1 15 1 https://www.nal.usda.gov/exhibits/speccoll/files/original/45198a645e3110b9b805baa41c65db6d.pdf 58b76261f0cc9ccdf32dcbb44955e332 PDF Text Text Item ID Number 0292G Author Verret, Jacqueline D Corporate Author Report/Article TitiB Statement of Dr. Jacqueline Verrett, Food and Drug Administration, Department of Hew JOIirnal/BOOk TltlO Hearings Before the Subcommittee on Energy, Natural Year Month/Day Color Number of Images D 85 Descriptor! Notes Thursday, November 01, 2001 Page 2926 of 3007 �EFFECTS OF 2,4,5-T ON MAN AND THE ENVIRONMENT ' WTO HEARINGS iBKPOKK TUB SUBCOMMITTEE ON ENEMY, NATURAL RESOURCES, AND THE ENVIRONMENT^ 0 / 1 ''<•>•-/ OF THE ° ''^COMMITTEE ON COMMERCE/^ '' UNITED STATES SENATE NINETY-FIRST CONGRESS SECOND SESSION ON EFFECTS OF 2,4,5-T ON MAN AND THE ENVIRONMENT Al'lUL 7 AND 15, 1070 Serial 91-60 Printed for the use of the Committee on Commerce U.S. GOVERNMENT PRINTING OFFICE 45-362 WASHINGTON : 1970 �190 191 Senator HART. Our next witness, as the Surgeon General has alj-eady indicated, is Dr. Jacqueline Verrett. weight losses occurred after 3 weeks, and death at 4 weeks. The pathology observed was congestion of the lungs and mottled livers. Dogs fed 10 percent toxic fat in their rations lost hair on their backs and shoulders (alopecia), and there was poor reproduction and lactation performance. Whelped pups were either dead or weak, and the mothers seemed to have an insufficient milk supply. Pups removed before weaning and fed a normal ration showed an immediate and dramatic increase in growth. Other litters maintained on the toxic fat ration postweaning demonstrated inferior growth performance. Monkeys have demonstrated considerable sensitivity to toxic fat materials. In one study nine monkeys received a toxic triolein at a level of 25 percent in their diets. One monkey died at 1 month, and four died at 3 months. At the 3-month period, corn oil was substituted for the toxic triolein, but the other four monkeys died from 3 weeks to 5 weeks later, in spite of this substitution. Of the nine monkeys fed the toxic triolein, eight were autopsied and showed signs of jaundice, pancreatic atrophy, and fibrosis, hemosiderosis, fatty liver with necrosis, bile duct proliferation, and gross hemorrhage in the intestinal tract. No such pathology was seen in the control monkeys in this study. A second study with 36 monkeys given a toxic fat at levels from 0.125 to 10 percent of their diet, demonstrated an inverse relationship between the concentration of the toxic fat in the diet and their mean survival time. Those given the highest level (10 percent) had a mean survival time of only 91 days, while those given the lowest level (0.125 percent) had a mean survival time of 445 days. It has been estimated that the highest level provided approximately 728 micrograms total chick edema factors, while the lowest level diet provided approximately 100 micrograms total intake. The toxic fat was lethal at all levels studied, and the animals were sacrificed when possible just before death. During the last 30 days of life, all monkeys developed alopecia, generalized subcutaneous edema, accumulation of fluid in the abdominal and thoracic cavities, and hydropericardium. There were decreases in red and white blood cell counts, total serum protein values, and altered serum-protein ratios. There was also cardiac dilatation and myocardial hypertrophy and edema. Finally, the experimental monkeys had reduced hematopoiesis and spermatogenesis. degeneration of the blood vessels, focal necrosis of the liver and gastric ulcers. Limited experimentation with mice, pigeons, and turkeys, indicated that toxic fat in the diet led to reduction in growth without hydropericardium or accumulation of abdominal fluid. Similarly, swine, fed toxic fat at a level of 9 percent of their ration, showed poor weight gain, but one pig sacrificed 6 weeks after the start of the study showed no gross or microscopic lesions attributable to the ration. One important finding in the studies with chickens, is the apparent storage of the chick edema factors in chick tissues. The unsaponitiable fraction of carcasses (exclusive of intestines, head, and feet) of chickens fed the toxic fat was very potent in producing hydropericardium in other birds when incorporated in their rations. Other investigations of the distribution of the chick edema factors in the STATEMENT OF DR. JACQUELINE VERRETT, FOOD AND DRTJG ADMINISTRATION, DEPARTMENT OF HEW Dr. VERRETT. Thank you. Mr. Chairman, for this opportunity to discuss our investigations of the relationships between chlorinated phenoxy herbicides, chlorinated dibenzo-p-dioxins, and the chick edema factors. Chick edema disease was first recognized in 1957, when large numbers of broiler flocks in the United States suffered what appeared to be an epidemic disease. The affected birds appeared droopy, with ruffled feathers, and had difficulty breathing. In many flocks, more than 50 percent of the birds died as a result of the disease. Of the millions of birds affected, those autopsied consistently displayed hydropericardium (accumulation of fluid in the pericardial sac), accumulation of fluid in the abdominal cavity, subcutaneous edema, and additionally liver and kidney damage. In 1958 the investigations of a number of laboratories indicated that the causal agent was contained in fats, and specifically in the unsaponifiable fraction of fats in the commercial poultry rations. In laying hens the toxic fat caused a rapid drop in egg production. Pullets receiving toxic fat during the full growing period did not come into production, and mortality was very high. Hydropericardium. the most common lesion found in young birds, was not found in birds of laying age. The chick edema factor found in the toxic fat in the 1957 outbreak was presumed to have arisen as a byproduct of industrial production of stearic and oleic acids, since the unsaponifiable materials from this process were the components of fat in the poultry ration. Subsequently, the toxic substance was found to be present in several different types of fats. It was demonstrated to be present in samples of commercially produced oleic acids and triolein. in acidulated vegetable oils, and in inedible animal tallows. The demonstration of the presence of the chick edema factor in commercial fats led to the ruling by the Food and Drug Administration in 1961 that higher fatty acids intended for food additive use must be free of the chick edema factors. The presence of the factor was to be ascertained by a chick bioassay based on the volume of pericardial fluid in birds fed the fat under investigation. Beginning in 1958, fat that had first been proved to be toxic to chicks was used by various investigators in experiments with other species, and was demonstrated also to produce deleterious effects in rats, mice, turkeys, pigeons, guinea pigs, swine, dogs, and monkeys. Early investigations of feeding toxic fats to rats indicated that they are more resistant than chicks in short term feedings, but when fed in sufficient dosage, extracts of the toxic fat produced definite deleterious effects as shown by growth depression, enlarged and fatty livers, marked involution of the thymus. and enlarged adrenals. Guinea pigs fed 2*4 percent toxic fat stopped growing at 6 weeks, and death losses occurred at 8 weeks. At a level of 4*4 percent the �192 193 chick tissues indicated significant levels in bone, heart, intestine, kidney, liver, and skin. The liver contained more than 80 percent of the total detected. A similar determination of the distribution in rats indicated the presence of chick edema factors only in liver and in the feces. During the years that the previously described toxicity investigations were taking place, the toxic fats were undergoing intensive chemical analyses to concentrate, purify, and finally determine the nature of the compounds responsible for chick edema disease. At all steps of these procedures, the path of the toxic material was confirmed by assay in young chicks. This proved to be a time-consuming and difficult job because of the complexity of the fatty materials. A major breakthrough in this effort came when it was found that a highly purified crystalline material possessing the properties of chick edema factor contained chlorine. This indicated that it was not a natural component of the fat in which it occurred. AVork in several laboratories obtained similar results, and examination of the purified material by a variety of analytical techniques suggested that chick edema factors could be highly substituted (chlorinated) derivatives of naphthalene, biphenyl, anthracene, or even structures common to the chlorinated pesticides of the DDT family. These latter compounds were ruled out when tested in the chick feeding assay, but some derivatives of the former classes of compounds were tested and found in some instances to be toxic, and indeed produce similar lesions to those observed with authentic toxic fat. However, none of these compounds demonstrated the high order of toxicity. or the complete chick edema syndrome when so tested. Finally, by means of single crystal X-ray crystallography, it was demonstrated that a pure compound isolated from a toxic fat was a hexachloro-dibenzo-p-dioxin. This structure was verified by infrared, ultraviolet, and mass spectrometry data. Final confirmation came when this particular compound was synthesized and found to produce the same lesions in chicks as the compound isolated from the toxic fat. The finding that a chlorinated dibenzo-p-dioxin was a chick edema factor explained why different investigators had isolated materials similar in their capacity to elicit chick edema disease, but yet in their purest forms, had slightly different chemical properties. The large number of isomers possible (more than 60) in this family ranging from mono- to octa-chloro-dibenzo-p-dioxins illustrates the complexity of the problem. It then became a problem of determining whether some or all of these compounds are in fact chick edema factors, and what their relative capacities in this regard might be. The chlorinated dibenzo-p-dioxin structures have been known in organic chemistry many years, and became particularly noteworthy, when in manufacturing processes with chlorophenols. their formation as byproducts posed serious occupational hazards. The most potent in this regard seems to have been the symmetrical tetrachlorop-dioxin which was formed in the manufacture of 2.4.5-trichloro' phenol. These chlorinated compounds were found to cause a serious and persistent disease referred to as chloracne. This disease was first described in 1899. Associations of this disease with chlorinated di-benzo-p-dioxins were made by the Germans, who had several out- breaks of this disease in their factories. There have also been similar occurrences in the Netherlands and in this country, in factories manufacturing chlorophenol compounds. It should also be pointed out, that other compounds, such as the chlorinated naphthalenes, anthracenes, biphenyls and dibenzoftirans, are known to be acnegenic, but as in the case of the toxic response in chicks, these materials are less potent than the chlorinated dibenzo-p-dioxins. In the case of the chloracne associated with dioxin, the human symptomatology extends to other mucous membrane irritation, porphyria cutanea tarda. hirsutism, hyperpigmentation, increased skin fragility, severe damage to the internal organs, particularly hepatotoxicity, and central nervous system disorders, as indicated by neuromuscular symptoms and psychologic alterations, and other systemic symptoms. Most of these occupational exposures in Germany occurred in the 1950's, and followup examination of these affected workers in recent years indicate that the recovery period is lengthy, with many workers still having demonstrable adverse effects from prior exposure. Similar observations have been made on exposed workers in the United States of America. The tetrachlorodibenzo-p-dioxin was demonstrated to produce the chloracne in humans after the application of only 20 micrograms. The rabbit ear is especially sensitive, with concentration of 0.001 percent to 0.005 percent producing severe reactions after local application. This assay using the rabbit ear is apparently used as an indicator in some plants of the content of this particular dioxin in the manufacturing process. Hence, the serious health significance of these compounds for humans has, inadvertently, been clearly documented. Research in Germany and Japan indicated that the magnitude of this problem was indeed large, since the formation of the chlorinated di-benzo-p-dioxins would be facilitated in the saponification procedures used in various processes involving chlorophenols. A further complication is that a given chlorophenol preparation is generally contaminated with other isomers, increasing the possibility of formation of a wide spectrum of chlorinated di-benzo-p-dioxins beyond those to be expected from the predominant component. Evidence that this does occur will be discussed shortly in connection with the chicken embryo studies of these materials. During the time the previously described investigations of the chick edema factors were underway, many of which were carried out by FDA investigators, methodology was developed for detecting the chick edema factors using sensitive gas-liquid-chromatographic (GLC) techniques. It became apparent that authentic toxic fats consistently gave peaks with specific retention times, and these peaks were used as an indication of chick edema factor in a suspect sample. Confirmation of this was obtained using the chick feeding assay. In the light of recent knowledge of the chlorodioxins as chick edema factors, it has been possible to establish that the materials being detected were hexa-, hepta-, and octa-chlorodibenzo-p-dioxins. Although toxic fat samples did indeed contain peaks corresponding to dioxins of lesser chlorine content, i.e., di-, tri-, tetra-, penta-, these are not detectable with this particular analytical procedure because their particular peaks are obscured by other components, j : ' .' , �-', ' ' , 194 including pesticide residues present in the samples. Other GLC procedures are being developed to detect these latter dioxins. " In the early 1960's the chicken embryo was being used in toxicological evaluations of a wide variety of materials. It was hoped to develop a rapid and sensitive screening system to pinpoint compounds with significant toxic and teratogenic effects for further study. In view of the demonstrated sensitivity of the chicken to chick edema factors, the chicken embryo was used to assay toxic fat samples, and found to present the same syndrome as observed in the chick feeding assay. A high mortality was observed with toxic fat extracts, and additionally, hydropericardium. generalized and massive edema, eye, beak, and leg defects, and necrotic livers were apparent on gross observation. No microscopic studies have been conducted on embryos or hatched chicks in these investigations. In parallel with other investigations, the chicken embryo was used to test the toxicity of the chick edema factors isolated from toxic fats. It was also found that the chlorinated biphenyls. naphthalenes, anthracenes, and other compounds did indeed elicit a toxic response, and in some instances, the chick edema syndrome was present. But in no case were any of these materials as potent as the toxic components isolated from toxic fats, and were generally less potent by a few orders of magnitude. After the identification in 1966 of a hexachloro dibenzo-p-dioxin as a chick edema factor, studies were initiated in which various isomers of chlorinated dibenzo-p-dioxin were prepared and tested in the embryos. Although the investigation was not extensive or complete, it illustrated that isomers prepared by pyrolyzing selected chlorophenols did give chloro-dibenzo-p-dioxins with GLC retention times duplicating those in the authentic toxic fats, and likewise, produced the chick edema syndrome in the treated embryos. It was also apparent from this study that the various isomers (that is, those with different chlorine content, and those with identical chlorine content, but with chlorine atoms positioned differently on the molecule) varied in their toxicity, although in all cases only microgram or less quantities were required to elicit the toxic response. It is not possible to give exact figures for the toxicities obtained in this study, since most of the individual dioxins were contaminated with traces of others. Nevertheless, it was apparent that the symmetrical chlorodioxin prepared from 2,4,5-trichlorophenol (2.3,6,7-tetrachloro dibenzo-p-dioxin) was more potent than any of the others tested, even recognizing its lack of purity. During this investigation, samples of the chlorophenols. both technical and reagent grades, were examined by GLC to determine if preformed chloro-dioxins were present. The presence of chlorodioxins was demonstrated by GLC, and these materials, which can be removed by appropriate techniques, were then tested in the chicken embryo system and did indeed produce chick edema. A current study of similarly contaminated chlorophenols. containing from 18 ppb to 95 ppm of chloro-dioxins with six or more chlorine atoms are currently under test. This investigation was not pursued further in view of the fact that there had been no known occurrences of chick edema disease since • the late 1950s and. hence, such research had a low order of priority. 