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.

GOSSYPOL 243 gossypol as the toxic substance in cottonseed. These and the subsequent publications by Schwartze and Alsberg,'4-!7 who claimed a positive correlation between the toxicity of cottonseed and gossypol content, led to the general belief that the toxicity of cottonseed is due solely to gossypol. Gossypol and other pigments of cottonseed are concentrated in pig- ment glands (sometimes called resin glands), distinct morphological entities, spherical or ovoid in shape and 100 to 400 uv in diameter, that appear as dark spots in a cross section of the seed. Although these cottonseed pigment glands were first studied in 1887!8 and have been described in detail in succeeding studies,6-!9-22 and later with the aid of the electron microscope,” their function is still unknown. The pigment glands constitute 1 to 3 percent of the weight of the raw cottonseed kernel and contain 25 to 50 percent gossypol. With the availability of both pure gossypol24> and cottonseed pig- ment glands separated in an essentially unaltered condition by a flota- tion process,2627 it became possible to make toxicologic evaluations (acute oral LDso determinations) on samples of known gossypol con- tent. In 1947, Boatner?® made a preliminary report of a collaborative study in which it was found from LDspo studies that samples of unaltered cottonseed pigment glands containing 33.7 to 40 percent gossypol were more toxic than pure gossypol itself. The full report by Eagle et al.29 concluded that the toxicity of cottonseed pigment glands is attributable to some component of the glands other than, or in addition to, gossypol and gossypurpurin. In spite of two dozen papers listed in the review by Adams et al.’ that point to gossypolas the sole toxic agent, Eagle and co-workers produced further evidence from LDs0 studies on 30 samples of untreated, stored, and treated cottonseed pigment glands?°-33 that the toxicity cannot be accounted for solely on the basis of their gossypol content and ques- tioned the common practice of considering gossypol analyses as a yardstick for toxicity. The acute oral LDso values for the rat for un- treated pigment glands varied from 925 to 2,170 mg/kg and bore no relation to their gossypol content which varied from 27 to 37.8 percent. The LDso values for samples of pure gossypol varied from 2,400 to 3,340 mg/kg. Additional evidence that gossypol is not the sine gua non of the toxicity of cottonseed has been obtained from experiments in which levels of gossypol or pigment glands were incorporated into the diets of experimental animals.32 34.35 The literature contains reports of a large number of feeding trials with various types of processed cottonseed meals on many animal

244 EDWARD EAGLE species that have yielded extensive physiologic information despite their limitations as sources of precise toxicologic data. But several in- vestigators had noted that a poor correlation existed between the gossypol content and the nutritional results obtained when cottonseed meals had been fed to rats and to chicks.*’ It is possible to determine toxicologic variations between different samples of cottonseed meal by feeding them at sufficiently high levels in an otherwise balanced diet to exaggerate any toxic effects while the protein content of these diets is confined toa narrow range. After feeding more than 100 different cotton- seed meals at the exaggerated level of 67 percent in the diets of rats? 8 no correlation was found between growth performance and the gossypol content of these meals, which ranged from 0.01 to 1.29 percent. Indeed, many meals with fairly high gossypol content were not only nontoxic to the rat but yielded body weight gains that in some cases were superior to those of control rats fed stock or soybean meal diets. These results reconfirm that the toxicity of cottonseed products cannot be accounted for solely on the basis of analyzed gossypol content. In 1963, Lyman and co-workers?? reported the oral LDso value in the rat for a sample of cottonseed pigment glands to be 1,120 mg/kg, while that for gossypol was 2,570 mg/kg, and concluded that the toxicity of pigment glands cannot be accounted for entirely on the basis of their gossypol content. It has been shown that prolonged storage of pure gossypol and cottonseed pigment glands at 2 to 10°C for from 4 to more than 9 years caused little change in their free gossypol content or in their oral LDso0 values for the rat.33 Marchlewski had noted in 18993 that gossypol is quickly oxidized in a solution of sodium hydroxide, but it was not until 1912 that Withers and Ray“? showed by rabbit feeding tests that detoxification of cotton- seed can be accomplished by boiling with alcoholic sodium hydroxide. Since then there have been numerous reports of varying degrees of successful detoxification of cottonseed by cooking, steaming, and autoclaving (at least 26 references cited in the reviews by Eagle*! and by Adams et al.’); by treatment with alkali ;9-1! 13 42.43 with iron salts ;!! 44-5! with aniline and other amines;!252-54 by combination with pro- tein ;!2.48.55.56 and by use of combinations of steam and various chemi- cal agents (cited in reference 7). It has been emphasized that successful detoxification of cottonseed meal by overcooking is in reality a failure if the favorable effects from decreased toxicity are offset by the un- favorable effect from protein damage in the process.*8 Some adverse effects attributed to gossypol have included depression of appetite and body weight,22:31.32.34.35.37.57-60 djarrhea,29-31 35.57.58

GOSSYPOL. 245 hemolysis,6! hemolytic anemia,62 and hypoprothrombinemia.@ An unfavorable effect of gossypol on hatchability and weight of eggs has been reported by Heywang ef al.,64.6 and twelve reports of discolora- tion within several months after storage of eggs from hens fed cotton- seed meal have been listed.” Fatalities from gossypol have been noted by many authors. !!.17.29-33 46.57-59.61.66-68 The death of the experimental animals after oral administration of pure gossypol prompted a warning note in Science against its suggested use as an appetite depressant in man.57 In 1919, Alsberg and Schwartze™ stated that cardiac irregularity is the most prominent effect of gossypol poisoning with death generally caused by circulatory failure. In 1956, Eagle and Brewer‘? studied the effects of intraperitoneally injected pure gossypol and cottonseed pigment glands on the blood pressure, respiration, and electrocardio- gram of anesthetized dogs connected to an ink-writing oscillograph- electroencephalograph. Three different samples of pure gossypol injected at a level of 150 mg/kg produced no effects in a total of eight dogs on whom continuous tracings were taken over periods lasting from 4 to 5 hours. On the other hand, 13 different samples of cottonseed pigment glands administered in amounts sufficient to provide a gossypol level of 150 mg/kg caused death in 23 of the 29 dogs tested within 80 to 278 minutes, in most cases from circulatory failure. Ten of the dogs developed pulsus alternans and three others showed less striking evidence of heart block. Postmortem findings after gossypol intoxication in various experimen- tal animals have been described by many authors.!! 17.29.46 57.58 .62.67.70—73 Some of the more important and consistent findings were pulmonary edema and congestion, hydrothorax, cardiac dilatation, ascites, in- flammation and hemorrhage in the gastrointestinal tract, and hepatic and renal congestion. The biological value of cottonseed protein is high, and cottonseed meal is one of the most important concentrated protein supplements for ruminants, who are generally less sensitive than nonruminants to the toxic constituents in raw cottonseed. In 1964, the production of cotton- seed meal and cake in the United States was approximately 2.6 million short tons, and a potential source of about 40,000 short tons of gossypol was available if important uses for this unusual chemical existed. Commercially processed cottonseed meals undergo marked reduction in gossypol content and it has been commonly accepted that cottonseed meals that have 0.04 percent or less free gossypol can be fed in unre- stricted proportion in balanced diets even to chicks, broilers, and

