-SLATER brought about an investigation of this “‘toxic fat” factor in several laboratories. It was found in certain batches of oleic acid and triolein used in feed manufacture and resulted, in chickens, in hydropericardium, edema, ascites, liver and kidney damage, and testicular degeneration. As little as 5 ug of the toxic factors result in death for the chick, and these factors have also been shown to be toxic for rats and monkeys.4—27 Investigation of the nature of the substances showed them to be present in the nonsaponifiable fraction and to consist of two chlorinated hydrocarbons with formulas C;sHioCls. The probable structure has been shown by Wootton and Courchene*8 to be hexachlorohexahydro- phenanthrene. Since compounds of this type do not occur naturally in fats it seems certain that they were introduced during the production or processing, but the actual source is still unknown. REFERENCES 1. R.A. Peters, R. W. Wakelin, A. J. P. Martin, J. Webb, and F. T. Birks, ““Obser- vations upon the Toxic Principle in the Seeds of Dichapetalum toxicarium. Separation of a Long-Chain Fatty Acid Containing Fluorine,” Biochem. J., 71, 245 (1959). 2. R. A. Peters, “Fluorine Compounds in African Plants,” Biochem. J., 76, 32p (1960). 3. R. A. Peters, R. J. Hall, P. F. V. Ward, and N. Sheppard, “The Chemical Nature of the Toxic Compounds Containing Fluorine in the Seeds of Dichapetalum toxicarium,” Biochem. J., 77, 17 (1960). 4. R. E. A. Dear and F. L. M. Pattison, “Toxic Fluorine Compounds. XVIII. The Synthesis of the Toxic Principle of Dichapetalum toxicarium (18-Fluoro- cis-9-octadecenoic Acid) and Related w-Fluoro Unsaturated Acids,” J. Am. Chem. Soc., 85, 622 (1963). 5. T. Kaneda and H. Sakurai, “Studies on the Nutritive Values of Lipids. X. The Formation of Seborrhea in Albino Rats Fed the Esters of Fatty Acids with Higher Alcohols,” Bull. Japan Soc. Sci. Fisheries, 19, 1168 (1953-54). 6. I. A. Hansen and J. F. Mead, “The Fate of Dietary Wax Esters in the Rat,” Proc. Soc. Exptl. Biol. Med., 120, 527 (1965). 7. S. B. Tove, “Toxicity of Saturated Fat,” J. Nutr., 85, 237 (1964). 8. P. D. Klein and R. M. Johnson, “‘Phosphorus Metabolism in Unsaturated Fatty Acid-Deficient Rats,” J. Biol. Chem., 211, 103 (1954). 9. D. C. Herting, P. L. Harris, and R. C. Crain, “Effect of Saturated and Un- saturated Fatty Acids on Dietary Lipogranuloma,” J. Nutr., 70, 247 (1960). 10. R. D. Coleman, L. A. Gayle, and R. B. Alfin-Slater,“‘A Nutritional Evaluation of Acetostearins in Rats,” J. Am. Oil Chemists’ Soc., 40, 737 (1963). 11. F. J. Baur, “Acetin Fats. II. Preparation and Properties of Diacetin Fats from Some Common Vegetable Oils,” J. Am. Oil Chemists’ Soc., 31, 196 (1954).