195 However, in early 1969 there was another large outbreak of the disease in North Carolina and the toxic factor was traced to the fat component of the feed. The toxicity was confirmed by the chicken embryo test, with fractions of the hexa-, hepta-, and octa-chlorochbenzp-p-dioxms from the crude fat. An investigation of the processing plant in which the toxic vegetable oil products were produced revealed a proximate operation for the manufacture of chlorophenol formulations. However, it is still not possible to conclude that this accounted solely for the presence of dioxins m the fat, or whether they were at least to some extent in crude oil from a prior contamination. Since that time FDA has initiated an investigation of the oils of other manufacturers and processors and in a few cases GLC analysis has indicated the presence of chloro-dioxins. These have not as Vet been confirmed by chicken embryo bioassay; however, it is noteworthy that in the case of these subject samples there is no known adjacent manufacturing operation that would give rise to direct chloro-dioxm or chlorophenol contamination, so that entrv of the chloro-dioxms from other sources must be considered. This 1969 outbreak of chick edema disease, coupled with the question of contamination of the herbicide 2,4,5-T by chlorinated dioxins led us to renew this investigation. In the past 6 months the herbicides •2,4-L> and 2,4,5-1, as well as the particular tetrachlorodioxin purported to be the teratogenic agent responsible for the effects in the Bionetics study of 2,4,5-T have been under study. In an effort to assess the edema-producing capacity, the teratogenic activity, and the acute toxicity of these materials', samples of the original Bionetics 2,4,5-T, from Diamond-Alkali, were obtained for comparison with a sample representative of the current manufacture of Dow Chemical Co. The Bionetics 2,4,5-T is reported to contain 27 plus or minus 8 parts per mil ion of the 2,3,6,7-tetrachloro dibenzo-p-dioxin, with the Tr8^ 1° nK16r dl°xms unkno™- The current production of Dow jJ.4,5-1 has 0.5 parts per million of this dioxin, with no analysis for higher chloro-dioxins reported, but does contain almost 5 percent of other impurities, mostly isomers of 2,4,-D 2,4,5-T. and chlorophenols and chlorophenoxy compounds of undetermined structure All investigations using the chicken embryo involved administration o± the compounds by injection through'the air cell of the we either premcubation or at the 4th day of incubation A comparison of the Bionetics 2,4,5-T with the Dow 2 4 5-T indicates that with respect to the ability of the materials to produce embryonic mortality, the Bionetics 2,4,5-T is more potent, The Bionetics 2,4,5-T has an LD50-that is, kills 50 percent of the treated embryos-of approximately 25 micrograms per egg, 0.5 parts per million, while the Dow 2,4.5-T LD/is approximately 100 micrograms per egg, 2 parts per million. With respect, to teratogenic effects, both samples produce chick edema syndrome m the nonviable embryos, and 'hatched chicks in eluding e>e defects beak defects-predominantly cleft palate-short and twisted feet-the result of tendon slippage-and diffuse and localized edema in various parts of the body. uuiu&e ^na �196 "With both of these samples of 2,4,5-T these teratogenic effects are "observed at levels inducing no significant embryonic mortality. The Dow 2,4,5-T still produces the chick edema syndrome at 50 micrograms, one part per million, a level where only 12 percent mortality is observed, while the Bionetics sample has similar effects as low as 6.25 micrograms per egg, 0.125 parts per million, a level inducing only 16 percent mortality. Both of these mortalities are close to that induced by the solvent alone. It should also be emphasized that the chick edema syndrome is not observed in the embryos treated with the solvents only, at any level, or in the control flock. A sample of 2,4,5-T from a chemical supply company was subjected to three recrystallizations before test. With the present GLC techniques no chlofodioxins are detectable in this purified sample. When tested in embryos it produced chick edema syndrome at 5, 10, and 25 parts per million, all levels which induced no more than 15 percent mortality in the embryos. This same sample was subjected to an additional purification by seven extractions to remove dioxins that might have been present, but were below the current detection levels. This repurified sample is still clearly teratogenic in the embryos since when tested at a level of 2.5 parts per million it produced 20 percent incidence of the malformations previously described, though no significant edema was seen grossly. The mortality induced was 24 percent, which is higher than that of the sample prior to the extensive extraction procedure. It is also noteworthy that the embryonic mortality occurred soon after treatment, and the hatched chicks had bleached down, indicative of an aberration in the normal pigment formation. With respect to the 2,3,6,7-tetrachlorodibenzo-p-dioxin, early investigations of this compound in a preparation containing some 2,3,7-trichlorodibenzo-p-dioxin indicate a high order of toxicity and teratogenicity. Whether prepared by pyrolysis of 2.4,5-trichlorophenol. or direct chlorination of dibenzo-p-dioxin. the test preparations, which contained approximately 50 to 55 percent of the tetrachlorpdioxin and 20 to 25 percent of the trichlorodioxin, produced significant mortality, that is greater than 20 percent, and chick edema syndrome in more than 40 percent of the treated embryos at levels offivetenmillionths of a milligram per egg, or 10 parts per trillion. More recent investigations, with two samples of the tetrachlorodioxin, both of purity greater than 95 percent, indicate edema and terata at 20 parts per trillion. These samples have only become available within the past month and additional testing is underway at lower and higher levels. It should also be mentioned that the herbicide 2.4-D as a commercially available sample, and a purified sample, a mixture of the N-butyl esters of 2,4-D and 2,4.5-T. and a sample of silvex, a related herbicide, have been tested. Terata and chick edema syndrome have been observed with all of these materials at levels of 10 parts per million and above. Lower levels are under investigation and the levels of dioxins in these samples are also being determined. 197 Studies have recently been initiated in the FDA using pregnant golden hamsters, mtubated on day 6 through 10 of orfanogtnesis with the test compounds. 1JP18 ?°W-2,'f'5;T', °'5 parts Per milli°n tetrachlorodioxin, tested at 100 m./k. yielded about 80 percent fetal deaths and those pups born alive had gastrointestinal hemorrhages ^iStf -rf^st,a.lliz.ed 2>4>5-T sample referred to earlier, with no detectable chlorodioxms, when tested at a level of 100 m /k oroducd an average fetal mortality of 55 percent. ' Among 38 live pups, three abnormals were found: One with a deformed hind limb and two with inadequate fusion of the skull At lower doses the fetal mortality was less, but still higher than that observed in control hamsters. ^ When the extensively repurified sample of 2,4,5-T was tested in hamsters no gross terata were observed at 100 m./k., but the number of early fetal deaths was 70 percent, indicating a definite embrvotoxic effect and corroborating the observation of increased mortality J studies are underwJ ° - Additional tests with this compound A dioxin preparation containing approximately 51 percent 2,3.67tetrachloro- and 21 percent 2,3,6 trichlorodibenzo-p-dioxin vielded 98 percent fetal deaths at 9.1 micrograms per kilogram. Gastrointestinal hemorrhages and eye anomalies—absence of lid—were present in many of the pups. Tests with the purer tetrachlorodibenzo-p-dioxins are underway The numbers of animals in the hamster tests are too small to be considered statistically valid, but there are definite indications that alterations in fetal viability and gastrointestinal hemorrhages do 6 occur at the levels tested. In summary- the chick embryo studies and additionally the preliminary hamster data indicate that the current production 245-T containing 0.5 parts per million of the 2,3,6,7-tetrachlorodibenzo-pdioxm is teratogenic and embryotoxic in these test systems Further an extensively purified 2,4,5-T sample, with no chlorodioxms detectable with the present techniques, has indicated significant embryotoxicity in the hamster and chick embryo, and Idditionally produced gross terata in the chicken embryos, making it impossible at this point in time to exonerate it of teratogenic or other adverse effects on the embryos that may have some health significance. The data for 2,4-D in chick embryos likewise demonstrate these effects in current production materials ^nlwl5di^hrY?T"0 ™yassess*d another and perhaps more complicated aspect of this problem, and that is the interactions of the various ch orodioxm isomers with each other in the many combinations m which they are likely to occur, or the possible interactions including potentiation or synergism, between the chlorodioxins and the chlorophenols, herbicides and other materials in which thev J are found. At this point, with your permission, Senator Hart, I would like to insert m the record the documents containing the data from which this testimony was derived. wmui Thank you. That is the end of my prepared statement. �198 199 Senator HAKT. Yes, they will be received and placed in the record after your oral testimony Dr. VERRETT. I realize I presented a rather lengthy and complicated piece. I will be happy to answer questions and elaborate if you wish. I also have some samples of embryos, if you would care to see them, or chicks. Senator HART. Yes, I heard about them. The statement is not an easy one for one not trained in your discipline. I take it. Doctor, these are some of the chicks that have the deformities that you are talking about ? Dr. VERRETT. That is right. These are chicks that have been treated with the tetrachlorodibenzo-dioxin. I would like to put them out on the table but it is not safe. As you can see, they cannot stand up. These chicks hatched early this week. They have noticeable edema. This is really the ankle, if you wish, of the chicken and this is really a result of the malformation. These are normal chicks: as you can see. they have no difficulty standing and walking. These are chicks that have been treated with the ethanol. the solvent alone or have been untreated. We always run some at the same time that have had no treatment and I think the comparison between that one and this one, if you wish, is quite obvious. Senator HART. Your last statement referred to a series of compounds. Dr. VERRETT. Yes. Senator HART. Exactly what are these ? Dr. VERRETT. It is a family of compounds which I may say very briefly is a series of phenols having a varying chlorine content. They go from dichlorophenol, which is the precursor or the compound from which 2.4-D is made, all the way up to pentachlorophenol. All of these materials are very widely used as herbicides. They also have a very broad use in indiistry and. in other words, are capable of being in the environment and if they are contaminated with dioxins, of also putting dioxins in the environment. What I was trying to say is that we have looked at these chlorophenols. apart from just the herbicides which are derived from them, and we know from past study, and there is a paper in the material that I have included in the record which demonstrates, that these materials are already contaminated with dioxin. The amount of the contamination at this point is Senator HART. It is very difficult to hear the witness. I do not know what the distraction is. but please be patient. Dr. VERRETT. We do know, at least by GLC (gas-liquid-chromatography) just chemical techniques—I will distinguish between chemical and biological techniques-—by chemical analysis, we know these materials are contaminated with chlorodioxins. In addition to that, we know there are many of these chlorinated dioxins that are involved and not simply the single one which has been the subject of the -2,4.5-T. There are approximately 60 or more. We are aware from our previous work that these materials are already in chloroplienols as they are being used. This is the point I was trying to make. Senator HART. The point I was trying to make by raising it is the necessity that there be no more delay with respect to establishing the dangers, the hazards, the potentials for harm in this whole variety, this whole family of products, given the kind of damage that you have so dramatically demonstrated here from one single element of the environment. Dr. VERRETT. Yes. I might say—of course I could not bring very many-—-these are chicks treated with the tetradioxin in question, which is the most pertinent to the hearing. There are some here which have been treated with 2,4-D. Thes are 2.4,5-T treated chicks. We do have underway an investigation of all the other dioxins. It is a question of having to synthesize these materials and test them in pure form. But by inference at least, in our experience with the chick edema problem in the past, we know all of these dioxins are in fact toxic. We do know that. We have that information. As I said, at least in chicks, we know there is a storage and possible transmission from chickens that have been fed these materials to the human. The possibility does exist. Senator HART. By handling, as you just did? Dr. VERRETT. No. I hope not. The tetradioxin is potent. I do not know whether handling this bird will harm you. but chloracne is a very serious disease and persistent disease and there is no known cure. We are not aware that the other dioxins are as potent as that in this regard, but they have yet to be evaluated. To our knowledge the herbicides we have been discussing have not actually been tested to see if any other of the dioxins are also present. The concern, apparently, has revolved around the particular dioxin because it is extremely potent: even if the others are less potent, of course, they may still be important. Senator HART. I understood your explanation with the rats and the monkeys to indicate that this—that the dioxin in 2,4,5-T has an accumulative effect. Dr. VERRETT. The effects that are seen would indicate that. I would he unable to say that has been proven, but the fact that, for example, in the one monkey study, when they were fed for 3 months and then the animals that did not die during that course were put on a ration free of it, still died in subsequent months, and that would indicate that either the damage had been done by the initial exposure or there was storage such that it finally did have its effect. But it was fatal in all cases. So the animal data would indicate— again we cannot apply that with certainty about the human—storage or persistence, if you wish to use that word. Senator HART. Then your attitude, which you say is not conclusive and certainly does not relate directly to humans, points to the persistence, the cumulative character of the dioxin, rather than in the other direction, that it is not cumulative? Dr. VERRETT. I would say it indicates it is probably cumulative. �; . ' . 