246 EDWARD EAGLE swine.22 But the processing procedures or treatments that cause these great reductions in gossypol content could hardly be so selective as to have no effect on any other factor(s) present.38 By selective breeding, geneticists have developed an almost glandless cottonseed, essentially free of gossypol. Large-scale production of such gland-free seed could be a reality within the next decade. Cottonseed flour has been used in various types of human food in different parts of the world. In 1964, the U.S. Food and Drug Administration reissued a color additive order clearing the use of partially defatted cooked cottonseed flour or similar products derived from cottonseed intended for human consumption with the specification that the free gossypol content shall not exceed 0.045 percent by weight.74 REFERENCES 1, F. Kuhlmann, “Memoirs sur une nouvelle couleur bleue préparée avec I’huile de coton,” Compt. Rend., 53, 444 (1861). 2. J. Longmore, “Cottonseed Oil; Its Coloring Matter and Mucilage and Descrip- tion of a New Method of Recovering the Loss Occurring in the Refining Pro- cess,”’ J. Soc. Chem. Ind., 5, 200 (1886). 3. L. Marchlewski, “‘Gossypcl, ein Bestandtheil der Baumwollsamen,” J. Prakt. Chem., 60, 84 (1899). 4. R. Adams, R. C. Morris, T. A. Geissman, D. J. Butterbaugh, and E. C. Kirk- patrick, “Structure of Gossypol. XV. An Interpretation of Its Reactions,” J. Am. Chem. Soc., 60, 2193 (1938). 5. J. D. Edwards, Jr., “Total Synthesis of Gossypol,” J. Am. Chem. Soc., 80, 3798 (1958). 6. C. M. Boatner, in Cottonseed and Cottonseed Products, A. E. Bailey, ed., Interscience, New York (1948). | 7. R. Adams, T. A. Geissman, and J. D. Edwards,“*Gossypol, a Pigment of Cotton- seed,” Chem. Rev., 60, 555 (1960). 8. A. Voelker, cited (p. 521) by A. C. Crawford, ‘*A Poisonous Principle in Certgin Cottonseed Meals,” J. Pharmacol. Exptl. Therap., 1, 519 (1910). 9. E. Eagle, -H. F. Bialek, D. L. Davies, and J. W. Bremer, “Detoxification of Cottonseed by Salts and Alkalies,” J. Am. Oil Chemists’ Soc., 31, 121 (1954). 10. W. A. Withers and F. E. Carruth, ‘“‘“Gossypol, a Toxic Substance in Cottonseed,” Science, 41, 324 (1915). 11. W. A. Withers and F. E. Carruth, “‘Gossypol, the Toxic Substance of Cotton- seed Meal,” J. Agr. Res., 5, 261 (1915). 12. W. A. Withers and F. E. Carruth, ‘“‘“Gossypol, the Toxic Substance in Cotton- seed,” J. Agr. Res., 12, 83 (1918). 13. F. E. Carruth, “Contribution to the Chemistry of Gossypol, the Toxic Principle of Cottonseed,” J. Am. Chem. Soc., 40, 647 (1918). 14. E. W. Schwartze and C. L. Alsberg, “Quantitative Variation of Gossypol and Its Relation to the Oil Content of Cottonseed,” J. Agr. Res., 25, 285 (1923).

GOSSYPOL 247 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 29. 30. 31. 32. 33. E. W. Schwartze and C. L. Alsberg, ‘‘Relation between Toxicity of Cottonseed and Its Gossypol Content,” J. Agr. Res., 28, 173 (1924). E. W. Schwartze, ‘““Gossypol and Cottonseed Meal Poisoning,” J. Oil Fat Ind., 3, 173 (1926). E. W. Schwartze and C. L. Alsberg, ‘Pharmacology of Gossypol,” J. Agr. Res., 28, 191 (1924). F. von Bretfeld, ‘‘Anatomie des Baumwollen und Kapoksamens,”’ J. Landw., 35, 29 (1887). E. E. Stanford and A. Viehoever, ‘‘Chemistry and Histology of the Glands of the Cotton Plant, with Notes on the Occurrence of Similar Glands in Related Plants,” J. Agr. Res., 13, 419 (1918). C. H. Boatner, C. M. Hall, M. L. Rollins, and L. E. Castillon, “‘Pigment Glands of Cottonseed. II. Nature and Properties of Gland Walls,’ Botan. Gaz., 108, 484 (1947). C. H. Boatner, C. M. Hall, R. T. O’Connor, and L. E. Castillon, ‘‘Pigment Glands of Cottonseed. III. Distribution and Some Properties of Cottonseed Pigments,” Botan. Gaz., 109, 108 (1947). A. M. Altschul, C. M. Lyman, and F. H. Thurber, in Processed Plant Protein Foodstuffs, A. M. Altschul, ed., Academic Press, New York (1958). A. T. Moore and M. L. Rollins, “‘New Information on the Morphology of the Gossypol Pigment Gland of Cottonseed,” J. Am. Oil Chemists’ Soc., 38, 156 (1961). C. H. Boatner, R. T. O’Connor, M. C. Curet, and C. S. Samuels, ““The Pigments of Cottonseed. III. Gossyfulvin, a Native Cottonseed Pigment Related to Gossypol,” J. Am. Chem. Soc., 69, 1268 (1947). L. E. Castillon, C. M. Hall, and C. H. Boatner, “Preparation of Gossypol from Cottonseed Pigment Glands,” J. Am. Oil Chemists’ Soc., 25, 233 (1948). C. H. Boatner and C. M. Hall, ‘“‘The Pigment Glands of Cottonseed. I. Behavior of the Glands toward Organic Solvents,”’ Oil Soap, 23, 123 (1946). H. L. E. Vix, J. J. Spadsro, R. D. Westbrook, A. J. Crovetto, E. F. Pollard, and E. A. Gastrock, “‘Pre-pilot-plant Mixed-solvent Flotation Process for Separating Pigment Glands from Cottonseed Meals,” J. Am. Oil Chemists’ Soc., 24, 288 (1947). . C. H. Boatner, ‘“‘Nutritional Investigations with Gossypol, Cottonseed Pigment Glands and Gland-free Cottonseed Flour,”’ Abstracts of papers, 38th ann. mtg., Am. Oil Chemists’ Soc., New Orleans, La., May 20-22, 1947. E. Eagle, L. E. Castillon, C. M. Hall, and C. H. Boatner, ‘“‘Acute Oral Toxicity of Gossypol, and Cottonseed Pigment Glands for Rats, Mice, Rabbits and Guinea Pigs,” Arch. Biochem., 18, 271 (1948). E. Eagle, C. M. Hall, L. E. Castillon, and C. B. Miller, “Effect of Fractionation and Treatment on the Acute Oral Toxicity and on the Gossypol and Gossy- purpurin Content of Cottonseed Pigment Glands,” J. Am. Oil Chemists’ Soc., 27, 300 (1950). E. Eagle and H. F. Bialek, “Effect of Single and Repeated Doses of Gossypol on the Rat,” Food Res., 15, 232 (1950). E. Eagle and H. F. Bialek, “‘Toxicity and Body Weight Depressing Effects in the Rat of Water-Soluble Combination Products of Gossypol, Gossypol and Cottonseed Pigment Glands,” Food Res., 17, 543 (1952). E. Eagle and D. L. Davies, “‘Effect of Long-Term Storage on Acute Oral Toxicity and Gossypol Content of Cottonseed Pigment,” J. Am. Oil Chemists’ Soc., 35, 36 (1958).