FOOD FATS 241 12. 13. 14. 15. 16. 17. 18. 19. 21. 22. 23. 24. 26. 27. R. B. Alfin-Slater, R. D. Coleman, R. O. Feuge, and A. M. Altschul, “The Present Status of Acetoglycerides,”’ J. Am. Oil Chemists’ Soc., 35, 122 (1958). N. V. Lovegren and R. O. Feuge, “Permeability of Acetostearin Products to Carbon Dioxide, Oxygen and Nitrogen,” J. Agr. Food Chem., 4, 634 (1956). R. H. Barnes, W. O. Lundberg, H. T. Hanson, and G. O. Burr,“‘Effect of Certain Dietary Ingredients on the Keeping Quality of Body Fat,” J. Biol. Chem., 149, 313 (1943). T. Kaneda and S. Ishii, “‘Studies on the Nutritive Value of Lipids,” Bull. Japan Soc. Sci. Fisheries, 19, 171 (1953-54); T. Kaneda, H. Sakurai, and S. Ishii, ibid., 20, 50, 658 (1954—55). N. Matsuo, ‘“‘Nutritional Effects of Oxidized and Thermally Polymerized Fish Oils,” in Symposium on Foods: Lipids and Their Oxidation, H. W. Schultz, E. A. Day and R. O. Sinnhuber, eds., AVI Publishing Co., Westport, Conn. (1962), Chap. 17. J. S. Andrews, W. H. Griffith, J. F. Mead, and R. A. Stein, “Toxicity of Air- Oxidized Soybean Oil,” J. Nutr., 70, 199 (1960). E. W. Crampton, R. H. Common, F. A. Farmer, F. M. Berryhill, and L. Wise- blatt, ““Studies to Determine the Nature of the Damage to the Nutritive Value of Some Vegetable Oils from Heat Polymerization. I. The Relation of Autoxida- tion to Decrease in Nutritional Value of Heated Linseed Oil,” J. Nutr., 43, 533 (1951); 44, 177 (1951). E. W. Crampton, R. H. Common, F. A. Farmer, A. F. Wells, and D. Crawford, “Studies to Determine the Nature of the Damage to the Nutritive Value of Some Vegetable Oils from Heat Polymerization. III. The Segregation of Toxic and Nontoxic Material from Esters of Heat-Polymerized Linseed Oil by Distillation and by Urea,” J. Nutr., 49, 333 (1953). . J. B. Brown, “Changes in Nutritive Value of Food Fats during Processing and Cooking,” Nutr. Rev., 17, 321 (1959). M. K. Horwitt, Nutr. Rev., 18, 351 (1960). M. K. Horwitt, C. C. Harvey, B. Century, and L. A. Witting, “Polyunsaturated Lipids and Tocopherol Requirements,” J. Am. Dietet. Assoc., 38, 231 (1961). J. A. Witting, “Lipid Peroxidation In Vivo,” J. Am. Oil Chemists’ Soc., 42, 908 (1965). N. V. Raju and R. Rajagopalan, “Nutritive Value of Heated Vegetable Oils,” Nature, 176, 513 (1955). R. Allcroft, R. B. A. Carnaghan, K. Sargeant, and J. O’Kelly,““The Examination of Fats and Fatty Acids for Toxic Substances,” Ver. Record, 73, 428 (1961). H. Delongh and H. Beerthuis, ’Response of Chickens to Prolonged Feeding of Crude Toxic Fat,” Biochim. Biophys. Acta, 65, 548 (1962). W. D. Salmon and P. M. Newberne, “Isolation of Three Hydropericardium- Producing Factors from a Toxic Fat,”” Cancer Res., 23, 571 (1963). J. C. Wootton and W. L. Courchene, “‘Structure of Two Hydropericardium- Producing Factors from a Toxic Fat,” J. Agr. Food Chem., 12, 94 (1964).

EDWARD EAGLE Gossypol Cottonseed, a waste product of cotton production for several thousand years, has been a source of cottonseed oil and cottonseed meal in the United States since the latter part of the nineteenth century. In 1861, Kuhlmann! published a report on a blue pigment that he had recovered from “‘cottonseed foots.” The latter material, also known as soapstock, is a by-product of alkali refining of edible oil and consists of soaps, pigments, and other minor constituents originally present in the crude oil. It was 20 years later that the English chemist Longmore? reported the extraction of a pigment from the same source, and in 1899 the Polish chemist Marchlewski° isolated and purified a polyphenolic material from cottonseed foots which he named gossypol, a contraction of the words gossypium and phenol. These investigators were interested in this yellow substance as a dye for silk and wool and made no mention of physiological activity. In 1938, Adams and co-workers* proposed a chemical structure for gossypol showing its tautomeric forms, and in 1958 Edwards? reported the total synthesis of gossypol and authenti- cated the structure formulated by Adams and his group 20 years earlier. The Chemical Abstracts designation for gossypol is 1,1’,6,6’,7,7'- hexahydroxy-5,5’-diisopropyl-3,3’-dimethyl(2,2'-binaphthalene]-8 ,8’- dicarboxaldehyde. Its chemistry has been reviewed in detail by Boatner® and more recently by Adams et al.’ The earliest recorded statement concerning a harmful effect of cotton- seed is attributed to Voelker® in England in 1859. In the intervening years, many strange materials have been blamed for the adverse findings in animals after cottonseed feeding.9 Between 1915 and 1918, Withers and Carruth published a series of papers!®-!3 whose titles referred to 242