200 Senator HART. Do you know any experiments that point in the direction that it is probably not cumulative ? Dr. VERRETT. I am not aware of any, no. The human data I cited with respect to occupational exposure would also indicate that it is probably cumulative in the human, of course. Senator HART. You used a word here which I take it means burning. Dr. VERRETT. Pyrolysis. Not exactly bxirning. In the sense I used it, it means reacting xuider conditions of elevated temperature, but not the actual burning of the material itself. It indicates a high temperature reaction, perhaps, heat applied to the material in order to make the reaction take place, but not in the sense of actually igniting it. But it does indicate that heat, in other words, facilitates the formation of these compounds (dioxins). if that is what you are arriving at. Senator HART. What if some of this material is just put in the city dump and burned? Could that burning inadvertently produce dioxin ? Dr. VERRETT. I could only say that the likelihood is there; yes. Again, lack of actual experimentation does not give us any evidence or proof of this, but the fact is that these materials are formed when chlorophenols are subjected to heat and that would indicate to me that that is definitely a possibility. Senator HART. What common products, or what common articles contain chlorophenol ? Dr. VERRETT. There are so many that I wouldn't be able to name all. But, every piece of newspaper or paper of any kind probably has chlorophenol used in the manufacture of paper. I am sorry I am not in an area which would enable me to give you total usage figures, but they are considerable. This is washed out to some extent, but there probably are some chlorophenol residues, and paper would be one item and one which we can say is very widely used. Another example—well, leather is cured by using chlorophenols in the tanning process. So leather materials contain it. There wotild be any number of other everyday items that would possibly have it. Senator HART. Well, just as I am reluctant to have pictures taken, I am reluctant to make these contrary statement and yet this one is not inappropriate. If the materials that we customarily—and for generations, centuries I guess—have been throwing into a fire contain chlorophenol, when you burn them, it is your opinion that dioxins can result Dr. VERRETT. Could possibly result; yes. Senator HART. Is it possible that some of the birth defects for which there has been no medical explanation to date are the result of this kind of thing, where we have always done it and it has never seemed to hurt us ? - Dr. VERRETT. I would say it is a possibility. It would be for others to assess this situation, but I think it is a distinct possibility because of the fact that these materials (chlorophenols) are ubiquitous, and j.f, in fact, chlorodioxins are formed in the environment, this is a possibility. 201 I should point out the studies done in mammals have been done by feeding, while the largest exposure may come from inhalation or dermal contact. That was the source of exposure occupationally to the tetrachlorodioxin. So here we have to be concerned not only about eating, but inhalation and perhaps contact exposure. This brings up the subject of other materials. For example, we are also investigating Senator HART. I was just going to say that it is hard to visualize a substitute for paper or leather, but can one have leather and paper without this material ? Dr. VERRETT. I would think so. I should point out some of the materials (chlorophenols) are washed out in the processing. The total amount used in the processing does not always remain in the products. I did not mean to imply that. That brings up another question: Are they washed out into the rivers et cetera ? Nevertheless, I am not really sufficiently knowledgeable of the technology to say whether something else could be substituted or not. Senator HART. Mr. Bickwit ? Mr. BICKWIT. I have heard several people criticize these chick embryo studies as being overly sensitive. I am not sure of this, but I believe the Secretary of HEW has criticized them. Can you respond to that criticism ? Dr. VERRETT. Well, I have not as yet. As I mentioned in the testimony very briefly, this techniqiie was started with just that idea, of finding a sensitive technique, if at all possible, for assessing toxic and also teratogenic response. One of the great difficulties in all of the toxicology animal work, using mammals and primates, even, is the difficulty of relating, of course, to the human. There is perhaps another generation gap, if you can use that word, between the chicken or an avian species. I would say without hesitation that these studies, for example, proved that this material is not for the birds, if I may phrase it that way. I would certainly not say that you can conclusively state that there is a human hazard from the results with the chick embryo. However, inferentially we have evidence of that, and we have the inadvertent tie-in because of the chlorodioxin toxicity already known in primates and humans. I would certainly hope the human is less sensitive than the chick, but I feel what we really need, and I think that is pointed out very much by these hearings, is a very sensitive test, and then it remains to show this is not the case in the species more closely related to man. I would rather demonstrate that something has an adA-erse effect in a sensitive system and then, by appropriate study, find out whether it will be relevant to man, than miss it altogether in an insensitive test. Although we do not try to make direct correlations, we feel certain anything seen in this system is worthy of further study, and I shoiild also like to add we are trying to keep the study in the proper perspective. That is, we use levels (doses) when possible that are relevant to the human exposure or other animal exposure and not simply try to produce effects with excessive amounts of material. �202 «. ' ', Mr. BICKWIT. You state on page 4 that the most potent dioxin in the production of serious occupational hazards, seems to have been "the tetradioxin. What actual experimental evidence is there that suggests or confirms that tetra is the most potent of the dioxins ? Dr. VERRETT. You mean in animal work ? Mr. BICKWIT. Yes. Dr. VERRETT. Of course the tests referred to earlier by Dr. Steinfeld show in that system that the tetradioxin is extremely potent in rats. I am not aware that the other dioxins have ever been studied. Mr. BICKWIT. Have the other dioxins ever been compared in potency in chick analysis ? Dr. VERRETT. Yes. As far as I know only in the chick embryos. Mr. BICKWIT. So, unless we can rely on the chick studies, we may be in very serious trouble, even more serious than it now appears on the basis of available evidence. Dr. VERRETT. That is right. Mr. BICKWIT. Are there any data available, either in mammalian studies or chick studies on the relative toxicity of dioxins compared to thalidomide ? Dr. VERRETT. If you used the chick as an exhibit or even the mammalian studies with the tetradioxine. which is the only one being studied at the moment, you would have to say this material is some 100,000 to a million times more potent in these particular species. N\rw, I should add that the abnormal effects fterata) that we are seeing are not the same as we see with thalidomide, but the potential for producing abnormalties that we do find, it is of that order. Senator HART. Doctor, thank you very much. (The material referred to follows:) BIBLIOGRAPHY 1. 'The Chick Edema Factor': Anon., Nutrition Reviews 26, 28 (1968). 2. 'Studies of the Chicken Edema Disease Factor': Friedman, L.. Firestone D., Horwitz, TV., Banes, D., Anstead, M., and Shue, G., JAOAC 43, 129 (1959)! 3. 'The Occurrence of the Chick Pericardia! Edema Factor in Some Oleic Acids and Products Therefrom': Ames. S. R.. Swanson. TV. J.. Ludwig, M. I., and Brokaw, G. Y., J. Am. Oil Chemists Soc. ST. 10 (1960). 4. 'Studies of the Chick Edema Factor. II Isolation of a Toxic Substance': Yartzoff, A., Firestone, D., Banes, D.. Honvitz. TV., Friedman, L., and Nesheim S., J. Am. Oil Chemists Soc. 38. 60 (1961). 5. 'Collaborative Bioassay for Chick Edema Factor': Douglass. C. D., and Flick, D. F., JAOAC 44, 3 (1961). 6. 'Progress in the Chick Edema Problem': Friedman, L., Feedstuffs, March 17. 1962. 7. Occupational Intoxication Occurring in the Production of Chlorophenol Compounds': Bauer, H., Schulz, H., and Spiegelberg, U., Archiv fur Gewerbepathologie and Gewerbehygiene 18, 538 (1961). S. 'A Technic for Testing Acnegenic Potency in Rabbits Applied to the Potent Acnegen, 2,3,7,8-Tetrachlorodibenzo-p-dioxin': Jones, E. L.. and Krizek H., J. Invest. Dermatol. 39, 511 (1962). 9. 'Studies of the Chick Edema Disease. 2. Preparation and Biological Effects of a Crystalline Chick Edema Factor Concentrate': Flick, D. F., Firestone, D., and Marliac, J. P., Poultry Science 44. 1214 (1965). 10. 'Chick Edema Factor: Application of Microcoulometric Gas Chromatography to Detection of Chick Edema Factor in Fats or Fatty Acids': Firestone, 203 JAOAC (1063) : AOAC 11. 'The Injection of Chemicals into the Fertile Eggs Prior to Incubation is a Tenacity Test': JIcLaughlin, J., Marliac, J. P., Vetrett TI T MutcWer TI Tlutchler K., and Fitzhugh, O. G., Tox, AppL Pliarm., 5, 760 (IMS) - M. U12T £ rf theTC£icke» Embryo in the Assay of Aflatoxin ToxidtV : Terrett clleu M. J., Marhac, J. P., and McLaughlin, J., JAOAC ',? 1003 (1964)' 13. 'The Role of Toxic Fat in the Production of Hydropercardium and Ascitfts in Chickens', Allen, J. R, Am. J. Vet Res 25 1<>10 (1964) 14. 'Industrially Acquired Porphyria' : Blieberg, J.. Wallen M Brodkin R and Appelbaum, I., Arch. Derm. S9, 793 (1964) »J"en, M., Brodkin. R., eCtr n licrosco ic °T * l> Alterations in the Liver of Chickens Fed Toxic L, *?• Tn Fa: £ ™L ^,R" aniactor: Soi r- A., Lab. Inves. 15, 970 (1966) ^ Carstens> 1C ema * ? « i Tissue Distribution Data and Toxicoloeic 0 1 "e1 " '" " FDA Internal Preliminary Report, October 1906 <• , u. e., ana Kess, J.. F«f 'Toxf^Fat- I<,nClr0nTJIiCrOSC1)iS Ol)servations.,in llacaca mulatto Monkeys fl967) ' ' an Carstens' T- A" Am- J- vet. Res. 28, 1513 19. 'Note on an Improved Cleanup Method for the Detection of Cai)ture «• SflS' ?a!S' an(1 Waxes' : Xeal. P., JAOAC 51, 489 (1968) 21. Chemical and Toxicolosical Evaluations of Isolated and Synthetic Chloro Denvative.s of Dibenzo-p-dioxin' : Higginbotham, G. R Hvia "" A Firestone D, Terrett. J., Ress, J., and Campbell, A. D, Nature 220. 70- n96S)" 22. Analysis of Commercial Chlorophenols for Trace Amounts of The r Con densation and Po ymerization Products': Higginbotham, G. R and Ress T FDA Internal Preliminary Report. November 196S ' 23. 'The Identification and Crystal Structure of a Hvdropercardium Pro : ™ e n z o . 1 ^ I o x i n ' : Cantre^ J . S " s 45 Xo 5 10 (1967) an(1 Acta ' - ' , ' 24. •Clinical Picture and Etiology of Chloracne' : Schulz, K. H Arbeitsmedi zin-Sozmlmedizin-Arbeitshygiene 3 (2) : 25 (19&S) ^roeir.smecn25. 'Report on Methodology for Chlorinated Aromatics in Fats Oils and SS rlsT U970): ' ' Hi^inbothan1' G- R- and Firestone, D, JAOAC "n 26. Fecleral Register December 9, 19C1, 26 F. R. 11S28, 121.2->4 • also Code 121 lom f rther - " (CWck 28. Table: Preliminary Report on Teratology Studies wi ' T.-P. X. ' 29. Table: Preliminary Report on Teratology Studies with <>45-T Samples G in, FDA, of 12/4/69 re 31. Letter: G. E. Lynn, Dow Chem. Co., to M. J. Verrett FD\ of re composition of Dow 2,4,5,-T sample. verren, i«UA, ot 32. Memo : FDA Internal from I). Firestone to A. D. uampue11 ot d/ / Campbell of 3/-V composition of chlorinated Dibenzo-p-dioxin Standard ' 33. Memo: FDA Internal from D. Firestone to A. D Campbell of n> samples for Chicken Embryo Testing ^mpueii, ot 34. Memo : FDA Internal 'from J. Ress to A. D. v ampDeu ot 3/<>6/7n re Campbell of chlorophenol samples for Chicken Embryo Testing. ' ' ««/-*>/ «0 re > n _ ?n _ 1 4 �204 THE CHICK EDEMA FACTOU nzic A toxic fnctor In fine \ec-d rj.a :t tali and tatty ncith produces r.-jj,-uix-ricarji<iin cius ... . chirks when 'J . . and asciu* in yo-..ng „ . . *g. per laiaffTani body Height ore li-rl per day. g There are u number oj these t'-ric cotHp.mnds. containing large amounts (»/ chlorine. One has been ch-iracterized and synthesized. In 2957, large numbers of chickens suf- ' cumulation of fluid in the pericardial and fered from what appeared to be an epidemic abdominal cavities of flic affected birds, the disease. I.os.-es a t t r i b u t a b l e to tin- epiilomie hematocrit, blood volume, and moisture have been estimated in the million- of content of the heart, skeletal muscle, skin, dollars. The affected birds appeared droopy, and kidney were normal. However, the total showed ruffled feathers, and had ditliculty body water content of the birds was signifibreathing (L. Friedman. FiCiiftitfJ*. Mttrch cantly increased. On the basis of these observation!-, it was suggested that ihc chick 17, J«?-?!. In some flocks, over 50 per rent of the edema factor increa.-oil permeability of the birds died with typical symptom- of the cardiac vascular bed (Flick. Dongl:'..-s, and disease (.V- L. Sangor d «/.. J. Am. Yet. Gallo. loc. cit.). Additional support for this Mcd. Assn. 133, 1~2 \lfi.jS)). \Vln-n aii- hypothesis came from the finding that the topsied, the birds had pale hearts with hy- toxic far-tor produced no appreciable change dropcnrardium, and livers that \u-ie pale, in the proportion or level of pla>ma proteins mottled, and had an it-vegular granular sur- (Flick, D. Firestone, and .1. P. Mavliac, face. In. the advanced stages, the abdomens PovlU;i Ke!. 4-1. 111.', U.%'-'5)). The type of diet had a marked effect on were distended, and contained 100 to 500 ml. of clear, straw-colored fluid. The peri- the rate at which the chicks' bodies aceardial cavity is most susceptible to fluid cumulated water. Chicks fed a natural accumulation, with the abdominal cavity grain ration containing 4 per cent toxic fat next, and then subcutaneous tissue (D. F. showed an increased body moisture content Flick, C. D. Douglass, and L. Gr.llo, Paul-, < I7S versus 72 per cent for the controls! try .?n. 42. SoS (1963)). The kidneys were only at the end of the third week. Bird' pale and swollen: the fatty tissue of the fed the same level of toxic fat in a semigizzard was edematous: and the rhirvlenum purified diet showed a marked increase in was swollen and soft (Sanger ct al.. lac. body water content after seven days (79 cit.). Hydropericardium was not as promi- versus 73 per cent) (Flick, Douglass, and nent in the older birds as in the broilers Gallo. loc. crt.l. During 19"iS, :i number of laboratories (Friedman, loc. cit.). Ko acceptable explanation is available traced the disorder to the presence of a for the abnormalities associated with feed- toxic substance in the unsaponi(table fracing the chick edema factor. Despite ac- tion of fats added to commercial poultry 205 ration?. This fraction contained « toxic subi'sm'Cts'i which presumably was present in i.vtiroiluct of industrial production of <.>r&".< anil (1'c'l> acids. Tin toxic sub-nance was present in a \srioty of types of fats. These included a sample of triolein, which on the basis of fhenv.cal analyses was found to lit- of "ex.f'.K-nt iiiw'ity" <-\- YarUoff (I til.. ./. Am. ,'•.: Cl.(m. NT. 38. CO l l f ' i ! l \ > . AYhni this :r;o:<iii v':1* incorporated into a chick ration a: .1 level of 15 per i-c-nl, tho bird- showed -,'-.Yre symptoms of liydropericaniitim. Similar -yr.iptoms were produced when chirks •xrr? fed rations containing distillates or rt-i'iucs secured in commercial production ,; fatty neids. The toxic substaticc<Vi was .;..! found in inedible animal tallows, acidijla'.fi! vegetable oils, and several comi'!-:.'ir:lly produced oleic acid< and triolein • Trojione. ^V. Ilorwifz, Friedman, and G. M. Slme. J. Am. Oil Chcm. .Sot. 38. .',IS nf->i'\<. A report from another laboratory ii-: firmed the presence of the. chick edema :V:i--or in various industrial fats (J. C. .A!.x;::irler,R. J. Young, O. M. Burnett, and .'I. IX Hathaway, Poultry Sri. 41, 22 Tv: presence of the chirk edema factor .•i foinintrcial fat5 led to the rul'ma by the Fw.! ::ri« Drug Administration that methyl <•;'.•;•= of higher fatty acids intended for use •-.-; ;ood additive must be free of the chick . :-:.!fi factor I Federal Register Dec. f>, ! •:;. XP.K. US1S; ]-?l°?4). The presciice •:' ih«r iVtor was to be ascertained by a •'••••-'x i.:o::-say lia««l on (he volume of peri- 1 'jri!:-t! fluid in birds fed the fat under in-' •'•;:.'»• ion. Tii.it -idt was necessary in the chicks' .v.'ju for flic development of hydroperif i:'!;ii!n v/hcn toxic fats v.vrc fed was su£: -t'-'i li> Alixander and co-worker-. They '"•"•rv(.-(! typical symptoms of the di-;urb:•(•!• or.ly ivlirn (he ration containeil sodium 1 '.ri'i.'-; rxfrs talt accentuated ihe con•.'ion. .Altliout'h llii- syndrome had many '.' <'.<• f.'innrirk- of a vitamin E rleficicp.'-y. neither antioxidants (including vitamin E) nor selenium had any curative effect. Sensitivity of different species to the toxic factor in these fats varies considerably. When monkeys were fed a sample of triolr-in that produced hydropericardium in chicks, the five monkeys died within three months (Yartzoff et nl.. loc. cit.'). The ration contained 2o pr-r cent triolein. The monkeys -howrrl sign- of jaundice, pancreatic atrophy and fibro.-is. licmo-idcro. a i»~, fatty liver with necrosis, and gross hemorrhage in the intestinal (ra^t. Piss fed a rariun composed of equal parts in" a highly toxic chick ration and a normal -wine ration L'aineil weight and appeared to show tin abnormalities I'.SanjjM et o'.. l>>c, C'V.i. Jlowver, when a toxic fat was incoinorated into a .