248 34. 35. 36. 37. 38. 39. 41. 42. 43. 45. 47. 49. 31. 32. 53. 54. 35. EDWARD EAGLE E. Eagle and D. L. Davies, ‘‘Feed Value and Protein Quality Determinations on Cottonseed Meals,” J. Am. Oil Chemists’ Soc., 34, 454 (1957). A. M. Ambrose and D. J. Robbins, ‘‘Studies on the Chronic Oral Toxicity of Cottonseed Meal and Cottonseed Pigment Glands,” J. Nutr., 43, 357 (1951). W. D. Gallup, ““The Relation of p-Gossypol to the Toxicity of Some Cottonseed Products,” Ind. Eng. Chem., 20, 59 (1928). C.H. Boatner, A. M. Altschul,G. W. Irving, Jr., E. F. Pollard, and H. C. Schaefer, “The Nutritive Value of Cottonseed for Chicks as Affected by Methods of Processing and the Content of Pigment Glands,” Poultry Sci., 27, 315 (1948). E. Eagle, H. F. Bialek, D. L. Davies, and J. W. Bremer, “Biological vs. Chemical Evalution of Toxicity and Protein Quality of Cottonseed Meals,” J. Am. Oil Chemists’ Soc., 33, 15 (1956). A. S. El-Nockrashy, C. M. Lyman, and J. W. Dollahite, ““‘The Acute Oral Toxicity of Cottonseed Pigment Glands and Intraglandular Pigments,” J. Am. Oil Chemists’ Soc., 40, 14(1963). W. A. Withers and B. J. Ray, ““A Method for the Removal of the Toxic Prop- erties from Cottonseed Meal,” Science, 36, 31 (1912). E. Eagle, “‘A Review of Some Physiological Effects of Gossypol,” J. Am. Oil Chemists’ Soc., 37, 40 (1960). E. Eagle and F. A. Norris, ‘Detoxification of Solvent-Extracted Cottonseed Meal,” U.S. Patent 2,740,718 (Apri! 3, 1956). W. H. King, L. T. Wolford, F. H. Thurber, A. M. Altschul, A. B. Watts, C. W. Pope, and J. Conley, “Effect of pH During the Cooking of Cottonseed on the Properties of Meals and Oil,” J. Am. Oil Chemists’ Soc., 33, 71 (1956). W. A. Withers and J. F. Brewster, ‘Studies on Cottonseed Meal Toxicity. II. Iron as an Antidote,” J. Biol. Chem., 15, 161 (1913). W. A. Withers and F. E. Carruth, “Iron as an Antidote to Cottonseed Meal Injury,” J. Biol. Chem., 32, 245 (1917). J. P. McGowan and A. Crichton, “Cottonseed Meal Poisoning,” Biochem. J., 18, 273 (1924). W. D. Gallup, “The Value of Iron Salts in Counteracting the Toxic Effects of Gossypol,” J. Biol. Chem., 77, 437 (1928). W. L. Robison, “Cottonseed Meal for Pigs,” Ohio Agr. Exptl. Sta. Res. Bull., 534 (1934). A. D. Swenson, E. A. Fieger, and C. W. Upp, “The Nature of Egg Yolk Dis- coloration Produced by Cottonseed Meal,” Poultry Sci., 21, 374 (1942). E. Eagle, ‘‘Detoxification of Cottonseed Pigment Glands with Ferrous Sulfate,” Proc. Soc. Exptl. Biol. Med., 72, 444 (1949). J. L. Fletcher, B. F. Barrentine, L. J. Dreesen, J. E. Hill, and C. B. Shawver, “The Use of Ferrous Sulfate to Inactivate Gossypol in Diets of Laying Hens,” Poultry Sci., 32, 740 (1953). F. E. Carruth, ‘‘Methods for Approximating the Relative Toxicity of Cotton- seed Products,” J. Biol. Chem., 32, 87 (1917). E. P. Clark, ‘Studies on Gossypol. II. Concerning the Nature of Carruth’s p-Gossypol,” J. Biol. Chem., 76, 229 (1928). J. V. Rice, ““Feed Composition,” U.S. Patent 2,607,687 (Aug. 19, 1952). W. D. Gallup and R. Reder, “The Influence of Certain Dietary Constituents on the Response of Rats to Gossypol Ingestion,” J. Agr. Res., 51, 259 (1935).