-wine ration at a level of 9 per cent, the five shoats gained 0.72 puun.N per day while the controls gained 2 pounds (I.. C. Scott, J. Am. Vet. Mcd. Amn. 137. ?',ft tlHtiO)). Despite the poor weight cain<, the one pig i-ar.-'ificed about >ix weel<> after the start of the study showed no gro.-s or microscopic lesions attributable to the ration. • The fat from the latter pig was rendered and incorporated into a chick ration at a level 'of •) per cent. None of the chicks fed tlx1' ration containing the rendered lard deTc-lopcd toxicity signs or symptoms. This finding suggested th:it the pig did not store appreciable amounts of the toxic substance in its adipose tissue. The apparent absence of tho toxic factor in pork fat is in contrast to chick-' fat, win-re storage appears to occur. The unsaponifiable fraction of carcasses of chicken-s fed the tuxic fat was very potent in producing hydropericardium in other bird* 'Friedman ct al.. J. Assn. Official .I;/.-. ("Item. 42. lin (/.0.7.9)). A more recent sluJy of tho response of different species to the chick edema factor utilized tho unsaponifiable fraction of a toxic fa I. Thi< was fed to young chicks and rats iT. C. Campbell and Ftl'.-dman. Proc. �206 Soc. E-rp. KM. Mid. 121. 1?S3 (I'.iW)}. The w e a n l i n g rut- fed the un-apomfiabic: fraction at a level of 9 /<g. per kilogram body Wcitdit per day of tin- chick edema f a n u r showed a 3.7 per rent los< in weight over the fix day feeding trial. Xo gross t i - - u c a l t e r a tions \vcrc apparent on autopsy. The Ini-rs, however, were 21 per cent heavier than those of the controls when expressed on a body weight basis, while the adrenal- wcu50 per cent heavier. The heart, kidney, and spiecii were of normal size. Chicks fed a comparable amount of coiiceinrate developed hydropcneardmni in six days. The livers of tht-e bin'..- \vi iv ]."> pc'' rent heavier than those \yi t i n - n J i . n o i - Tinstatement \\.1S made i ! , . i t " t h e iucu-a-i' 111 liver weight in the c h i c k- was not dm- lu moisture or fat." isich an u h - e r v a t i o n should have been documi nu.d a i u i cliei-Ued by liistolngical studies. A few years :\«o. \{. ]'.. H . i r n n n .m'i i o i hiborators (J. Am. (View. 5or. 82. :?')•;$ (IVGOt) reported i-ol.ition of a ciy.-t.illir.o substance f:om a iccd-gi..do taHov.. Tlncoinpound produced I i v d i o p c n c a r i i n n n in chicks when incorpoiaicd i n t o a c o n i ' r c r c i f ! ration at a level of U.I p p . n i . A report (Yartzoff ct al., loc. cit.} i n d i c a t i d t h a t the crystalline substance contain. <l 47 per ( c u t chlorine. Apparently, there are a t n i m i u - : of compounds that behave !i!.e the chick edema factor. Those compon:-d- I'luve cin-rly w i t h t h e toxic factor d u r i n g m o l i r u l a r d i s t i l i a t i o n . on thin layer ehro'n.rumraphy. and show similar peaks on ga- chromato'iraphy. Xot all these compound- are toxic, ami t o x i c i t y appears to vary i t i tho-e t h a t are. Part of thr-e di/m-uhie- may i)e n.-.-olvi-fJ w i t h chaiacten/.atitm and synthe.-:- nf i)i!t of the toxic -ubstances. By mean.-' ol -in^'L cry-tai x-ray crystallography,.!. S. (..'-1:1!^.^ X. C. \V<-1,1), and A. .!. Mabi^ rcpi>>i/-; (Al'flrncds, Am. Cnj.^lnlldui'aiihic .Us,, Merl'H'i. p. 27 (/•%'7i| t h a t t h e conipnutnl isolated in their laboratories Was i .'J.I!,7.. S.O-hexachiom-dibenx.o-p-dioxin. This work w.i- supported by infra-red, ultra-vinlft and ma.-s -pectrometry principally by .1. C. Wooti'ii ai'd \\'. V\. t'ourchene i7. .\n,\ I'nod Cim.'i. 1^, 04 (lUd.1,)). This stiiiclurn \vai verified by synthesi/.ini; the compouml. The l a t t e r pro-'.uceil the .-aim: lesions, in chicki a~- i he ; ouipound i.-olait'il iroin t t . r to\it- fat > 1 ;iinrn':iiit progre-s ha- been made. i<< i n d e n l i i ' v i n s r one uf i!ie toxic siibstitiu'cpu-om in some batcln-s of f a t ^ miended j'ur a u i n i . d foo'is. There are still a number of imanswi'U'd questions, the mo-t imporhmt of tlie-e !>-.'inj;: What is t h e pos.-ible si;j;i:-i;. caiu-i' oi' ih ( -e compound- to h u m a n health' 1 Since suli.-tam-Os w i t h jihysioki^ical pivip. ertir< s i i i i i i a r to the chick rdema fai-mr i-^a be slorcif in (he ailiposc tis-U' x of rliick 1 -, can they a!-o accumulate in human tisme when minimal amounts arc inadvertently ins;e-te<l? Ano;liei' ini|)ortant question is: \Vliaf i* the source of lliesO compound*? Apparently, the chick edema factor has been a.--ociateil only with fa!= and f a t t y acids subjected tci a c«ii.-ide.":iblt' amonnt of licat (hiriuu: tlirjr pioces-iiiL:. If the source of the chlorine in t h e toxic compound conld bo identified, it mijtht be po-sible to devi.-i inelliii'!- for IK icmoval i)!' elimination before marketing the fat. 207 Reprinted from Journal oj the Association o/ Ofii'-ial Affriculti-.ra! February, l!>i:» Chemists Studies of the Chicken Edema Disease Factor* By L. FRIEDMAN, IX HKESTONE, W. HOIIAVITZ D IHNES M G. SHUE (1'ood and Drug Administration, Washington 25, D.C.j ' ' , *»d Introduction we successfully produced the cha:.>cteristic When the pioblcin of "X" or "Edema" symptom* of the "chide edema d:.-:e:;^e" inin poultry was brought to our attention in volved a sample of fat that h.:d been colDecember of 3!'5r, a considerable amount lected by one of our Food and JJiug inof work had already been done. It was es- spectors al a feed inaiiiifai'tmiii:: p'ant and tablished that dnijts added to (he feed, as which had been knov.n to ha\-.* pro>'i;ced well as ciuit.'iinin'-linn by drugs or heavy the- ('-'sc-ase, and :d-n a sample of a "larry ' metals such as lead or arsenic, Vas not re- bv-pn>d;:f( from the mrmuf.'icton: of oleic sponsible. The disease was not. caused by and .-'ti'aiic r.-ciil, v.-hich presumably h.i'l ben; bacterial, viral, or parasitic infection, and mixed into this fat .-vrnpi" to -\;i: extci.t of evidence was accumulating thai the oi.iy feed about id'i. In tliis experiment *!;<• v;'::m;r. ingredient associated with all t h e various B hypoiiie-is was tc-ie.i. Al.-o, il,u .•:;il::b:!outbreaks was the fat. The tunptom., have ity of the A.O.A.C. ration for thi- ijive-ti::;!been describ.-cl in detail in the preliminarv tion ««.-• checked ap'iji-t a fee'- ic-'I^'I!;;IK report of Schmitlie, d al. (J. An Y<( Mat more clo-cly a iir:iciic,-J conimcr--!. ! :..•:•';:. Assoc., 132, 21(> (I'.v'S)), which al-o sum!',,t*nl Jlalion and ilic Yilam'n :'. llii/H'l!:marized some of the work leadirjL- to the «.•;>.—Table 1 shows thr coinjvi-:!i.'in «•' the incrimiuation of the fat added to (he ration. A.O.A.C:. basal ration and the --;;..cir;I bav.il The characteristic symptoms of this dis- ration. The major (Hfi'erencr? ;-rc '.!.;• ,-ubease are the presence of excessive fluid m stitiition c>f Uraclceir (an if''.,'a-} 'oybivui the pericardia! sac, in (he abdominal cavity jirolcin) for the c::=ein of the A/'.A.O. die;, (water belly), and less often snbciitancoash", and ihe incl'.ision of alfalfa Jr-rif m r a i and accompanied by high mortality bccinnir.w lin.-ced oil meals. T'lom lla; re •.;!(.? shown 3ppro:;im.if<>!y in the fhird wr-efc. The strilr- in T;ible 2, it is clear that: (/) <he fatiy ins roscmblanco of tiie "cdcm-i disease" aciJ (K.A.) by-jii'odiict is much ;i;oic ••fj'ot;symptoms to those recoided for the exuda- tive in producing tho disea.-e th:ni the ^.•iIlItive diathesis syndrome of vitamin 10 de- ple of fa;. (IXV. Fat) that h.id jiffu.-Jiy been ficiency in chicks nr.de attractive the used by feed manufacturers; (2> the ^i-'-cial hypothesis that the di?e.ise outbre..,!:s were l\is:d ration is to bo preferred to the the result of an induced vitamin K deficiency A.O.A.C. die', .-i'lce. the .-"verity ••:" the (de.'ji.n. ?ymptoa>s is irpica.'ed, growt.'i . --rfuraiance caused by the Use of poor quality fat;. i-- improved, arid "iz/ird cro-i-vi i: eliniiIt was coi;vcma3l, Iicrau.-e of oilier work nated; (!!) the fecdim-; of vit.nnh: I; ;n J.Ttc in progress in our I.iboraforv at the time, to use the A.O.A.C. vitamin D, chic!: test 'raTahlcl. Basal cliick n,i;.,, l v tion and fiiiRle comb white 'Ujrhorr. chic1:? .<•(.,-;..| in an attemj.t to pn,.-.!m- the fMnrniou. Our first fcfdm s trial with a f anple of feed coller-lcd from one of the "d;,e:,.-e ar-as" v.vus a false start, ^!K:5 n f ( c r f ; x ,, f ,. k , ()n the ration il,o «':ie.',.-j were normal .•-•,<• showed no unusual symptoms i;poa p:»sim»ri,-pi ey.iniimlion. The first experiment ;„' ^j,;^, III 2.M x . : Ko.,.irr,,.i...tj yo,,t Mi-Si' • ',!' i: 2.01 :i.2 �209 208 FATTY BY-PRODUCT UOOOgrrO p i* Saponification Tab).r 2. K|}r<:l of .si/.spect fut and lij'-prodi:ct fat on chicksAv . Cio'ip 1 2 3 4 5 A.Ucrf Fat ]>,et A.O.A.r. A.O.A-C. Special Spc-ci.'i 1 lard 1XV. f:.t G% 1NV. fat •'3.6% lard -j\2.4% K.A. bv-prutiuct I '6 Qf/c hiTil +" <2.4% F.A. by-product 4; vitamin L * (i'/f. l.'0-D^y i «' 147 1-3'i Drail. No. of Chirk* W!.i»iliK ICdpnn:l Syiupto:ns .Mi!.! No. Mark ...1 Xonital — 10 7 2 20 12 4 (() b 7 9 — 0 5 5 9 ir>;t 0 0 0 M7 143 — 5+ . v:.l l;mc, 29.2 da>«.. doses docs not prevent the occurrence of the disease, but docs seem to decrease tbe severity of the symptoms. However, in an mdepcnd' ••; study by P. C. Underwood and C. G. Durbin at the licltsville Laboratories of our Veterinary Medical Branch, with lihoile Island Reds, the oral administration of I ing per day per chick of rf'-alplw tocopheiol acct.'ite h.id no efi'oet, the symptoms being slightly more pronounced in the suppJcmc-ntcKl group. The conclusion that sonic material associated with the ''fatty acid by-product' 1 is responsible for the di.— ease, and not n simple \ibrnin E deficiency, was confirmed in a subsequent test, when the fatty by-product was fed at '% of the diet and all chicks showed marked to severe symptoms at autopsy; only 2 of 19 chickssurvived tli? twenty-third d:iy, and the average survival of 57 chicks was 19.4 days. In a similar group fed 10 mg of <K-alpha tocophero! pei chick per day, the average survival W.MO 20.5 day.- for 15 chicks; 4 chicks survived to the fv.enty-third day. All showed marked to s-tvere symptoms. Chcmir.a! Findings.—Part, of the chemical data on thr-.=e samples is shown in Tab's 3. Findings of particular significance were that the=-e "fats" (F.A. by-product) were actually TaMr 3. Comparison of * ivith larj fat^ I'nj-ap. 0.35 Lard Inv. fat 3-b 10. r ]-'. A. i.v- ' ' "' l'0-lli''lj J.j.,.,. ,.,;,,,[ !. I fj of Vn.-:>|.. ',r Of t II- • [ ' • 23.0 70.0 21.7 yi.'.i I 1 UNSAP 10.7% !07gm. 7.0 2.S free laity arids containing an unusually large amount of unsaponiiiaUe material. Furthermore, the un.saponifcible fraction is qualitatively different from that found in frc-h animal fat such as hr*l, by virtue of the large percentage that reacts in the Liebrrmanu-Ilurchard t^-st, compared to the much smaller proportion that reacts wiih • li^itonin. In the lard the proportion is almost one to one. Urea Fractionalion.--r£lie f;*tty acids from the fatty by-product, after saponification, were fj'acfio;ia(ed (Fig. J) \\iih ii'-ca into tv.-o portions: the normally Recurring fatly acicii that form urea adchu-t?, and iriodifici! c>r abnormal fatty acids th..l do not form urea adduets. About 70% of the fatty acid-: form urea adduets; the rra.iinder C-4.%) comprise ihe urea filtrate. Effect, of l;>ra Fraction* en Rats anil Chic!;s.—The urea filtrate frtty acid fraction was fatal to weanling rats (40-GO g) njion oral administration. Tv.o successive daily do.-es of 0.4 nil resiilti.i! in marked loss of weight and death by tho, fourth day. Two doses of 0.2 ml caused a marked v.-cight loss from which rcco\cry be?an at the fourth day. This effect is similar to that observed in our laboratory w i t h the i.i'ty acids thai do not form urea addur.ts v.-'uich were derived from hr.tU'd colion-ecc' oil, and to thn observations recorded by ^rnraptun. et a!. (J. Xiilritk'ii. 44, J77 ( l i V i i ) ) with lieatPil linseed oil. Cliicks react sii^ilariy. althousli they re'-m to be more rcsi.*(:::it to the lelhai effects of direct, feeding thsn (he r.-its. ]fowfver, ivil rn<ai!:h oKservatip.; - were made (o e-tr>blish the relative MI«C- ; libility of Hie two speri'.'.-. AH polymers .•••rid hydrojremttion jiiodu'-ls (too Fi;. J ) v.\"!t- lu.\ic. FATTY Ci/ACIDS 8 O /o O 830 gm. Urea Fractionafion ADDUCT 76% (669am.) M.W.Me. Esters: 288 VOLATILE, !00°C 1.3% I3grn. NON -ADDUCT 24% (p.llgm) MAV. Me. Esters: 323 Molecular Distillation l30°C.,7;j POLYMERS 1% (2gm.) MONOMERS "1 99% ||09grn) Fig. 1—Fruciionatum of jatly odd by-product. (Kumbefs fallowed by + siyn nrc tivicily scores; see footnote c, Table 5.) The ure.i adduet. fraction of the fatty acid.-; of the liver markedly iiicreawM', v.-hich was produced no unusual effects either in ra'.~ or accounted for only in part by HI . increase in in chicks. liver fat. No other significant ii iier^nees in the relative weights of org.-Ui.-; (kidneys, Kflccl of linsajionlfiable Fraction on Rats. thyroids, hearLs, adrenals, etc.) vrcre noted, —The significant findings of ferdinj the unalthough puwll differences cn!:no', be Uemon? -iIxinifi:ib!« fraction of the fatty acid bystrated by suf!) .-maTi groujw. product (o rats arc f-wm in Table 4. One Although deleterious effects v>- ri: produced ml of the. iiR.--':!ponifiubIc fr.-.ction was fed to each of four rats every d:iy for 22 day?. in rats both by the- ;iroa filimt. r>nd by the Control animals received no supplement to unsnjmnifinblc; jiorlion, Eyinpiosj:.- siinihr to ihe basal ration. The experimental group those of the "I'deina di.sea»'.-" sy.'hriine were ,untamed two mates from different litters, not ok-erved in this specie-. \vllh liiler mate".' in (lie conirol t'l'duj). Th'.Kijccl v,i C/»'eto.—When a-JJ.- \ to ihe diet frniales of Ivjih the. tpf-t nnil control gioups of chicks at a level eijuival-. :ii t. • 7^ of fhe \\VTC fill from one littei. TJ:e !T7-ow(!i rate origin:i! fatly by-produc'., the. •••rci illnaie v.as s'p.iiii'.'f.iiilly dejiu-s.-cd—females wi-ie jiroduc'-d a significant grov.!h d.-, vc-.^inn l>ul, more draMic.iIly afi'ccfcd than inalc;; t!:e. as Tabln 5 shuv.1?, very fe\v PI' pp.a dise.^o -\m of '.he thyimis dtcroar-td: and the- size i". 'IT.e Miwll anwmt of a c t i v i t y �210 211 Table 4. Kcsults of feeding uiisnponitiahlr-s from "fatty aoiil l.\ -^iroil to ^vcanlin^ ruts l»y t-toinach tuln: (1.0 ial/rat/<i:ty> 1 Cll'!iLlOV Cm. trol 22-day wt Klin, K rnnja Liver wt, g/100 <n g body wt K-it in dry live r T!l y m u s , Vcd 13S 1)2 0 .00 5.77 12 .0 0 220 0 .25S i ioo g body wt j ill in d. Go 10 .03 11 .8 0..005 0. 1H2 C'iml!.. 114 nr. 5.00 o.:'»i 10. y n.2;u 0.10 " Table 5. Effect of by-product fnt and its fractious on chicks IV. i 49 48 10. on 9. t;3 14. !) 0. OS-1 0.073 ' M;'Vl i .::m-d l.v li't'-r: •-' li;>,.,,. " K-mu!. •! a l l f n > . . i .in..' I.'.:.! E\|)t. 2 2 2 2 2 4 4 4 4 Sample % in Diet F.A. by-product I F.A. by-product -f b vitamin E 7 Urea adduct 4.8 Urea filtrate 15 Unsap. 0.70 Unsap. 0.50 Unsnp. 0.25 Unsap. 0.125 Unsap. O.OC2 Deal!,* 17(19.4) 15 (20.5) 100 50 10 3 s 1.0 0.5 0.1 10 i-3 ?-Q 25 FF:V:/!!CW NO. 30 2 2 — 15 (20.2) 3 (20.3) r> c -N'o. N(jrinal Tox. Score' 4.9 2 — — 5 — indicated in the urea iiltratc was shown in later experiment.; to reside in the unsaponifiuhle re.-idue thru had nol. been complexly ext.r.iclod. It is clear that the edema d t r ease-producin;; material i.- associated with (he un-.ip(ii' ; fiab!e portion of the fatty byproduct. In Table o :m: ali-,0 .~ho\vn the variations in ^verity of the dioease produced by a scrie? of graded do-e--. Although such a tabuI.i'KMi a= ihov.ii l,vre cannot give a co:iip!e;c picture ff the r:inje. of observations encountered, it indicate- how the pattern of symptoms seen iu any one feroup depends upon the dose of active material fed. In these studies the dose was usually high enoi:i'.'i to produce were symptoms, tnd as we became more f".n:ihir with the di.-iase we localise more confident of the result.- obtainable u i i h smaller jucmps. Whenever possible \ve now use five chicks per poup; No. Chick* fslinw-;,,.. Kdc rna Sj'nij'toims .Modulate Mild severe 4 J 3 2 20 14 3 10 4.7 0.2 4.7 4.0 2.4 2.0 ~— occasionally we have u.;e.- as few as two, when the amount.of samj..'"-.- was limited. Concentration of Tc-- -:;c Factor Alumina Chromalofiripti v-.—The first step in our aUeuipts to isoiau the factor lh.it caused the "edema di.-c.:-.- from the uns.nponifiable jionion of tii'j n t y by-produc' \v;:s chromatography on r." r.iina (1:0;. A chroi:i.i!o;;rarn (I-'ij:. 2) •• :-' obtained by plotting the \vcisht of in "riiil cluted bv increments of FoKvnt. 1'e:; •leum elhcr \vas H=ed i!)iiil thi) cluiioJi curve 1 f-c-emed to Jwe readied a mini'inun, then •"."s followed by ?o^o fthyl t'.her in poiro; -.•in ether. The eiuiion was complelrd \:,:': 100% ethyl ether tu cive three diitiir • fractions of inCK:.-ii)!r jio!-irily. In ™.^r • ..d, Frtc.tion I Cfi-jld be ch:i!~icleri/ed a- h; 'rorarbon (IIC). Fraction ll :>s coutaii!!!;^. larbonyl compound?, and Fraction III n> fontniniiig s!frols. Tho ij] tt a violet ab=orj,> •-•:i sj'ectra show that Fr. cticn I lias characti '..stic absorptioii peaks at 223, 227, 234, and J-Io m/.i, sugge?i. ing the pretence of E-isnifi. :iit amounts c: diolcitnuiene, either the 2,-t ar 3,5 isomcr or both. In anlk-ipjlion of the pr. -'•bility that the toxicity v.onld ic.-ide in i ijction I ($3%, of the i!ii-?a;)0itifiable niatr: 1 ' d) a famp'e of 3,;>-eUo!o-i:idie;:e v/as j'-rc; red by >V. ].. Hail in tinie to be incli;.' • ! in the third feeding trial. The re.-nlt-, i. ; tin-: experiment indicated c'f.irly that l l i r i i - wa- i!o resj>o!isc K) rhnKM.idiena Init t l u t .'11 the activity n.-spoiij:!i'f- for the: chick i • 1.1 a dkcasc w,i< i:t Fraction H, which u>,. ''.".tod appui\iiiiriti-'.y t;'','. of the, un>-i,}o •'iablt 1 or alxnit ().S-*)''.- of the original prn n; ;.—Three difiVi-- eat approaches \»crc made simultaneously to the problem of further purifying Fraction II: (/) counter current distribution (iro-octane:' mctluniol solvent system with 500 transfers 1 ), (g) chemical separation of carbonyl containing compounds by the Girard Reagent T, and (S) re-cnromato?raphy on alumina with a higher ratio of adsorbent to sample and more gradual dution. For the counter-current distribution experiment S.4 grams of Fraction II were loaded into the first 10 tubes of a 500 tube Craig all-glass counter-current apparatus. After 500 transfers, the contents of every 10 consecutive tubes were combined, and after evaporation of the solvents the weight of solids was determined in each of the 50 frictions obtained. The resulting distribution curve, i.e, solute vs. tube numbers, is slinn-n in Fig. 3. potent, showing as much activity at a level of 4.2 mg per 100 grams of feed as had been observed with 5 grams of original product. It is of interest that the toxic fractions 5 and G did not react with digitonin and did not show the L.B. reaction. In preparing material for the chemical separation (Fig. 5) Fraction II was cut into two approximately equal portions; a characteristic yellow band was used as the landmark for the separation. Each of the fractions (IIA p.nd IIB) was treated identically. First, after evaporation of the ether the material v.-as heated with methanol, (hen cooled. When the hot meth.'nol solution cooled, white crystals were obtained. The infrared spectrum for these crystals indicated a cnrbonyl compound with a long aliph :lic chain. A synthetic compound, dipalmilone (prepared by Jonas Carol), had a very similar infrared spectrum. The 50 fractions were combined accord:::g to the peaks indicated by the distribuApproximately 16% of Fraction IIA was te, curve into 10 fractions f or biological :•--'. in; (Fig. 3). Kach fraction was added insoluble in mefhanol. To the methanol••> the test diet at a level equivalent to 1% soluble portion was added twice the weight :;. the original fatty by-product. Figure'4 of Girard Reagent T and to the mixture =::ows the separation procedure diagrammati- was added }Q% its volume of g.'ari.il acetic .:-!iy, and indicates the amount of material acid. The mixture was allowed to stand for r.revered in each fraction. The activity two hours at room temperature before the • -.-tfrf to Iw equally distributed between water-soluble derivatives were separated i ' l ' - f i o n 5 (tubes 2-10-320 containing 2.23'J from the unreacted material ("non-l;etonic."). .. .:n-) and Fraction G (tubes 32I-3-i3 P0n- The "non-kcfo;iic" portion was dissolved in • ;.;.'i? 0.000 grams), as indicated by the hot acetic acid and heated with tm times • x ' i f y score?. Ft.ieiion f. v.-as tl.u'mosf its weight of Girard Hwswil T on the .-team bath for 10 minute, and the unreackd ma'f'fitlly a'-li'iotvl.----!!;!. onr di-lit Jo 3'rorMr (eri:d was separated from the ws!pr-so!nWe. rkTS for i-Ui-:;'",-f:ii£ the i-ol-.-rrf 5ys-f»m. derivative-. Approximately 23% of sample �212 213 UNSAP !0.7% I07gm. AUO-z Chromatography (l!6) in Ha 4.2gm. 200 500 400 300 Tube No. Fiff. S—Counter-current partition o] Fraction 11. Methyl alcohol:iso-octane eystem, 500 tubes, &2 hours. lO.Sgm. MeOH Insoluble 0.67gm. MeOH Soluble 3.53gm. Girord-T Reagent (I.-2) FATTY BY-PRODUCT lOOOgm. Saponification HOT KETONIC 0.28 gm. (UN) NON KETONIC 2.44gm. 4.4* Fig. S—Further froctionation of Frncl:rni II. 1 21 1 .03 COUNTERCURRENT DISTRIBUTION 500 Tubes MeOH ~ Iso octane 25 .26 4 5 .49 2.57 O.I* 3.9* 6 .69 3.8* \ 7 8 .55 1.13 1 1 9 10 1.97 -82 gm. 0.1 + Fig. .',—Pmlilinn oj Fraction II with methyl alcohol :uo-nctnne counlcr-cnr.-cn'. fjstcm. (h\mbers j'jll'wd by + aljn are toxicity scores; sec joo'notc c, Table o.) lutJ icacictl iu the cold and S% more with heat; 69% did not react under these conditions and was classified as "non-ketonic" (UN). The various fractions derived from r'raction 1IA in Fig. 5, as well as a similar jet derived from Fraction 1IB, were also fed at a 7% equivalent level in the diel. The -ynthetic: diji.ilmitone was fed at .1 level of in ins per 300 grams. As indicated, there v:ni no tli.-T3>e-produeing activity in any of the fractions obtained from III3 and only the "non-kclonic" Fraction UN was very aclivc. ?nrpriflngly, however, the infrared spectrum of Friction HN definitely indicated (he • •nvi-ncc of r/ii-honyl groups. This eo-eatted ••non-kctoisic" materi;!l v,-;is again treated with Girard Reagent T, this time with fivr •.imcs 1hc weight of re.igent instead of two •i:nes, and was heated at SO0 C fur ).ri min- ute?. After the water-solubl'1 derivative? were separated from the unrcactcd nonkefonic material, two-thirds of the 1IX fraction proved to have reacted. The nonfcetonic residue was again treated with excess Girard reagent and heat, and this time 50fJ of the "non-kctonic" mate-rial reacted. When the original material from the Girard derivatives was regenerated and the reaction was repeated with excess Girard reagent, again only 67% reacted to form hetoneGirard derivatives. It was c'car that our active "non-kctonic" Fraction UN consisted largely of compounds contaiuin; "hindered" carbon}-! groups- that would react with Girard reagent v.-ith great reluctance. Of additional interest If the fact that tluxe "hindered kc-tonp" fractions had characteristic ultraviolet spfdra wiih a major ali--orption jn.-ix- �214 215 UN NON KETONiC 2.44 gm. 4.4+ 1 Iso octane 0.41 gm. 2.2 + (3mc.% i 80% EiOH 0.22gm. AUO^ Chromatography (1'.40) 1 2% Ether in 2% Ether in Iso octane Iso octane l.!2gm. 0.63gm. I *.0 + Partition Chrcmatoqraphy {3 mj.%) Siiane-Treated Cefite 80% EtOH 0.24 gm. 90% EtOH 0.47 gm. 1 4% Ether 0.27 gm. CHC13 0.19 gm. 4.2* Fig. 0—Cliro?natoyfiiphy oj Fraction UN. (Ttixicity fcoics at the levels fed appear below boxes.) imum at 233 m/ t acd a seconds iy maximum Xcxt, this "phenaiithren'.-." material was at- 200 ny. further concentrated by partition chroinaThe disease-producing activity of Fraction tography on a column consisting of 5 grams UN was further concentrated (Fig. 6) by of silane-treated Celite plus 4 m! of isochromaiography, first on alumina. A ratio octane. Mobile solvents vere 80% ethyl of one part simple to 40 parts alumina was alcohol saturated with iso-octane, followed used, and the column was ehited successively by f'0% ethyl alcohol sat'.'Kitcd with isowith iso-ociane, 1% ethyl ether in i.-o-octane, octane. The first and second 80% alcohol and 4% ethyl ether iii iso-octane. The frac- fractions were both very toxic at a level of tions wore monitored by ultraviolet spoe- 2 mg per 100 grams of feed. These livo trophotometry as tbi-y were eluted and then steps (Fig. 0) represent approximately an were combined for feeding on ihe basis of S-fold concentration of the toxic material spectral characteristics. The first part of of Fraction 11X. The first S0% alcohol cut the material eluted with 2% ethyl ether in had a typical phenanthrene ultraviolet speciso-octa.':e had a definite phenanthreue type trum and no peak at 23i ni^, whereas the of ultraviolet spectrum with characteristic second S0% alcohol fraction shows a major maxima at 259 m/i and 300 m,u, and with no absorption m.ixinium at 234 m/i, in addition peak at 231 m/ t . The fraction immediately to the characteristic phenantinene peaks a: preceding, eluled with 1007o iso-ot-tane, did 259 and 300 m^. not show a phonanthrene type of si)ccirmn but did hnvo a i>:aj.»r absorption ;v-;!k at Simultaneously with the work just «i<. 234 m :l r,r.d a Mivl'.-'r one at 200 r.u.. The cuisou, another st-i-iy of cbx-.^toav.'.plvy i™ and (hat only the first, cut contained edema disras-c-prodiiciii;* material (Fig. 2). This their ultraviolet spectra. Fraction S2A consisted mainly of material with a phenanFraction HA was elutn! from the second thrcnc spccfrum, and 821'. showed the major column with large volumes of petroleum poak at 23-1 m//.. Fraction S2A was reether (three arbitrary mis), followed in succhroraatojiraphcd on the riiane column, using cession by largo volumes of 0.2%, 1% •>•:•„ 75% ethanol as the mobile solvent, and again and 5% and 25% ethyl ether in'ix:trol«mi' ether. The seven fractions indicated in Fig 7 cut according to the ultraviolet spectra into •.vere tested at a level of 3 m? per 100 grams a forerun, a fraction with a phenanthrene of feed. Only Fraction-; 2 and 3 showed spectrum, an overlap fraction showing approximately equal absorption peaks at 2;"<0 activity. The methyl al.-ohol-in.,ohible porand 23-t m/t, and a fraction with a major tion from Fraction 3 showed activity. This peak at 23-1- m//., approximately 0 times as finding did not sgrec with earlier experience. intense ns at 259 m/i. All of (he fractions hut ni.-iy be explained by (he brier quantities were fed, and each of the major fractions of material involved in these particular opcrl was found capable of producing (he chick siions as compared to (he cm her expeiiment Apnarrady the active maieri.il has « more edema disease when fed at a level of 0.5 nig limited solubility in mefbanol than first sup- per 100 ;:ram« of diet with a severity apposed. In one of our most recent experi- proximately equ.il to that ob-:rvcd in chicles fed 5% of original fatty by-product. This ments, the mcthanol-soliible portion of Fracis the first instance where fractions with tion 3 from the second alumina column was distinctly different absorption characteristics subjected to partition chromatography on had apjiruxiniafeiy the samn biological acsilane-treated Celite as previously described tivity. (Fig. S). This time only S0% othanol was Another recent experiment involves the used as the mobile phase and the eluate was partitioned into forerun, tailings, and two reduction of one of our potent fractions with major fractions, 82A and S2B, according to fodium borohydride. After ih-3 borohydride was separated, the treated material was Chromctogrophy 4.4 + 3.0 rig % 3.0 mo % fiy. 7-~ Chromatography nnd cltilinn of Fraction IIA (1:100 column). �216 217 MeOH Soluble 0.220 gm. Parlilion Chromalogrcphy on S i l a n e - t r e a t e d Ceiite ( 8 0 % E10H) 4.4* Tailings 2.0mg%[ I47mg. Partition Chromotogrqohy on Siiane-trected C'elire (75% E t O H ) 9 ma. .2.9* 1.2 + O.25mg% 0.5mg% /'"/. c?—Hcversc piia?e jiirtition c!"'<>i'"it.uQrap iiy tit VIOLET BLUEWHITE WHITE PALE BLUE BLUEWHITE BLUE WHITE 340 300 260 • chrumatogTttp\y. nr.d •nHf orsci'Kt fpots oblaiitt'l i McOH-fi'iliible (U.V) 234m;j. if Fraction i'. curbonyJ group present, ir.uicsting that borohydriiie al-o reacts reluctantly and incompletely with the hinderc. i kctone?, and the third fraction contained hydroxyl gioups. 'Flu-re was very Jittic d;:---.isc-;>roc!i!cii)g activity left, all of it. limited to the- <irst fraction. This may mean th.\" P.:'.her the activity is ]o.-t upon reduction '"":~ that mil all of the material reacts under i;.? conditions u.«ri, or that the. reduced co:u>lund retains only ].-irt of the original :;ctiv;-y. In either c.ve. llK'-c data arc addition;)' evidenri- tint tin active material contain- a oarbonyl compound. Karli of the fraction: • .ii-eu.-.-cd hero, and many others have beca. studied by ultraviolet and infr.ircd ?pc.-.J.rophotomctry and the peitinent ob^'rvaiir-'.s have been mentioned. K. O. H-ienni has beer- ?tmiyi:iR she (luon'-rence sjiec-tr.-., both a tivation and tmi,-. -ion, of each fraction, -:;•! will rn'on ili» re-nlt^ when the d;ita h-'i .0 been analysed. In ::di!:iion, as tlie jio r '.-:icy of the disea.~e 220 irndueii'K lac'or.s has be--si conceutratefl, w li:ive tested the p'.irity •: f each fiaction Iv )j:i)" - r rhrmnatoi;rap;iy. \Yc are ti-ins; Sx chrumatogjaphf'd on alumina into thi^-- ;:plirpximatcly equal (30:30:-40) fractions 01 increasing polarity. The infrared spcctr.i indicated that the first fraction v.as brsely hydrocarljon, the second fraction Ire! the Sample ASa3c O.Srng P = Phenanthrene SpectrjA= Absorption Maximum ct •J'/a Whatman No. 31 paper, washed with mefhanol and then coated with mineral oil dial has been pre-washed with sulfuric acid and mclhaiio!. The papers are developed with incthanol iu an ascending system for periods varying from 2 to 20 hours. For the longer periods the solvent is permitted to evaporate from the edge of the paper through the slotted glass cover. After the papers are dried, they are examined under ultraviolet iifdit and the fluorescent spots are traced. Figure 9 shows a typical paper chromatcgrarn. The spots ;,re out out and duted with methanoi, and the ultraviolet spectra of these solutions are determined. Figure 0 represents a fiaction of high purity that has been tested by a chick feeding (•\-periment; it is of considerable interest to note that 1'ie ultraviolet spectra of these spots are of both types discussed earlier, namely, the phenanthrenc type with an absorption maximum at 259 ny and the type with a maximum at 23-1 m^. The ultraviolet curve of the original solution was of the phenanthrone type. itble material from the toxic fat and fatty by-product, we compared the vmsaponifiable fraction of the carcasses of chickens that had been fed these products with the unsiponifiable fraction from the carcasses of normal control chickens. About 25% more imaiponifiable material per unit of fresh chicken weight was obtained from the test chickens than from the control group. When the unsaponifiable material was •chromatographed on alumina, about 18 times more hydrocarbon material was obtained from the test group. The test group hydrocarbon fraction gave a strong positive reaction in the Liebermann-Burchard Test whereas comparable material from the control group v.'f.s negative. Furthermore, upon comparing the hydrocarbon fractions by differential ultraviolet spectrophotometry, definite absorption peaks appeared which were similar to tho?c observed from Fraction I of the ursaponifiab'e portion of ihc fatty by-product. This evidence indicated beyond doubt that some of the material from the diseaseproducing fat was deposited in the flesh of If we were to speculate about, the chemical chickens when they received it in their diet. nature of the disease-producing material This, of course, did not prove that the from the evidence available at this pcint, chicken meat contained the toxic material, n-e would visualize a pair or a series of especially after it was learned that the closely related compounds, with a 2-rinjr disease-producing activity was in Fraction II naphthalene structure, or more likely a 3- and not in Fraction I. In order to settle rins nucleus, similar to ]>lienanthrcne; n this point, the unsaponifiable portion obcarbonyl group in a long aliphatic side chain; tained from freshly ground chicken carand molecular weights of 500-000. casses, not including intestines, head, or feet, Appearance of Poison iu Flesh of Chickens was fed to chicks at levels of 1.0, 0.5, 0.25, 0.125, 0.05, and 0.025% in the te=t ration. Once it had been established that the As can be seen in 'fable 6, symptoms of chicken edema disease was caused by a toxic edema-disease were observed at all levels fed. principle, the possibility that edible chicken The unsaponifiable material from normal flesh might contain this toxic material had control chickens fed at 0.5% of the diet io be considered. As soon as we discovered produced no abnormal symptom? of any the unusual ch.-.n-.ctcristics- of the unsanonifi- kind. Table 6. Kcsitlls of fetf'dinp inisap^>nitial>le extract of chicltens that had Jjc«'ii fr<l fal or fatty acid hy-pi-oiluct i:»iit. X" 3 4 o n '; in IXc-t 1.0 0.5r ().2. , 0.12.T 0.050 0.02;. IVatli2 (15.5V 5 (10.6^ 5 C-19.0) 4 (21 .0) No. O.'icl;* Showjnc Kdr-ii.a Svmptoins fcvorr Mixlento Mild — — — j • Fi^ury? in -psr'-n'l.fM-s *!n>w avfi.i^<» Aiiri-j'A-d lime ttt t\-.y* „ .