GOSSYPOL 249 6 82 g 63. R R R S 74. E. Eagle and J. W. Bremer, “Detoxification of Cottonseed Material,” U.S. Patent 2,797,997 (July 2, 1957). E. Eagle, “Chronic Toxicity of Gossypol,” Science, 109, 361 (1949). E. Eagle, “Effect of Repeated Doses of Gossypol on the Dog,” Arch. Biochem., 26, 68 (1950). R. J. Lillie and H. R. Bird, “Effect of Oral Administration of Pure Gossyp l and of Pigment Glands of Cottonseed on Mortality and Growth of Chicks,” Poultry Sci., 29, 390 (1950). J. R. Couch, W. Y. Chang, and C. M. Lyman, “Effect of Free Gossypol on Chick Growth,” Poultry Sci., 34, 178 (1955). P. Menaul, “The Physiological Effect of Gossypol,” J. Agr. Res., 26, 233 (1923). R. H. Rigdon, G. Crass, T. M. Ferguson, and J. R. Couch,“Effects o Gossypol in Young Chickens with Production of a Ceroid-like Pigment,” A.M.A. Arch. Pathol., 65, 229 (1958). W. S. Harms and K. T. Holley, “Hypoprothrombinemia Induced by Gossypol,” Proc. Soc. Exptl. Biol. Med., 77, 297 (1951). B. W. Heywang, H. R. Bird, and A. M. Altschul, “The Effect of Pure Gossypol on Egg Hatchability and Weight,” Poultry Sci., 29, 916 (1950). B. W. Heywang and H. R. Bird, “Egg Production, Diet Consumption and Live Weight in Relation to Free Gossypol in the Diet,” Poultry Sci., 33,851 (1954). C. L. Alsberg and E. W. Schwartze, “Pharmacological Action of Gossypol,” J. Pharmaco!.(Proc.), 13, 304(1919). I. G. Macy and N. M. Alter, “Further Observations on the Results of Feeding Cottonseed Meal and Kernels,” Am. J. Physiol., 55, 304 (1921). I. G. Macy and J. P. Outhouse, “Further Studies on Cottonseed Meal Injury,” Am. J. Physiol., 69, 78 (1924). E. Eagle and N. R. Brewer, unpublished observations. E. P. Clark, “Sondies on Gossypol. I. Preparation, Purification and Some of the Properties of Gossypol, the Toxic Principle of Cottonseed,” J. Biol. Chem., 75. 725 (1927). W. D. Gallup, “Sundies on the Toxicity of Gossypol. L The Response of Rats to Gossypul Administration during Avitaminosis,” J. Biol. Chem., 93, 381 (1931). H. A. Smith, “The Pathology of Gossypol Poisoning,” Am. J. Pathol., 33, 353 (1957). R. H. Rigdon, T. M. Ferguson, V. S. Mohan, and J. R. Couch, “In Vivo Produc- tion of a Cerosd-like Pigment in Chickens Given Gossypol,” A.M.A. Arch. Pathol., 67, 94 (1959). Federal Register, 29, S304 (July 21, 1964).

D. W. FASSETT Nitrates and Nitrites The occurrence of inorganic nitrates in animal forage and occasionally in some water supplies has, at times, given rise to serious acute toxic effects, especially those resulting from methemoglobin formation. The accidental use of excessive amounts of nitrate—nitrite mixtures in the preservation of the color of meat products has also been associated with this type of injury. Many investigations have been concerned with the factors influencing the nitrate content of plants and the mechanism of toxic action, especially in ruminants. Some authors have considered the possibility that, under some conditions, the intake of sufficient quantities of vegetables of high nitrate content by human beings might result in methemoglobin formation secondary to reduction of the nitrate to nitrite. As early as 1895, an investigation by Mayo!2 of acute deaths in a herd of cattle in Kansas established clearly that the cause was due to the unusually large quantities of potassium nitrate present in the corn stalks used as fodder. The corn was grown near a barn on land with a high organic nitrogen content and contained as much as 25 percent dry weight of potassium nitrate. The toxic manifestations of tremors, diuresis, collapse, and later of cyanosis were reproduced by dosing cattle with potassium nitrate at about 1.3 g/kg. Although methemo- globin formation due to nitrites was not clearly understood at the time, particular mention was made of the very dark color of the blood, and chemical studies showed evidence for conversion of some of the nitrate to nitrite in the blood and especially in the bile. This remarkable study thus established the now well-recognized essential elements in the acute toxicity of plant nitrates to livestock, 250

NITRATES AND NITRITES 251 namely, the large increase in nitrate content of some plants with high nitrogen fertilization and the reduction of nitrate to nitrite in vivo by bacterial action in the rumen.'3 Although most work on the problem has been done with cattle, the same type of acute effects have been seen in sheep,!® although they appear less sensitive than cattle. Turkeys!7 also show nitrite toxicity after ingestion of high nitrate feeds. In addition to acute effects, there has been some concern in recent years as to the possible relationship of high nitrate animal diets to vitamin A deficiency, conversion of carotene to vitamin A, and possible effects on the thyroid.4 Experimentally, some interference by nitrates or nitrites with carotene utilization or vitamin A shortage has been found with calves,!3 rats,4:15.20 lambs,§ and chicks.!9 When comparisons were made, nitrites were more potent in these respects than nitrates. Thyroid weight was increased in chicks!9 and rats.24 Since several thousand parts per million of nitrites and much more of nitrates were used in these experiments, the practical nutritional significance of the findings, although suggestive, does not seem to be clear at present. The hazard of excessive nitrates in water supplies has been generally recognized since Comly2 described cyanosis in two infants resulting from the use of well waters containing 64 to 140 ppm of nitrate nitro- gen. These waters contained only traces of nitrite ion (0.4 to 1.3 mg/ liter). The total intake of nitrate ion was of the order of 0.2 g/day. The possible role of coliform contamination and digestive disturbances was recognized. Many outbreaks have occurred and the subject has been reviewed by Burden,! Steyn,?! and Walton.?3 It is generally agreed that nitrates in water supplies are especially hazardous to young infants because of their relatively higher gastric pH, which facilitates the reduction to nitrite by bacteria. In addition, an infant has a relatively high fluid intake compared with an adult. Water containing in excess of 10 ppm nitrate nitrogen (equals 45 mg of nitrate ion/liter) is con- sidered hazardous for use in infant feeding.22 Livestock have been affected by nitrate nitrogen in water at levels of 70-150 ppm. Sodium nitrate and nitrite are permitted in a limited number of meat and fish products at levels of 500 and 200 ppm, respectively. Bacterial reduction of nitrate in these products causes the formation of nitric oxide hemoglobin or myoglobin and results in a comparatively stable red or pink color. Outbreaks of poisoning from the accidental incorpo- ration of excessive amounts of nitrate—nitrite mixtures in meat products have occurred.!6 Two of the ten cases described by Orgeron ef al.!6 were in children 2 and 3 years old and followed the eating of wieners containing over 5,000 ppm of nitrite.