\ 1 Xo, Normal 2 1 ' 1 3 �218 We have examined samples of various type; of fats and fatty acid products in a survey that is. still coiitiiiuiiiR. The edema disea^c-producinc product is a characteristic not confined to the production of a single manufacturer. In summary, from the studies so far, it may be concluded that: (1) The chicken edema disease is c.nucd by a toxic factor in a fatty by-product of Elearic and oleic acid manufacturing operations. (•J) The disea-e-producing factor lias been concentrated approximately 10,000 times by sapohification, chromatogi-.'i phy of the unsapomfiable porliou on alumina, and n verse pha=c partition chrom-jlography on sil.:!Ktreated Celite. (3) The evidence pi canted indicate, (a) the presence of compounds with a two-ring nucleus .similar to naphthalene or a threering nucleus similar to phenanthrene, (b) the prc.-nu-e. of a "hindered carbonyl" croup, pinbably in n ride chain. Based on the i<-!al i \-rly low i n t r i i - i l y of the rarbc.n-,1 p> k at 5.8 /i. a compound of inolrruhr n c i u h i 500-000 i.- su«-e.-lcd. However, the cvidi-i.rf on the presence of carbonyl in tlie toxic compounds is not conclusive and does no'. completely uile out the possibility of a toxic hydrocarbon. (-!,! The fatty acid fraction from this fatty by-product contains significant quantities of ron-urea adduct-forming compounds that are toxic, although they do not produce edema di.-ea.-~c. (5) The edema disease-producing compounds are present in significant amount-? in the fie-h of chickens on rations v.-liich contain tcxif material. (fi) The cdi-ma disease-producing material is not confined to the product ion of a single manufacturer. Note 219 THE OCCURRENCE OF THE CHICK PEBICABDIAI. EDEMA FACTOR i>- SOME OLEIC ACIDS AXD PRODUCTS DERIVED THEREFROM 1 distinctive fractions. One of these fractions, pos-csiing the ultraviolet abcorption spectrum of s-iib-iiit'itril iiiii>hlliali-nr~> (X,,.,.. at 2;iO, 2SC, 200 iii/i; j~lu~-.ilili.~r at o2:; ni/;~, Xn.in. at 2,13 m-i; ,.1;30 — 14 x Asa), proved highly toxic. Tlie inf:v.icd p'-rctrum of (his n-at'.-ml showed little absorption in. the carhonyl .'ojdon. It is concluded that the "n-vphlhali'-ie" fraction compsises a mixture of hydioc :rboas. Assuming t!i.d alkyl-yubJluulc-l u::i'lit!.::!f!je is the only chromophore pre-eni, tin: average molecular weight, calculated from tht; Bi.-sorptivity at 23G m-i (F, I'/', 1 cm S: 3200) i-; iu the vicinity of 250. Stanley E. Ames, William J. Swanson, Marion I. Ludwig, and George Y. Brokaw, Research Laboratories, Distillation Products Industries, Division of Eastman Kodak Company, Rochester, New York A material causing the chick edema syndrome has been reported (1, 2, 3, 4) to occur in specific lots of feed-grade animal fats. These fats were reported to contain the edema-producing factor as a trace impurity produced during certain fat-processing operations (2, 3). Birds fed diets containing this unidentified factor develop an edematous condition, characterized by pericardial edema idistention of the pericardial sac with fluid, also termed "hydropericardium") and in more severe cases ascites and gross liver and kidney damage. In the young chicks the initial gross symptoms are characterized by abdominal distention and labored breathing. Manifestations of the pericardial edema factor appear to be limited to poultry (1). The active perieardial edema-producing factor has not .vet been isolated, but certain characteristics of the pericardial edema factor have been reported by Brew ct al. (2) as follows: a.) presumably a hydrocarbon derivative, possibly of cholesterol, b) a molecular weight of about 360, c) associated with fractions which give the Liebermann-Burchard test for sterol residues, d) it can be concentrated by molecular distillation of fats containing the factor. The edema syndrome is not produced by fat per sc but appears to be related to impurities in certain lots of fats subjected to special fat-processing operations. F. W. Hill of Cornell University rei>orted to us that high-level feeding to chicks of a sample of a glycerol ester of oleic acid resulted in some mortalit.v with gross symptoms resembling pericardial edema. Our studies have confirmed his observations and furthermore have shown that the pericardial edemaproducing factor is present in many samples of commercially-available oleic acid as well as in glyeerol esters made therefrom. Molecular distillation of an active oloic acid or of a jrlycerol mono-ester of an active oleic acid was found to concentrate the pericardial edema-producing factor. Ack-nowlcclsu-ci-t The exteiiaive work ou'JiiM'-l in this paper was made possible by the active cooperation of many oilier individual-, both within and outside tlie Food and l-'n'.c; Administration. We are indebted to W. C. Ault and I). II. Paundeis of the Eastern Ut's'i/ation He-search and Devcliipinent Divi-jiei! of the TJ.S. De]>artmcnt of Agriculture, v. ho jirepared 1,330 •cram.-- of unapoiiifiablr i,--..ilerial from tho fatty by-product, and to th-r.-c members of ihc Food and Drug Adnii;.:-ira1ion: Stanley X-.'-hcini, Divi.-ion of I-W!, v.!io jirepaK/d an I ' l j i r i l amount of (lie ur--'i{>(>nifial>lo in.-i11 r i a l ; . I I - H M I K - Mi-nur, l")i^ i. MI of Koud, \\lio helped ]u<'p:ire the charts; ,1. II. Jones, John Winnii;'.'!-, and Meyer Poii-'.-'-.y, Division of Cosmetic";, who assisted i". the couiilercurrent di.-iribution .study; Jonas Carol, Division of Pharmaceutic-u. Chemistry, \vlio heljKcl to interpret the infrared curves and to synthc-ixe a sample of diyalmitone; P. C. Underwood and C. G. Durbin, liureau of Medicine, who provided in-ii-Iiet-s-ize chickens that had lx-i-n fed the toxic mate-rial, for use in the caica>s re-idm- r'udios; W. L. Hall, T)ivi.-~io;i of Nutrition, \vlio synthesizcil a s:mi]i!c of 3,.~J-cho!estadic~n','; and the mimerou- in.-pec-iors of ihc l-'ood and Druj Admini-iratioii who pruvidr>! essential information without which tins work would no! have pruiie.-.'cd. We are also indebted tr the other indjvidu-d.- and hbor.-iorirs. \.-h-i are --imult.'.nco>;-iy imv.-fi-iilinr: th'.- l-n-l iirn. for difciiF>ii'i:- ih-i! piovided ir.foi': .tion Lcljifiil in <:iiid ; ni; iu in our studies. A'nr.iu; the.-e anl.''i!-lon l'ni~ii':i ('o:;i;':tnv, (lie IVorter ,^ daniblc ('"'iip.iny, and II.-!"- .-'.nd Ibmter. TECHNIQUE In this investigation a rapid, sensitive bioassay procedure, developed in our laboratories, was employed (5). The diet used is a modification of that described by Brew ct al. (2) and consists principally of purified casein, gelatin, clucose, and 16% test fat. Day-old chicks are fed this diet al lib. immediately mi arrival. Subgroups are sacrificed at 10, 14, and 21 days. In general, a higher incidence of pericardial edema is observed in the subgroups sacrificed at 10 and 14 days than in the group sacrificed at 21 days. At autopsy, chicks are examined for the appearance of pericardial edema, ascites, and the appearance of complications, such as liver and kidney damage or labored breathing. Bioassay groups consist of 10 or 15 chicks for each sample of fat. Using a weighting procedure described in Table I, a pericardial edema-activity score is calculated. This makes possible a semi-quantitative comparison between the relative activities of various fat samples. OCCURRENCE Oleic acid samples (U.S.P.) from four manufacturers were assayed biologically and chemically (acid value and unsaponifiable content). As indicated in Table I, several of the oleic acid samples were strongly active, but others showed little or no activity. The pericardial edema activity scores showed that no correspondence existed between the presence of the pericardial edemajiroducing factor and the acid value or the content of unsaponifiables. Tests for the pericardial edema-producing factor in certain commercial samples of glycerol esters (monoglycerides and monodiglycerides) of oleic acid arc summarized in Table II. Seven different manufacturers of glyct-rol esters arc rcprcsoiitivl. The majority of these ester samples wore active. Again, there is no correlation between unsaponifiable content and pericardial edema activity. In four instances, samples of both the nh-ic iirid ami the -~;lyivn>l osti-r proI, i I'oimmmlriilliMt \n "il.l f n. Mvl,'-,, «Mv|-,l,,,, ,, |,r,,ln,,„ L, .i ' I ntun'iili-rl.". i-r l-1-...h, �PERICARDIAL EDEMA ACTIVITY Bioassay Results Samplf , No. Cut. percent Pericaridal edema Analysis Acjd value Unsap. percent Plus death Plus complications' Plus ascites Uncomplicated Other deaths>> Negative Activity ratio1 pos./total Activity score,"! percent TABLE I.-OLEIC ACIDS AND THEIR PE (PERICARDIAL EDEMA) ACTIVITY 201 0 202 5 201 2 202 8 203 0 202 5 201 3 204 9 204 4 203 9 202 0 2 3 4 5 g 7 g g 10 ,1 0 45 0 68 0 77 0 52 2 1 0.18 1 4 0 28 0 51 0 SO 3 3 0.77 21 7 8 8 6 4 10 8 1 ... 1 2 1 1 3 4 10 10 0 52 0 20 0/21 3/21 1/10 2/10 3/10 7/11 0/10 1/9 10/10 0/10 0/10 0 110 33 60 133 200 0 22 300 0 0 TABLE II.—OLEIC ACID ESTERS AND THEIR PE (PERICARDIAL EDEMA) ACTIVITY i 2 1 1 2 2 3 4 5 g 7 8 9 10 11 12 2 6 7 2 0 5 0 3 2 8 4 7 7i 12 5 5 3 3 0.13 0.26 4 0 4 6 11 13 9 9 2 5 13 1 1 4 0 12 1 2 0 32 0 63 0.5 1 2 0.49 0.35 4 3 1 3 4 8 6 9 1 .. 0 46 12 . 3 3 2 6/6 6/10 24/30 0/11 1/14 2/11 0/9 5/10 6/10 1/9 0/9 2/14 320 130 230 0 14 55 0 160 200 33 33 50 TABLE IV.—MOLECULAR DISTILLATION OF AN OLEIC ACID (SAMPLE 6, TABLE I) Charge : F-2 F-3 F-4 F-5 r 11.9 10.3 4 29.11 29. 6/ 9.7 1 7.2 1.2 3 4 10 8 2 1 7/11 0/10 200 0 1/9 5 2 8/10 290 TABLE V.— MOLECULAR DISTILLATION OF A GLYCEROL MONO-ESTER OF OLEIC ACID (SAMPLE 2, TABLE II) Charge F-2 F-4 F-5 f-f „ 10.4 30.61 31. 61 2 1 4 5 1 3 1 4 1 6/10 9/10 130 350 6 11.6 1 2/9 67 1.0 a o n s n c u e gross liver and kidney damage or labored breathing. Early deaths in excess of control groups, PE not observed. ° Chicks with positive pericaridal edema/total chicks in test. rf The "PE activity score" lumber of chicks as follows: is a numerical index of the severity of the activity. The products of the number of chicks i . so e numer of chicks in each category limes its weighting factor are stimulated and divided by the total PE activity score=100 X 5X(PE+death)+4X(PE+compl.)+3X(PE+ascites)+2X(PE)+lX(other deaths) b Total chicks in test group —- �222 223 duced therefrom were bioassayed as summarized in Table III. The glycerol esters showed activity only when the corresponding oleic acid was active. With inactive oleic acids or those with low activity the corresponding glycerol ester did not have increased activity. Thus the edema factor was not formed during the manufacture of the glycerol esters. CONCENTRATION STUDIES Molecular Distillation of an Oleic Acid. A sample of oleic acid known to contain a moderate amount of the pericardial edema-producing factor was fractionated in a 14-in. molecular still into six fractions as indicated in Table IV. The first and last fractions and a composite of the middle fractions were bioassayed as described above. The first 12% strip cut was entirely free of the pericardial edema-producing factor. The middle cuts, representing approximately 60% of the input fat, were only slightly active. The last 7% fraction showed a concentration of the edema factor with an increased activity over the input oleic acid. Thus the pericardial edema-producing factor was concentrated in the last fraction after the bulk of the oleie acid was removed by molecular distillation. Distillation of a Glycerol Mono-ester of an Oleic Acid. A sample of a glj. eerol mono-ester of oleic acid, prepared from an oleie acid known to contain the pericardial edema-producing factor was fractionated into six fractions in a 14-in. molecular still, as described in Table V. The first 10% cut, the last 3% fraction, and an intermediate fraction representing approximately 60% of the input material were bioassayed as described above. An increased concentration of the pericardial edema-producing factor was observed in the first 10% cut. Some pericardial edema-producing factor was present in the other fractions examined. DISCUSSION The material which produces pericardial edema in chicks has been reported previously only in certain lots of fats subjected to special fat-processing operations (1, 2, 3). Therefore the finding that some U.S.P. oleic acids possessed a high degree of activity was unexpected. Feed-grade fats that contained the edema factor characteristically had high unsaponifiable levels ranging above 6% (2). However, in the present study, samples of the more active oleic acids had unsaponifiable levels ranging from 0.2 to 0.7%. This indicates a 10- to 30-fold concentration of the edema factor in the unsaponifiable fraction and suggests that the unsaponifiable fraction of an active oleic acid might be a starting material for isolation of the factor. Molecular distillation of a fat which contains the factor concentrated this factor in the more volatile fractions (2). Molecular distillation of an active glycerol mono-ester of oleic acid also concentrated the edema factor in the more volatile fractions, but separation from the monoglyceride fraction was not as complete as it was from the higher-distilling triglycerides. Molecular TABLE III.-COMPARISON OF PE (PERICARDIAL EDEMA) ACTIVITY OF OLEIC ACIDS AND THEIR CORRESPONDING GLYCEROL ESTERS Oleic Acid Sample No. (from Table 1) 5 (75%) 11 (25%) 5 1 3 Glycerol Ester PE activity ratio pos. /total 3/10 0/10 3/10 0/21 1/10 PE activity score Sample No. (from Table II) percent 133) Of 133 0 33 PE activity ratio pos. [total 2 6/10 3 4 5 25/31 0/11 1/14 PE activit, score percent 13C 23: u distillation of an active oleic acid concentrated the edema factor in the less volatile fraction. Since fatty acids, such as oleic acid, distill at temperatures lower than the corresponding inonoglyeerides, the distillation range of the edema factor appears to lie above that of oleic acid and below that of the glycerol mono-ester of oleic acid. SUilMABY edema °ieic wb either .ich produces The ssw ACKNOWtEDGMENT Appreciation is expressed to Louis T T O ^ ™ ducting the molecular distillations? to Wifem r" ryman^ EGarr Br for c°"' Arde11 and oorrelation of samples, to Herbert AV 5^,°; S f the Pr °<*s for Laboratory for chemical analyses and \n Dfmd £ Her n ^ r? °<Jucts Control and Norris D. Embree for advice °' «ng, Philip L. Harris, REFERENCES Washington, District of Columbia partment of Health, Education and Welfare In a symposium on the chick spectral propertied tories (1,2,3) presented rejorts o h e n o °Ct°ber' 1958' sevfraUabora,, ors elucidation of the toxic factor response for^hf tOWard the ^olation and an ,-yndrome. It was established that the dil ^aaeaee of this unusual y a toxic fa the unsapomfiable fraction of a fatty by n £ ctor fn S t r i a l stearic ,;le,c acid manufacturing operations, and [ f t f,, l n d U suggested that ™* furth . er the factor might posses a polynuclear or steroidal note later » an addendum to the contribution S ' Published (2) :he toxic factor was associated with eluates - ^Ported that and .-ehte" chromatographic columns whfch "si'ane-treated 10 ™ ™ absorption -^nn*^81?^** naPWhaTenes ft >6 and 296 m/t). Neighboring cuts from * secondary XM^ at c roma .-haractenstic spectra of phenanthrene f. tog^ms showeTthe °f Subsequent purification of substanr.A« v. absor w demonstrated that they were not the P«on * ?rae ons in peak at 236 f furthermore our computations showed l ? our materials lc fector m rten present in the diet at levels°of a f «s* be potent ,-ently Harman et al. (4) have rerTortJ ^ °ne partthe million Be »» factor in crystalline form frcTm a^ feSrade « <=Wck edema > Their :or,c to chickens at 0.1 p.p.m. in the ™bstanee was an <(«-trum with a major peak at 244 m! ""raviolet comm a and , , glosed that the crystalline ™ « o f substance Ames et al. • (6) • 'have nhec.^^ J.T. present „,„.•_ : . I?, observed the �224 225 ingredient in a series of dietary treatments involving changes in the level and types of fats to which a group of Cebus monkeys had been subjected. The following summary of experimental results relative to these monkeys was received from O. W. Portman and S. B. Andrus of the Department of Nutrition, Harvard School of Public Health. Of a group of nine monkeys that received this triolein in their diets at a level of 25% by weight, one died at one month and four at three months. After three months on the triolein diet corn oil was substituted for the triolein. The other four monkeys died from three weeks to five months later even though triolein had been discontinued and replaced by corn oil. Of 14 monkeys in the colony that did not receive troilein but were supplied other fats and oils at 25% of the diet by weight, there was only one spontaneous death. Eight of the nine monkeys fed triolein were autopsied and showed the following findings: jaundice (4.S?) : pancreatic atrophy and fibrosis (6): hemosiderosis (6) : fatty liver (5) : bile duct proliferation (3) ; extramedullary erythropoiesis (3) : necrosis of liver (2) : gross hemorrhage in gastrointestinal tract ( 2 ) ; and erythrocytophagocytosis (1). Several features including marked anemia in several instances suggested the possibility of a hemolytic process. Pancreatic changes were most pronounced in the two monkeys that survived longest (seven to nine months from the beginning of triolein feeding). The severity of the lesions in the pancreas was unrelated to that of the hepatic changes. With the exception of fatty changes in the liver, the above findings have not been reproduced in rats. These observations are from an experiment not designed to study a toxic principle, and it would be unwise to draw firm conclusions witli respect to the toxicity of the triolein from these limited data. The fact that, in our laboratory, marked symptoms of chick edema disease were produced by this sample of triolein suggests the possibility that the chick edema factor may have been responsible for the toxic effects noted in the triolein-fed monkeys. We now wish to report the ioslation of a highly toxic crystalline substance from this triolein and to describe its properties. alumina, using 250 g. in a 42x4.5-cm. tube. Several fractions which eluted with 5% ethyl ether in petroleum ether exhibited an absorption maximum near 245 m/« as well as the peaks previously observed at 234-236 m/t and 255-260 mn. There fractions were combined, and the eluted material was partitioned on 5 g. of silane-treated Celite, by reverse-phase chromatography, using 4 ml. of iso-octane in the immobile phase, and 80% alcohol saturated with iso-octane as the mobile solvent. Fractions exhibiting maximum absorption at 245 m/t, but not at 235 m/i or 255/t, were combined for two further chromatographic purifications on alumina. Jlerck alumnia was deactivated with 3% its weight of water, and the ratios of alumina to sample and column size 2,000:1 in a 25 x 1-cm tube and 25,000:1 in a 25 x 2-cm. tube, respecively. Iso-octane, redistilled and chromatographed over silica gel, was employed as eluent, and fractions were again monitored by ultraviolet spectral absorbance fractions devoid of inflection and 255 ni/t were combined and evaporated to dryness. The solid residue was dissolved in a small volume of boiling iso-octane and stored over-night in the refrigerator. White crystals weighing 2.64 mg. were obtained. The ultraviolet absorption spectrum of the crystals dissolved in iso-octane tvas characterized by maxima at 245 m/t (EJ*m = 718) and 300 rap. (E=**m SO) and inflections at 240 m/t (Ej? m .= 655) and 310 m/t (E;?m.=76). The substance sublimed on the hot stage about 239°C. The infrared absorption spectrum of the substance, incorporated into a potassium bromide disc, showed evidence of aliphatic and aromatic linkages but no bands characteristic of oxygen or nitrogen functions. The reported presence of chlorine in the toxic substance isolated by the Merck group (5) prompted a Beilstein test, which showed the presence of halogen. The presence of halogen in the crystalline material being reported here was confirmed by examination in a inicrocoulometric gas chromatograph.1 (In this instrument the sample is fractionated by gas-liquid chroma tography, and, as they are eluted from the column, the compounds are pyrolyzed at 800° C. under oxidizing conditions. The halogen acid formed from lialogenated materials is titrated in a mierocoulometer.) In the chick biossay the substance, when fed at en. 1 p.p.m. in the diet, produced severe hydropericardium, hydroperitoneum, and liver damage, with death occurring within 12 days. At 0.1 p.p.m. in the diet marked hydropericardinm was evident at autopsy after a three-week feeding period. The isolation of the toxic substance had been complicated by the presence of another material with an absorption maximum at 248 ni/t. This second substance was isolated by the same techniques of chromatography. monitored by ultraviolet spectrophotometry, and a yield of white crystals was obtained. The ultraviolet spectrum of this material was identical with that of the toxic substance except that it was shifted 3 m/t so that the major absorption peak \vas at 218 m/t. It behaved similarly to the toxic substance in the microcoulometric cas chromatograph, showing a similar retention time and a similar haloffpn content. However, it was completely inactive in the chick edema test \dien fed at 1 p.p.m. in the diet. In order to obtain additional crystalline material the portion remaining from nondestructive testing, fortified with the adjacent fractions from the final chromatography (of high potency, judging from the iiltraviolet spectrum), was further chromatographed according to the previous isolation scheme. Small quantities of phenanthrene derivatives were separated with a consequent diminution in the ultraviolet absorption in the 250-300 m/t region. Furthermore, although there was no indication of nonhomogeneity by paper chroma togniph.v, the ratio of absorbances of the shoulder at 240 ni/t to the !>eak at 245 varied from fraction to fraction, suggestion the presence of an additional component. The extinction values reported above may be in error because of inaccuracies in weighing on account of the difficulties in handling the small amount of material involved. There is no doubt however of the validity of the relative absorbances at various wave-lengths. The spectrum of the substance reported here differs somewhat from that reported by Harman et al. (4), particularly in the ratio of the absorbanee at the maximum (244 or 245 in/t) to the absorbance at the inflection at 238 or 240 m/t. EXPERIMENTAL The triolein sample was of excellent quality with an unsaponifiable content of 0.87% and a steroidal hydrocarbon (1) absorbance of 0.07. Only 0.01% of oxirane oxygen (epoxide) was detected. Examination of the fatty acids as the ethyl esters by gas chromatography showed that oleic acid constituted 70.9% of the total acids with 3-6% liuoleie, 0.3% linolenic, 13% palmitoleic, 9.4% palmitic and shorter-chain fatty acids, and 1.8% of a Cn fatty acid containing one double bond (by inference from relative retention time). Urea filtrate fatty acids were present to the extent of 4.73% (6a). When fed to chicks at a level of 15% in the diet, the triolein produced the symptoms of chick edema disease with a severity approximately equivalent to that observed with a diet containing the toxic fatty product used in our original studies (2) at the 5% level. The unsaponiflable fraction of the triolein was proportionately richer in the toxic factor than any other materials available to us. The toxic factor in 17.6 kg. of triolein was concentrated by molecular distillation under pressures of 10-20 microns at temperatures up to 200°C. The distillate (289 g.), which contained all of the toxic factor, was saponified, and 166 g. of unsaponifiable material were recovered. The portion soluble in petroleum ether (155 g.), was chromatographed on 3 kg. of Fisher Alumina A 540 in a 4-ft. x 3%-in. column, using petroleum ether as eluent. Two-liter fractions were collected, and substances with the absorption spectra of naphthalene and phenanthrene derivatives were eluted in fractions 6-15. The bulk of the material containing cholestadiene and related hydrocarbons (129 g.) was eluted in the first three days, and the remainder was recovered by the use of more polar solvents. These foreruns and tailings were devoid of potency in the chick edema test. Fractions 6 to 15 were combined (66S ing.) and chromatographed on 2,500 * g. of the more retentive Merck Alumina Xo. 71707. Xo material was eluted with petroleum ether. Foreruns were eluted with 1-2% ethyl ether in petroleum ether, and fractions exhibiting the spectra of naphthalene and phenanthrene derivatives (239 mg.) were obtained when 5% ethyl ether in petroleum ether " was employed as eluent. This material was again chromatographed on Jlerck »Dohrmann Manufacturing Company, Palo Alto, Calif. �226 227 DISCUSSIOIf The toxic substance which we have isolated from triolein resembles that recovered by Harman et al. (4) from animal feed tallows. However the divergences in their properties suggest either that we are dealing with two different but closely related compounds, or that one or both of the preparations is still a mixture of related compounds despite the fact that only a single spot could be obtained on paper ehromatography in a number of solvent systems. In view of the manifest difficulties involved in isolating a pure compound in minute quantities from a myriad of substances with similar properties it would be hazardous, as Harman warns, to infer chemical structures from spectral data. Nevertheless it should be pointed out that the spectra obtained by us and by Harman et al. are strongly reminiscent of those exhibited by highly substituted naphthalenes (7). Furthermore the toxic factor occurs in association with a bewildering array of aromatic naphthalene and phenanthrene derivatives, as we have previously noted (2). The detection of chlorine in large proportions in a toxic preparation, and the ultraviolet spectrum observed, suggest a possible relationship with chlorinated naphthalenes. Pentachloronaphthalene possesses an an absorption maximum at 243 mjt and a secondary maximum at 312 m^ (8), and it has been shown to cause hyperkeratosis in cattle (9) and several other species of animals including chickens (10). Other chlorinated naphthalenes also are toxic (11). The possibility that the chick edema factor is a chlorinated napthalene derivative cannot be ignored. Samples of tetrachloronaphthalene and hexachloronaphthalene, kindly provided by Engel and Bell* of the Virginia Polytechnic Institute, who had demonstrated that these compounds could produce hyperkeratosis in cattle, were without effect in the chick edema test. Furthermore these compounds, despite the similarity of their ultraviolet spectra and their chromatographic behavior to the toxic substance, showed considerable difference in the microcoulometric gas chromatograph. For instance, chlorinated pesticides, aldrin and heptachlor, showed retention times of 10 min., tetrachloronaphthalene 9 min., and hexachloronaphthalene 14 min., whereas the toxic substance, as well as its inactive analogue with the absorption maximum at 248 m/i, had retention times of 3T-3S min. It is tempting to speculate that the greater retention-time of the toxic material is related to a greater molecular weight or to a substituent conferring different solubility and polarity properties. We are continuing our studies toward the isolation of the toxic factor. It is necessary that the chemical nature of this substance be elucidated to make possible a rapid chemical test for its detection, to clarify its origin, to verify the suggestion of its severe toxicity to primates, and to study its action in other species. ACKNOWLEDGMENT Numerous individuals in addition to those mentioned in the text have contributed directly or indirectly to this project. We are particularly grateful to O. L. Kline for his sustained interest, encouragement, and helpfulness in the course of these investigations; to Benjamin Webb, who conducted the extensive molecular distillations; to Donald F. Flick and Linda Gallo for the chick bioassays: and to Raymond J. Gajan for the microcoulometric gas chromatography. The cooperation of the various commercial laboratories that were also working on this problem, in discussing their work with us, is also greatly appreciated. REFEREXCES 'Brew, W.B., Dore. J.B., Benedict, J.H.. Porter, G.C., and Sipas, E., J. Assoc. Offlc Agr. Chemists. -}2, 120-128 (1959). = Friedman, L., Firestone, D., Horwitz, W., Banes, D., Anstead, M., and Shue, G., ibiA 129-140. 3 4 Wooten, J.C.. and Alexander, J.C., ibid., 141-148. Harman, E.E., Davis. G.E., Ott, W.H., Brink, N.G., and Kuehl, F.A., J. Am. Chem Soc.. 82, 2078-20T9 (1960). 5 Tlshler, M., Merck and Company, private communication. July 19. 1960. •Ames. S.H.. Swanson, W.J., Ludwig, M.I., and Brokaw, G.T., J. Am. Oil Chem. 37 4(10) (1960). ' Abadir, B.J., Cook, J.W., and Gibson, D.T., J. Chem. Soc., 1953, 8; Mosby, W.L., J Am. Chem. Soc., 75. 334S-3349 (1953). 8 9 Blickenstaff. R.T.. and Callen, .I.E.. Anal. Chem.. 26, 1586-1589 (1954). Sikes. D.. and Bridges. M.E., Science. 116. 506-507 (1952). M Kohler. H., Archiv. Experimentelle Veterinarmedizin, 8, 163-198 (1954). "Bell, W.B.. Vet. Med., 48, 135-140 (1953). Collaborative Bioassay for Chick Edema Factor* A substance contained in certain processed fats and fatty products used in poultry feeds has been implicated in the outbreak of the condition called "chick edema disease" which •Presented as the report of the Associate Referee on Bioassay of %ick EdemiiTParto? Car £ Douglass, at the SeveSyTourtt in-' aua Meeting of the Association of Official D.cemi8ts' •Oct PLICK (Kv f *°* ° occurred in 1957 (1). A number of laboratories are actively working to isolate and identify the agent responsible for the disease (2-5) and each has developed its own bioassay, which in each case has admirably served its specific purpose. Ames and co-workers (6), early in I960 reported this factor in oleic acid samples destined for human consumption. The pres- �229 228 etice of this highly poisonous substance in human food products immediately resulted in a regulation from the Food and Drug Administration specifying that all such products must- be free of "chick edema factor" in order to be incorporated in food. From the point of view of the regulated industry and the regulatory agency, the need to develop an assay method of adequate specificity and sensitivity, the results of which are comparable from laboratory to laboratory, can hardly be overemphasized. Since the chemical characterization of the substance has not proceeded far enough to provide the basis for a chemical or physical assay, it is necessary to use the bioassay. This method has successfully proved itself capable of providing for the detection and semiquantitative estimation of small amounts of this factor. When the need for the standardization of a method became acute, investigators in eight laboratories that are currently engaged in some phase of work on the chick edema factor were invited to participate with us in a collaborative study and to submit suggested procedures. All graciously responded, and from the procedures which they forwarded, the authors selected what they considered the most desirable features of each and com- bined them into the procedure presented here. Three collaborative test samples were prepared by thoroughly mixing a fat, known to be toxic, with cottonseed oil in «ch proportions that the amount of tone fat in the samples was in the ratio of 1:2:4. Each of the collaborators was sent the procedure given below as well as four collaborative test samples identified by number. Sample 1 was USP cottonseed oil; Samples 2, 3, and 4 contained, respectively, 1, 2, and 4 grams of toxic fat in 16 grams of sample. Report sheets suitable for recording the following specific information were also sent: chick number, initial body weight, final body weight, weight gain, heart fluid volume, presence or absence of peritoneal and subcutaneous fluid, and dates of early deaths. Noteworthy features of the assay procedure are the use of white leghorn cockerels as the assay animal; a semi-synthetic ration in which the test samples are contained at 16%; and a 21 day feeding period. We included a request that groups be assigned to provide for observations to be made at the 14th day as well. The collaborators were asked to score the results visually, by use of a suggested scale of values, as well as by measurement of the pericardial fluid volume. CaCO, Ca.(PO.)= KSHPO. NaCl Na=HPO. MgSO..7HsO MnS0..4H=O Fe citrate ZnCO, CuS0.5H,O G MIXT. 15 14 9 8.S* 73 4.86 0.42 0.4 02 0.02 12 mg Avenge Hydropericardium AveraM Weight Gain Total (c) Vwial Score 14 21 14 21 14 21 14 21 12 12 12 12 12 12 12 12 0 2 1 1 0 2 3 5 39.5 ± 73.0 ± 52.8 ± 78.2 ± 47.0 ± 69.4 ± 20.2 ± 39.4 =fc 3.2 4.3 4.3 4.9 3.2 4.6 3.3 5.1 1 0 0.054 ±.004 1+ 0.12 ±.025 1 1 2 2 3 3 4 4 14 21 14 21 14 21 14 21 6 5 7 5 6 5 7 5 1 0 1 0 0 1 3 4 26.2 ± 59.6 ± 32.7 ± 70.0 ± 39.8 ± 63.2 =t 39.3 ± 57.0 ± 6.7 12.6 9.2 4.3 2.9 4.7 7.6 9.6 1+ 0.028 ±.013 0 .0.072 ± .013 9+ 0.095 ± .021 3 1 1 2 2 3 3 4 4 14 21 14 21 14 21 14 21 12 12 12 12 12 12 12 12 0 0 0 0 0 1 0 1 46.1 ± 104.1 ± 51.4 ± 118.2 ± 48.7 ± 89.5 ± 45.4 ± 86.1 ± 2.0 0 0.023 ± .001 0 3.3 0 0.085 ± .016 0 2.5 4+ 0.101 ±.018 1 + 6.3 3+ 0.145 ± .036 4+ 2.6 12+ 0.39 ±.13 9+ 4.9 7+ 0.28 ± .12 7+ 3.3 14+ 0.54 ±.21 14 + 4.4 23+ 1.46 ±.45 25+ 1 1 2 2 3 3 4 4 14 21 14 21 14 21 14 21 12 12 12 12 12 12 12 12 0 0 0 2 0 4 0 8 42.3 ± 91.3 ± 43.5 ± 41.8 ± 28.1 ± 35.0 ± 22.3 ± 32.3 ± 1.6 8.9 2.9 5.1 2.1 1.4 1.8 5.1 7 1 1 2 2 3 3 4 4 14 21 14 21 14 21 14 21 12 12 12 12 12 12 12 12 0 0 0 1 2 2 1 9 8 1 1 2 2 14 21 14 21 12 12 12 12 0 42.9 ± 0 100.8 ± 0 54.5 ± 1 j 78.1 ± 2 AMOUNT, 0 Folic acid (1.0% triturated in powd. glucose) 10 Biotin (0.1% triturated in powd. glucose) 75 Vitamin B,, (0.1% triturated in powd. glucose) 25 Niacin 2.0 Ca pantothenate 050 Thiamine 050 RiboBavin 0575 Pyridoxine-HCI 0200 Menadione 0.025 Celluflour (Alphacel), to make 500 (c) Fat-soluble iritamin mixture.