252 D. W. FASSETT Lehman!! does not consider the usual 200 ppm residual nitrite in meats such as corned beef hazardous for adults, but he stresses the lower margin of safety in children because of their lower total hemo- globin content. In addition, the molar ratio between nitrite and hemo- globin was noted to be such that | g of sodium nitrite could theoretically produce as much as 460 to 1,800 g of methemoglobin. He advises that the use of nitrates be carefully controlled and that extension of use to other products seems undesirable. Lehman also mentions some unpub- lished studies that showed sodium nitrate to be tolerated in the diet of rats for a lifetime at levels of 1 percent with no effects and only growth depression at 10 percent. Dogs tolerated 2 percent for 3-4 months. Sodium nitrite was tolerated by rats for 121 days at 93 ppm and by a cat for 105 days at 390 ppm. Druckrey eft al. recently considered the possibility that nitrites in foods might cause teratogenic or carcinogenic effects. They point out the possibility of formation of nitrosamines resulting from formation of the nitrosyl cation, NO+, in the highly acid conditions in the stomach. Long-term studies with rats at levels as high as 100 mg/kg/day of sodium nitrite did not give evidence for a nitrosamine type of carcino- genic effect nor was there evidence of teratogenic action. Some shorten- ing of life span and reduction in hemoglobin concentration were noted. They conclude that the present use is safe but that it might be desirable to reduce it because of the narrow safety margin. The need for better chemical methods of detecting nitrosamines in foods was mentioned. The possibility of methemoglobin formation in human beings from ingestion of vegetables containing high levels of nitrate has been con- sidered in a number of reports. Wilson,26 while studying the nitrate content of plants in relation to nitrate fertilization, noted the increased nitrate content of a number of leafy vegetables such as beets, broccoli, cabbage, cauliflower, and lettuce that results from high levels of fertilization. He suggested the possibility that some reduction to nitrite in the digestive tract might occur, although the quantities usually eaten would make it unlikely that nitrate from this source alone would prove hazardous. In a later paper,2 added information was given on the nitrate content of certain vegetables and baby foods. The content varied considerably but was found to be from about 1,000 to 5,000 ppm for such vegetables as spinach, broccoli, rhubarb, cabbage, and lettuce, but very low for some vegetables such as tomatoes. Strained baby foods contained lower concentrations; for example, two samples of spinach contained only 600-800 ppm. Concern was expressed that

NITRATES AND NITRITES 253 the increasing nitrate content associated with high nitrate fertilization might be hazardous to babies, especially if nitrate intake in some waters was considered. Gilbert et al.5 reported the high nitrate content of some vegetables but felt it was unlikely that any acute effects would result from normal intakes. Neither Gilbert nor Wilson were aware of any reports of actual cases in human beings. Kilgore et al.9 considered the possibility that the ascorbic acid present in such foods might act as a protective agent against nitrate toxicity. No evidence could be found for this in guinea pigs, however. An earlier study by Kilgore and her associates® demonstrated that rats and rabbits could develop some methemoglobin (about 20 percent) if given large amounts of inorganic nitrate in the diet (3-10 g/kg/day). Feeding lower doses of nitrate ion or collard greens containing 0.6 percent NO3— (dry weight basis) produced slight amounts of methemoglobin. A nitrate balance study showed only small amounts of nitrite in urine and some of the nitrate remained un- accounted for. In a study using human subjects, Kiibler!° fed seven infants a diet in which one fifth of the food intake was replaced by spinach containing 68 mg of NO3—/100 g of prepared vegetable. The daily intake of NO3— ion was about 16-21 mg/kg/day. Feedings continued for 1 week, followed by a week of low NO3~— intake. Urines were studied for NO3— content, creatinine, chloride, and blood nitrite (NO2-), and methemo- globin contents were determined. The urinary nitrate increased rapidly with high NO3— intake and subsided quickly after cessation of intake. No diuresis was seen. No nitrite or methemoglobin could be found in blood. Kiibler felt that the nitrate ion is so rapidly absorbed in the upper gut that normally no bacterial reduction to nitrite occurs. He pointed out that this might not be the case if a gastrointestinal upset was present. He concluded that there was little evidence of a risk to infants from nitrate in vegetables and that the nitrate ion was rapidly excreted without reduction to nitrite. Recently, Holscher and Natzschka’ reported two cases of methemo- globinemia in infants (55 percent methemoglobin in one case) after eating spinach puree. The spinach contained large amounts of nitrite (about 218 mg of nitrite ion/100 g wet weight) and only small amounts of nitrate. It was assumed that for some unknown reason the normal nitrate content had been reduced to nitrite during storage. Also of importance in this situation is the sensitivity of fetal and infant hemo- globin to the action of methemoglobin-forming agents such as nitrites.!4

254 D. W. FASSETT It may be concluded that, in contrast to ruminants, the risk of either acute or chronic injury to human beings from the naturally occurring nitrate content in vegetables is very low indeed. It is difficult to con- ceive of any situation in which adults might be affected, and it would seem to require unusual circumstances for methemoglobin to be pro- duced in infants from this source. Under normal conditions, the nitrate ion is very rapidly absorbed and excreted as such with little opportunity for reduction to nitrite. Some conditions under which a high vegetable nitrate intake might impose some added risk to infants are the simultaneous ingestion of high nitrate water or meat, intestinal disturbances, and anemias. While past experience indicates that such situations must have been rare, it is also clear that the unusual sensi- tivity of infants to methemoglobin-forming agents requires continued study and control of the total nitrate-nitrite content of their food and water. The possibility of accidental reduction of the relatively harmless nitrate ion during storage or processing of vegetables obviously needs further investigation as suggested by the cases described above.’ Continued attention needs to be given to vegetable nitrate content in relation to fertilization and growing conditions. The nature of any “protective” agents in vegetables as suggested by Kilgore and the existence of any significant thyroid or vitamin A effects are areas for further investigation. Additional information on the metabolism of nitrates and nitrites in human beings, especially infants, seems desirable. NOTE: In a recent letter to the editor [Lancet, i, 872 (1966)] Claus Simon of the Pediatric Department of the University of Kiel reports additional cases of nitrite toxicity from spinach. He believes that, if spinach with a high nitrate content from excessive fertilization is allowed to stand at room temperature after preparation, bacterial reduction may occur. This may even occur in frozen spinach. It probably does not occur with canned spinach. He suggests that prepared spinach should not be kept at room temperature, excessive fertilization should be avoided, the nitrate content of spinach used for infant feeding should not be more than 300 ppm, and spinach should not be given in the first 3 months of life. A useful summary of current conclusions regarding the effect of nitrate fertilization on nitrate concentrations in plants is provided in Agriculture Information Bulletin No. 299 of the Agriculture Research Service, U.S. Department of Agriculture [The Effect of Soils and Fertilizers on the Nutritional Quality of Plants, U.S. Govt. Printing Office, Washington, D.C. (Oct. 1965)].

NITRATES AND NITRITES 255 REFERENCES 1. 2. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. E. H. W. J. Burden, ‘“‘The Toxicology of Nitrates and Nitrites with Particular Reference to the Potability of Water Supplies,”’ Analyst, 86, 429 (1961). H. H. Comly, ‘‘Cyanosis in Infants Caused by Nitrates in Well Water,” J. Am. Med. Assoc., 129, 112 (1945). H. Druckrey, D. Steinhoff, H. Beuthner, H. Schneider, and P. Klarner, “Prufung von Nitrit auf chronisch toxische Wirkung an Ratten,” Arzneimittel-Forsch., 13, 320 (1963). R. J. Emerick and O. E. Olson, “‘Effect of Nitrate and Nitrite on Vitamin A Storage in the Rat,” J. Nutr., 78, 73 (1962). . C.S. Gilbert, H. E. Eppson, W. B. Bradley, and O. A. Beath, “Nitrate Accumula- tion in Cultivated Plants and Weeds,” Wyoming Agr. Exp. Sta. Bull. No. 277, Laramie, Wyoming (1946). . E. E. Hatfield, G. S. Smith, A. L. Neumann, R. M. Forbes, U. S. Garrigus, and O. B. Ross, “‘Interactions of Nitrite, alpha-Tocopherol and 1-Methy]-2-mer- captoimidazole upon the Vitamin A Nutrition of Lambs Fed ‘High Nitrate’ Silage,” Proc. Soc. Animal Prod., Western Sec., 12, 41 (1961). P. M. Holscher and J. Natzschka, ‘“‘Methamoglobinamie bei jungen Sduglingen durch nitrithaltigen Spinat,’’ Deut. Med. Wochschr., 89, 1751 (1964). L. Kilgore, L. Almon, and M. Gieger,‘‘The Effects of Dietary Nitrate on Rabbits and Rats,” J. Nutr., 69, 39 (1959). L. Kilgore, A. R. Stasch, and B. F. Barrentine, “Relation of Ascorbic Acid to Nitrate Content of Turnip Greens and to Methemoglobin Formation,” Am. J. Clin. Nutr., 14, 52 (1964). W. Kiibler, ““Die Bedeutung des Nitratgehaltes von Gemiise in der Ernaéhrung des Sauglings,” Z. Kinderheilk., 81, 405 (1958). A. J. Lehman, ‘“‘Nitrates and Nitrites in Meat Products,” Assoc. Food Drug Officials U.S. Quart. Bull., 22, 136 (1958). N. S. Mayo, “Cattle Poisoning by Potassium Nitrate,” Kansas Agr. Expt. Sta. Bull. 49, Manhattan, Kansas (1895). P. K. Mcllwain and I. A. Schipper, “Toxicity of Nitrate Nitrogen to Cattle,” J. Am. Vet. Med. Assoc., 142, 502 (1963). W. K. Metcalf, “The Sensitivity of Intracorpuscular Haemoglobin to Oxidation by Nitrite Ions. I. The Effect of Growth, Starvation and Diet,” Phys. Med. Biol., 6, 427 (1962). B. L. O’Dell, Z. Erek, L. Flynn, G. B. Garner, and M. E. Muhrer, “Effects of Nitrite Containing Rations in Producing Vitamin A and Vitamin E Deficiencies in Rats,” J. Animal Sci., 19, 1280 (1960). ° J. D. Orgeron, J. D. Martin, C. T. Caraway, R. M. Martine, and G. H. Hauser, *‘Methemoglobinemia from Eating Meat with High Nitrite Content,” Public Health Rept. (U.S.), 72, 189 (1957). C. W. Riggs, “Nitrite Poisoning from Ingestion of Plants High in Nitrate,” Am. J. Vet. Res., 6, 194 (1945). K. B. Sinclair and D. I. H. Jones, “‘Nitrate Toxicity in Sheep,” J. Sci. Food Agr., 15, 717 (1964). J. L. Sell and W. K. Roberts, “‘Effects of Dietary Nitrite on the Chick: Growth, Liver Vitamin A Stores and Thyroid Weight,” J. Nutr., 79, 171 (1963).

256 D. W. FASSETT 20. G. S. Smith, A. L. Neumann, and E. E. Hatfield, “‘Carotene Utilization and Vitamin A Nutrition as Influenced by Dietary Nitrite and ‘High Nitrate’ Silage: Laboratory Studies,” Proc. Soc. Animal Prod., Western Sec., 12, 42 (1961). 21. D. G. Steyn, The Problem of Methaemoglobinaemia in Man with Special Refer- ence to Poisoning with Nitrates and Nitrites in Infants and Children, Publ. No. 11, University of Pretoria, Pretoria, Union of South Africa (1960). 22. U.S. Public Health Service Drinking Water Standards (1962), p. 47. 23. G. Walton, “Survey of Literature Relating to Infant Methemoglobinemia Due to Nitrate-Contaminated Water,” Am. J. Pub. Health, 41,986 (1951). 24. C. Welsch, R. A. Bloomfield, G. B. Garner, and M. E. Muhrer, “Effect of Nitrate in The Diet on Weight of Thyroid and Adrenal Gland,” J. Animal Sci., 20, 981 (1961). 25. J. K. Wilson, “‘Nitrate in Foods and Its Relation to Health,” Agron. J., 41, 20 (1949). 26. J. K. Wilson, “‘Nitrate in Plants: Its Relation to Fertilizer Injury, Changes During Silage Making, and Indirect Toxicity to Animals,” J. Am. Soc. Agron.., 35, 279 (1943).

D. W. FASSETT Oxalates For many years case reports have appeared in textbooks and in the literature of acute poisoning from ingestion of oxalate-containing plants particularly those in the Polygonaceae family and particularly the species rhubarb (Rheum rhaponticum L.) and sorrel grass (Rumex acetosa L.). Because of the fact that these plants are known to contain a somewhat higher oxalate content than most other foods, the cause of the symptoms has almost invariably been attributed to this factor. A number of textbooks on poisoning or poisonous plants discuss the role of oxalic acid.3—5.17.18.28.31 Symptoms are said to be similar to those noted in human beings ingesting oxalic acid, and the authors stress the occurrence of corrosive effects in the mouth or intestinal tract, gastric hemorrhage, renal colic or hematuria, and sometimes convulsions. Other authors have expressed doubt as to the role of oxalate content in rhubarb poisoning. For example, Sollmann?4 doubts if the oxalate content of rhubarb leaves is responsible for the poisoning. Locket!3 states that fatal poisoning by rhubarb leaves is probably mythical and that a person would need to eat some 4 kg of rhubarb to get the lowest recorded fatal dose of oxalic acid. Drill6 mentions that the oxalate content of vegetables such as rhubarb and spinach has no “significance unless an unlikely degree of indulgence. . . is assumed.” He points out that this may not be the case in cattle eating very large quantities of vegetable matter. Although there is no question that the ingestion of sufficient (about 5 g or more) oxalic acid as crystals or in solution by human beings can be fatal with associated corrosive gastroenteritis, shock, convulsive symptoms, and renal damage,® a careful examination of some of the 257

258 D. W. FASSETT alleged cases of oxalate poisoning from rhubarb allows some question as to the etiology. A case that has been repeatedly quoted is that re- ported by Robb.22 This report, published in 1919, is a brief letter to the editor of the Journal of the American Medical Association. A housewife had prepared rhubarb for supper by frying some of the leaves for greens and by boiling the stalks. The husband ate the stalks, but only a small quantity of the leaves. Some 12 hours later the housewife developed some cramp-like abdominal pains, and by about 50 hours after inges- tion of the rhubarb she developed symptoms of shock, vomited a brownish fluid, and aborted a 6 weeks old fetus. The woman died some hours later, and it was mentioned that there was some bleeding from the nose after death. The husband appeared to have few symptoms except for feeling weak and dizzy the day following ingestion of the rhubarb. There was no postmortem examination nor was there any analysis of the ingested plant. No mention was made of oxalate crystals in the urine or of any disturbance of kidney function. No description was given of corrosive effects in the mouth. A more recent case is that reported by Tallqvist and Vaananen2’ of the death of a child from oxalic acid poisoning caused by eating rhubarb leaves. In this instance, a 5-year-old girl, along with other children in an orphanage, was said to have eaten some raw rhubarb stalks and in addition had been fed some rhubarb leaves by her friends. There was no evidence as to the quantity of leaves or stalks ingested. No immediate symptoms were reported, but later in the same day the child became drowsy and would not eat. She eventually developed vomiting of a dark material, was treated with sedatives, and developed coma and some reduction in urinary output. No oxalate crystals in the urine were mentioned, but it was stated that a qualitative oxalic acid test was strongly positive. No postmortem examination was done. Death, however, was definitely attributed to oxalic acid in the rhubarb leaves and stalks. A second case was mentioned of a 4-year-old girl with abdominal pain and vomiting the day after eating raw rhubarb. Vomiting persisted for some 5 days, following which the patient was hospitalized. It was established that uremia and anuria were present. An increase in potas- sium in the blood was noted, but the blood calcium was not determined. At autopsy this patient showed the typical picture of a lower nephron nephrosis, but no oxalate crystals were found. The authors consider the etiology of this case as not being clearly established. Certainly neither of these cases appeared to fit the typical picture of a corrosive gastroenteritis that is known to follow ingestion of

OXALATES 259 oxalic acid. Furthermore, it is rather doubtful that oxalic acid exists in the free form in plants.32 The oxalates in vegetables such as spinach and rhubarb exists principally as the calcium or potassium salts. An extensive review by Jeghers and Murphy!® on oxalate metabolism indicates that the oxalate content of spinach is actually not very different from that in rhubarb. Spinach, for example, is stated by various authors to contain from 0.2 to 0.9 percent total oxalate whereas rhubarb stalks contain 0.24 to 1.3 percent oxalate. The review by Watt ez al.3! states that the rhubarb stalks contain 0.4 to 1 percent oxalate and the leaf 0.3 to 1.1 percent oxalate. Other high oxalate-containing foods are almonds, 0.4 percent; cashew nuts, 0.3 percent; cocoa, 0.4 percent; and tea, 0.2 percent. It is thus apparent that the difference in oxalate content between rhubarb leaves or stalks and other common foods, such as spinach, is scarcely sufficient to establish rhubarb as a poison. Further- more, the form in which the oxalate is present seems unlikely to be capable of causing a corrosive tissue change. There seems to be a scarcity of experimental work on this subject. Tanner and Tanner,?8 however, quote a report by Maue!® that he and five other individuals repeatedly ate cooked rhubarb leaves without effect and that these leaves contained 0.4 percent oxalate. Maue be- lieved oxalates were not responsible for previously reported illnesses. Although it is possible that, as stated by Locket,!3 some of the reports of poisoning from rhubarb leaves are coincidental, there does seem to be a certain amount of evidence that rather severe symptoms have been produced at times from this source. It is somewhat remarkable that toxicologists have not generally considered the role that might be played by various toxic anthraquinone derivatives present in these species.32 In a recent monograph,’ there is considerable discussion of the chemistry of the anthraquinone type of purgative, and it has been pointed out that these compounds may exist in both the roots and stems of species such as rhubarb or sorrel grass. It has also been found that the glycoside is probably the active form of the compound and that the redyced form of the quinone may be oxidized on storage. It therefore seems possible that these substances may have been involved in some of the cases. A recent report by Streicher makes this suggestion. In this instance a 6-year-old girl and her 4-year-old brother ate the stems and leaves of raw rhubarb in quantities varying from 20 to 100 g. Both developed profuse vomiting within 2 hours and eventually slight icterus and enlargement of the liver. The girl developed a renal insufficiency and a 4+ albumin in the urine. It is significant that no oxalates were present

260 D. W. FASSETT in the urine in this case. Recovery followed extracorporeal hemodialysis. Streicher suggests that these cases could scarcely have been attributed to oxalates since the ingestion of the latter would only have been of the order of 0.2 to 0.8 g. He points out that 200 g of rhubarb stalk, 300 g of red beets, 150 g of celery, and 200 g of spinach would have furnished the same amount of oxalate as that eaten by the girl. It was also stated in this paper that the anthraquinone glycosides are present in the leaves especially in the early summer and that in their fresh state they are extremely irritating. The content of anthraquinones could be as high as 0.5 to 1 percent of the weight of the fresh leaf. A study with human volunteers by Schmid? showed that 10-20 g of fresh rhubarb leaves caused immediate vomiting. Streicher proposes that the highly irritating reduced forms of the anthraquinone glycosides are likely to be absorbed readily and are probably responsible for toxic effects in the liver and kidney. It is obvious from this review of the literature that, in spite of the many reports of poisoning attributed to oxalic acids in plants, those cases described in any detail bear little, if any, resemblance to those known to have occurred from the chemical oxalic acid. There seems to be little evidence tor a corrosive action on the mouth, esophagus, or stomach. The onset of symptoms seems to be delayed and there is no apparent uniformity regarding the matter of oxaluria. A review of inquiries at the National Clearinghouse for Poison Control Centers! shows eighteen reported episodes of eating of rhubarb leaves in the years 1959 through 1964 with no fatalities and only five instances of nonspecific symptoms such as vomiting, diarrhea, or abdominal cramps. It is possible that oxalates may play a role in some cattle poisoning, but it appears much more likely that toxic anthraquinone glycosides may have been involved in human cases. Seasonal variations in glycoside content and alterations during cooking would seem to be important variables in the toxic actions. Additional toxicologic work is needed to clarify the nature of any toxic principles in rhubarb leaves and in other species of the Polygonaceae family. With regard to chronic effects of oxalate-containing foods in human nutrition, there are a number of reports in the literature dealing with this subject. The principal question raised is whether there are any practical circumstances in which the oxalic acid content of food would sufficiently affect calcium metabolism to produce a deleterious effect. The same question has also been raised with the phosphate and phytate content of the diet, and considerations have also been given to other organic acids such as citric, malic, tartaric, and benzoic in relation to

OXALATES 261 the possibility of influencing calcium metabolism. The role of oxalates was reviewed by Jeghers and Murphy!° and also discussed briefly in a review by Nicolaysen et al.29 It is difficult to discuss the possible effect of oxalates in foods without some consideration of modern concepts of calcium metabolism and calcium homeostasis. There seems to be general agreement!53° that man and a number of other animals have an extraordinary ability to adapt to widely varying intakes of calcium without the production of deleterious effects. In many parts of the world the daily intake of calcium is perhaps half or less that occurring in the United States and yet the evidence for a true calcium deficiency is not readily apparent. It is also obvious that the dietary content of vitamin D has a major effect on the uptake of calcium from the intestinal tract. In addition, it is known that the efficiency of absorption of calcium may be greater in individuals on a low calcium diet than in those on a high intake. The higher percentage of absorption of calcium in the young growing individual is, of course, well known. Whether a borderline calcium deficiency in older individuals is related to the general tendency to osteoporosis is uncertain, and there seems to be no clear evidence that calcium deficiency is involved, although theoretically this is possible. In general, it is apparent that, because of the very large storage of available calcium in the bony structure of the body, many species including man adapt very successfully to widely varying levels of calcium intake provided certain diseases are not present and provided vitamin D intake is adequate for that species. One of the earlier and more‘comprehensive reports on studies of the possible effect of oxalic acid in food is that by Kohman.!2 This article reported an extensive study of the calcium and oxalate content of a wide variety of foods, and it was pointed out that there was no necessary relationship between the calcium and oxalate content. Some foods, such as spinach, although having a high calcium content have a sufficiently high oxalate content to interfere with the absorption of the contained calcium and with other dietary calcium as well. In this experiment 21- day-old rats were fed a basal diet of roast beef, peas, carrots, and sweet potatoes; the calcium content was only 0.093 percent. To this diet was added about 5 to 8 percent of various greens containing sufficient calcium to raise the total calcium level to about 0.22 percent. Dietary feeding was continued for a period of 21 days. Under these conditions the addition of spinach having an oxalate content of about 0.9 percent and a calcium content of 1.25 percent on a dry weight basis caused interference with growth and bone formation. On the other hand, if the

262 D. W. FASSETT rats were fed greens containing fairly high calcium content (2-4 percent dry weight) but a low oxalate content (0.14 percent dry weight), such as turnip greens, kale, mustard greens, and collards, they showed good growth and bone formation. If additional calctum carbonate was added to the spinach supplement in the diet, then growth and bone formation were comparable to those obtained on a greens diet low in oxalates. When the dietary feeding period was continued until the rats were 90 days of age, the group on the spinach diet had obvious bone difficul- ties including dental abnormalities and soft pliable bones. It seems probable from the results of this experiment that the animals may have been on a borderline vitamin D as well as very low calcium intake. On the other hand, MacKenzie and McCollum!* did not produce serious effects in the rats at an oxalate concentration in the diet of 2.5 percent unless there was also a deficiency of calcium, phosphorus, or vitamin D. For example, on a 1.7 percent potassium oxalate diet, containing 0.35 percent calcium and 0.35 percent phosphorus, some slowing of growth and bone formation was noted. If vitamin D intake was optimum and the calcium and phosphorus contents of the diet were 0.6 percent and 0.7 percent, respectively, then the presence of 0.9 percent potassium oxalate did not affect growth or percent bone ash in a period of 10 weeks. The 2.5 percent level of potassium oxalate did, however, cause a somewhat lower bone ash but no growth effects. It was concluded that it would be impossible for a person with an adequate calcium intake to ingest enough oxalate from foods in a normal dietary for it to be harmful but that, if for some reason there was borderline intake of calcium plus a high oxalate intake, calcium deficiency might result. One of the most extensive experimental studies of the chronic toxicity of oxalic acid is that by Fitzhugh and Nelson.’ In this case, addition of oxalic acid to the diets of rats at levels varying from 0.1 to 1.2 percent caused no significant effects over the entire 2-year period of the experiment. No mention was made of any type of bone deformity. While the calcium content of the diet was not specifically stated, it is believed that the diet consisted of a standard laboratory chow which usually contains calcium in considerable amounts (probably 1.5 percent). In any event, the conclusion could be drawn that under the circumstances of this experiment even massive quantities of oxalic acid produced no toxicity. Another pertinent study is that by Nicolaysen ef a/. in which rats were placed on a vitamin D-free diet containing 0.5 percent calcium and 0.5 percent phosphorus.!9 When 0.5 percent sodium oxalate was added

OXALATES 263 to this diet, some resorption of calcium from bones was noted and a negative balance occurred. However, if vitamin D was added to the diet, there was a large increase in uptake of calcium even in the presence of the sodium oxalate. In other words the vitamin D-fed rats adapted their calcium absorption when vitamin D was present to meet the deficiency encountered by reason of the presence of oxalates in the diet. Since the calcium requirements vary considerably in different species,> it is difficult to predict the effect of food oxalate on human calcium metabolism. For example, in the young male adult of various species, requirements for calcium in terms of mg/kg/day is stated to be as follows: man, 12; cattle, 22; dog, 260; monkey, 155; rat, 160; and mouse, 900. While there may be some question as to the validity of the very high values for dogs and mice, it is evident that several species require ten times as much calcium on a body weight basis as does man. Some of the more important human studies on calcium metabolism or the effects of oxalate on human calcium metabolism are pertinent to this discussion. Extensive long-term calcium balance studies were carried out by Malm!5 over periods of 70 to 812 days with carefully controlled calcium intake levels of 937 mg/day and 459 mg/day. The results showed remarkable adaptation of most individuals to this drastic restriction in calcium intake although the length of time for this to occur varied considerably between individuals. The adaptation was probably achieved primarily through an increase in absorption efficiency in the intestine as well as through some reduction in urinary calcium excre- tion. It seems likely, therefore, in view of this comprehensive study, that food oxalates would have to reduce calcium availability more than half in order to cause deleterious effects. Another human study was that by Johnston et al.!! Six college women, age 21 years, were placed for a 4-week period on a basal diet containing 820 mg of calcium per day. During a second 4-week period, 120 g of spinach was added daily to the diet. The spinach provided an added amount of calcium equivalent to 160 mg and had an oxalic acid content of 0.6 g. The diet was adequate, but probably provided a rela- tively low vitamin D intake. During a third 4-week period, the time of feeding the spinach was changed from morning to night; the milk in the diet was not taken in conjunction with the spinach. Aside from an increase in fecal output of calcium more or less equivalent to the calcium content in the spinach, no significant changes were noted in calcium balance. Urinary output showed no significant effect. All the subjects had a small negative balance throughout the entire period including the first 4-week control period.