— AMOUXT/KG MIXT. AMOUXT/KG DIET 900.000 100,000 2.00 1.00 9.000 USP Units 1.000 1C Units 20 mg 10 g USP Units 1C Units g kg 0 1+ 0.11 ± .009 0 0 12+ 0.26 ±.039 7+ 20+ 20+ 27+ 10+ 7+ 15+ 20+ 15+ Edema Incidence Score Fluid Volume From (ml) Volum e 1 1 2 2 3 3 4 4 1 (b) Vitamin mixture.— G/60 Vitamin A acetate, cryst. Vitamin D», cryst. D-a-Tocopheryl acetate Corn oil, to make CoU No. Aaaay No. of Period of Early Diet (Dayi) Chicki Death i 5 METHOD Reagent* (a) Salt mixture. Table 1. Summary data of the collaborative assay for chick edema factor 0.23 0.92 0.71 1.58 9+ ± .09 3+ ±.29 21 + ±.25 15+ ±.72 22+ 0 0 0 0.59 ±.24 7+ 0.155 ± .079 2+ 1.43 ±.49 14+ 2.05 ± .78 16+ 1.80 ±.66 15+ 0 0.059 ± .008 0 0 0.081 ± .034 0 0.127 ± .006 3+ 0 Aacites Sub- Hydrocuta- pericarneoua dium* 0/12 0/12 1/12 0/12 0/12 1/12 0/12 0/12 1/12 3/12 1/12 6/12 0/12 0/12 0/12 7/12 1/12 1/12 2/12 10/12 2/12 6/12 3/12 10/12 0/5 0/5 0/6 0/5 1/6 2/5 5/7 4/5 0/5 0/5 0/6 0/5 1/6 1/5 6/7 3/5 0/12 0/12 0/12 0/12 2/12 1/12 3/12 8/12 0/12 0/12 0/12 0/12 0/12 1/12 0/12 3/12 6/12 4/12 8/12 5/12 4/12 6/12 9/12 9/12 0/5 0/5 0/6 4/5 1/6 5/5 5/7 5/5 0.102 ±.021 2+ 0.66 ± .29 17+ 0.37 ±.17 9+ 0.80 ±.25 20+ 0/12 0/12 0/12 1/12 0/12 10/12 1/12 8/12 0/12 0/12 0/12 1/12 2/12 8/12 5/12 9/12 0/12 0/12 3/12 8/12 2/12 11/12 6/12 9/12 22.6 ± 5.9 0 0.05 ± 0 0 33.4 ± 10.0 0 0.054 ± .004 0 1I.2± 6.3 0 0.058 ± .005 0 16.2 ± 7.2 8+ 0.196 ±.093 3 + 25.4 ± 5.2 4+ 0.166 ±.078 2+ 35.9 ± 10.3 10+ 0.427 ± .203 9+ 12.4 ± 7.2 10+ 0.391 ± .14 9 + 22.0 ± 15.6 6+ 0.204 ±.078 6+ 0/12 0/12 0/12 1/12 1/12 1/12 3/12 3/12 0/12 0/12 0/12 4/12 2/12 2/12 5/12 6/12 0/12 0/12 0/12 1/12 1/12 4/12 5/12 5/12 0/12 0/12 0/12 0/12 0/12 0/12 0/12 0/12 0/12 2/12 2/12 4/12 (Continued) 10+ 0.306 ±.052 10+ 1+ 16+ 11+ 19+ 3.8 0 7.3 3+ 3.2 9+ 4.7 15+ 0.061 ± .015 0 0.093 ±.013 0 0.125 ± .019 2+ 0.183 ± .023 4+ �230 (d) Assay Ration.— Sucrose, commercial Corn starch, commercial Casein (vitamin-free) Fat (USP cottonseed oil or asay fat) Gelatin Salt mixt., (a) Vitamin mixt., (b) Salt, iodized Fat-sol, vitamin mixt., (c) Choline chloride, 25% aq. soln 205 203 20 16 13 6 2 12 1 02 Treatment of Experimental Animals Use day-old white leghorn, single comb cockerels. On day of receipt of chicks, tag individually, record body weights, and place in brooder cages equipped with heater. Use room with controlled temp, and humidity. Offer control ration contg 16% cottonseed oil as fat and H:O ad libitum. After 48 hr, weigh chicks and do not use any chick which is outside limit of mean body wt by ±5 g. Place 12 chicks in each brooder cage and record body wt and date of beginning feeding regimen. A&iay Period Continue 1 group on control ration contg 16% cottonseed oil and substitute test fats for all or part of the 16% cottonseed oil in assay rations. Check chicks daily for fatalities and for presence of adequate food and H=0. Record all deaths and cause of deaths. For each death due to other than accidental cause, autopsy and record presence of hydropericardium, hydroperitoneum. subcutaneous edema. and amount of heart fluid to 0.01 ml. Autopsy remaining chicks on 21st day and record findings as above. Postmortem Examination for Quantity of Heart Fluid (a) Sacrifice oj chicks.—Sacrifice by means of cervical dislocation and proceed as follows: (b) Exposure ol heart.—With dissecting scissors make small transverse cut in skin over full diam. of abdomen. Peel skin toward head and lay skin fold over head. This cutaneous incision permits wide field exposure of subcutaneous area over thoracic and abdominal cavities for examination for subcutaneous edema. Skin may be reflected caudally to afford wider field of vision. Record (+ or — ) evidence of subcutaneous edema. : Insert blunt tip of scissors thru body wall. 231 and make transverse incision of musculature to lower rim of rib cage, avoiding cutting into organs of peritoneal cavity. Lift breastbone with fingers, insert blunt tip of scissors, and carefully enlarge incision by cutting on each side of chest cavity thru rib joints up to clavicles. (Do not cut into subclavian vessels.) With sufficiently wide cut, fingers may be used to protract incised chest cavity and thus permit clear observation of substernal attachment of pericardium. Firmly clasp pericardial attachment with fingers and reflect flap of sternum so as to expose heart with pericardium intact. Estimate visually and record severity of pericardial edema according to following table: Pericardial Edema this value falls at the high end of the normal range of the volumes of pericardial fluid for Collaborator 1. ! M Days a Days chicks of the age and weight used in the Ktnoe>on*r From the observations reported by the collaborators it may be generalized that of the three anatomical sites in which fluid may accumulate, hydropericardium occurs with the greatest frequency, followed in order by peritoneal and subcutaneous accumulation. This is in agreement with earlier observations. The appended charts (Fig. 1) show graphically the response of the volume of peri- Score Collaborator 2. 14 Days 21 Days "Days Absent Slight Moderate Severe Very severe (c) Withdrawal of heart fluid.—(1) For voli estimated as <,1O ml.—Firmly clasp apex of pericardium with small forceps and make small incision in heart sac. Insert small spatula into incision of pericardium and push heart to one side. Carefully aspirate fluid into 1 0 ml tuberculin syringe. Blunt-end 18 gauge needle permits more complete aspiration of small vols. To reduce formation of air bubbles during aspiration, prerinse needle and syringe with n-butyl alcohol. Do not include vol. of n-butyl alcohol in measurement of fluid. (2) For vols estimated as >1J) ml.—Insert sharp hypodermic needle on 10 ml syringe into intact pericardium. Aspirate as much as possible of the heart fluid (n-butyl alcohol rinse is unnecessary). Collect and measure remainder with blunt-end tuberculin syringe. Record total vol. heart fluid. (d) Other observations.—Observe and record other obvious changes such as peritoneal edema (ascites), liver changes, kidney changes, etc. a Days E -0 l« —• 14 Days Collaborator 5. Collaborator?. 21 Days 14 Days |3 21 Days U fl) 0-2 o •> n Collaborators. 14 Days Collaborator 9. 21 Days 14 Days Results Seven of the nine cooperating laboratories submitted results in time for inclusion in this report. Their observations are compiled in Table 1. It is seen that mortality due to the toxic factor does not become appreciable until the third week on the diet containing the highest level of the toxic fat. Xote that three of the Collaborator 3. ! 4 1 2 Diets [g. f—Velum* of pericardia! fluid (— S.E. of the mean) versus level ' " [ periods. 21 Days �233 232 Table 1. (Continued) ' Edema Incidence Average Hydropericardium Weight Gain No. 1 of Aasay No. Early I of Period Total *g! Coll. 1 Diet (Days) Chicks Deaths i 2.8 ! 13 + 8.3 1 28 + 48.3 ± 5.4 : 28+ 75.1 =fc 6.4 ! 26 + i 3 14 12 ! \ 3 4 4 21 12 12 12 0 2 1 5 12 12 12 12 12 12 12 12 0 54.5 ± 104.4 ± 1 0 ' 57..'i ± 108.0 -r0 II 1 o:t.:t ± 1 104.0 =t 0 i 60.4 ± 89.6 ± 5 14 21 9 , 1 14 ! i 21 !4 21 II 21 14 21 2 •7 :i 3 4 4 j : Vi«ual Score 45.1 ± 72.5 ± Score Fluid Volume From (ml) Volume 0.203 ± 0.878 ± 0.903 ± 3.44 ± 0.09 3.3 i 0 0.10 9.6 : 0 1 + 0.15 5.6 7.1 ; 10 + 0.21) :t.o , 'I t D.I 11 Ascitei .066 5+ .339 17 + .303 20+ 1.40 19 + =fc .01 ± .015 ± .07 + .05 ±.04,1 11.8 ! 36 + 2.86 =t .55 4.8 : 21 + 0.84 ± .24 6.7 35 + 4.52 ± .86 1/12 4/12 1/12 7/12 5/12 8/12 6/7 6/7 0/12 0/12 0/12 0/12 0/12 9/12 6/12 ; 11/11 1 0 0 2+ 10 + 5442 + 22 + ! 34+ Sub- Hydrocuta- pericarneous dium* 4/12 8/12 8/12 7/7 0/12 0/12 0/12 0/12 0/12 1/12 0/12 D/12 0/12 •1/12 5/12 12/12 2/12 11/12 9/11 11/11 • Hydropericardium incidence: baaed on measurement of pericardial 3uid. collaborators reported deaths of four birds on the control diet. Since these deaths could not have been due to the factor, this and the foregoing observation indicate that mortality measurement alone is not a reliable index of the presence of the chick edema factor. It has been found that ,-igns of chick edema disease arc the development of hydropericardmm, hydroperitoneum, and subcutaneous edema, in that order, a.- the dose of toxic material is increafed. At higher dosage levels, death will occur from the 10th day. Birds dying of chick edema disease invariably exhibit these signs. Deaths unaccompanied by any of these sign.- cannot be attributed to the toxic agent. Weight gains of the chicks were somewhat depressed at the higher levels of toxic f a t . In the control groups, the average weight gains at the end of the a.-«iy periods were quite variable among the laboratories. This variability may be due to borderline nutritional inadequacies of the basal diet that possibly were aggravated by hereditary or environmental factor.- or a combination of both, as may have happened in the ra.-c of Collaborator 7'.- group. !Vrhap> it would have been wise to follow the suggest ion (if cerfcun of the collaborators and to have an antibiotic incorporated in the lest ration. Visual subjective scoring of the degree of hydropericardium according to the instructions given in the collaborative procedure is seen to agree fairly well with a score calculated from the actual measurement of pericardial fluid as shown in Table '2. The figures given in Table 1 represent the sum of the plus scores for individual chicks within the group. It appears that there was greater accuracy in visual scoring of large toxicity responses than in the smaller responses. While visual scoring may be adequate in judging the presence of a large amount of toxic material, it is inadequate in those borderline cases where controversy is most likely to develop. We have arbitrarily adopted ().'_' ml a.s the upper limit of normal heart-sac fluid volume. This choice is justified on the basis of the work of Shue and Hallo (7), who report that Table 2. Scoring by Referee of pericardial edema Volume of IVricardlal Fluid <0.20 0.21 0.40 O..II 1.00 1. 01 2.W >2.01 Toxicity Score 0 + +4 f 1- + + +++ cardial fluid to the level of toxic fat in the diet. It is seen to vary directly with this level. It is likewise seen that an appreciably greater response to the dose is obtained after a three-week feeding period than after a two-week period. Although there is a high incidence of negative responses in a group of chicks on a low level of toxic fat, the average response of the group as measured by the volume of pericardial fluid is proportional to the dose level within the dose limits of this experiment. of sodium chloride was likewise the subject of question on the basis that 2% of salt is too high and subjects the chick to extra "stress." It has been found in our laboratory that this level favors the development of the edema. Seiye and Stone (8) have found that extra sodium chloride in the diet of chicks accentuates the development of edema produced by certain steroids. Biester and Schwartz (9) state that 6-8 grams of sodium chloride per day is not harmful to 9 week-old chicks. We cannot see that a change here is justified. Cmiiiiirtil* of ( olUlx.rolori. One collaborator reported thnt flic ration was unpalatable to Ins chirks The weight Mont, of tin 1 comments were directed at the nutritional adequacy of the basal ration. KIUIIS reported by him were much lower than One investigator has suggested that dry, in any other laboratory. We have no explastabilized vitamins A, D, and E be used nation for this effect. It is suggested that this rather than the crystalline products used in laboratory may have some peculiar problem the procedure. This suggestion appears to be and that the performance of these chicks meritorious and has been incorporated in is not typical. the recommended nrocedure as optional. AnSeveral of the collaborators preferred to other objected to the mode of addition of use heavy breeds of chicks rather than legcholine to the diet and suggested a 25% dry, horns. We have retained leghorns for the free-flowing preparation which is commer- reasons of uniformity of response, widecially available. For the sake of convenience, spread availability, convenience, and the this has likewise been incorporated a.« op-' demonstrated sensitivity of this breed. tional. Objection was made by another colSummary and Recommendation** laborator to adding the water-soluble vitaA successful collaborative study of a promins on Olltiflour as the carrier. Since tnanv laboratories routinely use other of the dietary cedure for the detection and assay of the ingredients as carriers for the vitamin mix- chick edema factor in fate and fatty mateture, we see no reason why any of the major rials has been carried out. The results indicomixments cannot be used for this purpo.se. cate that the method as studied is satisfacWeight, gains obtained on the diet are not tory for the intended purpose. Although it optimal Tlii.s, however, is nnl crilir.-il I rum is ileiiion.-lrated here t h a t (lie iiiellmd :-lmllnl the standpoint „! t h e assay except, as ,,,,e i - cap.'iblr "I ililleienli.'ilini; riinlmnimiled from mieoiil.-iiimi.'ilcd fair-, il would be de collalxinifor pomf* l m i, l|i;,i ,|, ,. ea.-ier to r;irry out Uie procedure ,,| w i t h d r a w i n g t|,,. 'Ullllilr III c.'lllv ( M i l ,'|i|illl|iill,'il rnllMlinrilllvc heart-sac fluid from a larger bird than from slmlies mi samples of lower [xitencies t h a n a smaller one. Since the diet has given excel- those ll-ril here. It. is recommended 1 — lent pericardial fluid responses, major (1) That the method for bioassay of chick changes cannot be justified at this time. Other collaborators have suggested that edema factor, presented in this report, be an antibiotic be incorporated in the ration, adopted as first, action. <2l That collaborative studies Ix- conthat, crude casein he substituted for the vitamin-free casein, and tliat commercially avail- tinued. able sill, and v i t a m i n mixtures be UMI|. TlieM' HKKKIIKM KM n|i|iear In be desirable from the sl:irn|p,,,nls (I) K d i l o i V N'nle, Tin* Jinn mil. 42, 120 (MI.W) of iTommiv and convenience •,,,d will fnrtii ninwtvrtl liv (lie IBIM-C nf a mollification „( the diet to be Ihi'' Tdi'Mp riTHimiti'iliitiiiHM li.vWfff (i'-iii'i-iil Iti-feriT iinif HiibcoimniltiT *\ used in further collalx>ralive work. The level anil \vrri- mlnptril l>v tin1 Association. Si'i* Thin Juiiriiiil. 44, 70 ( I t M t l ) . �234 235 PROGRESS IN THE CHICK EDEMA PROBLEM (2) Brew, W. B., Dore, J. B., Benedict, J. H., Potter, G. C., and Sipos, E., ibid.,- 42, 120 (1959). (3) Friedman, L., Firestone, D., Horwitz, W., Banes, D., Anstead, M., and Shue, G., ibid., 42, 129 (1959). (4) Woolen, J. C , and Alexander, J. C., ibid., 42, 141 (1959). (5) Hannan, R. E., Davis, G. E., Ott, W. H., Brink, N. G., and Kuehl, F. A, J. Am. Ghent. Soc., 82, 2078 (1960). (6) Ames, S. R., Swanson, W. J., Ludwig, M. I., and Bookaw, G. Y., /. Am. Oil Chemists' Soc., 37, 10 (1960). (7) Shue, G. M., and Gallo, L., This Journal, 44, 456 (1961). (8) Selye, H., and Stone, H., Proc. Soc. Exptl. Biol. and Med., 52, 190 (1943). (9) Biester, H. E., and Schwartz, L. H., Diseases of Poultry, 4th Ed., The Iowa State University Press, Ames. Iowa, p. 112. Collaborators J. C. Alexander, The Procter & Gamble Co., Research Division, Cincinnati 39, Ohio Stanley R. Ames, Biochemistry Department, Distillation Products Industries, Rochester 3, N.Y. Carl D. Douglass, Food and Drug Administration, Washington 25, D.C. O. F. Hixon, Laboratory of Vitamin Technology, Inc., 7737 S. Chicago Ave., Chicago 19, 111. Walter H. Ott, Merck Institute for Therapeutic Research, Rahway, N.J. C. E. Poling, Swift & Co., Union Stock Yards, Chicago 9, 111. H. C. Sehaefer, General Research & Control Laboratories, Ralston Purina Co., St. Louis 2, Mo. By Dr. Leo Friedman, Food and Drug Administration It has been almost four years since I first became aware of the problem that we know today as "chick edema disease." Because of our activity on this problem, iny colleagues and I have had the opportunity to become acquainted with many scientific groups and individuals working in the same area with whom we have enjoyed a fruitful cooperation and pleasurable association. We hope that they feel as kindly toward us as we do to them, but sometimes, I known, they wish as we do that they had never heard of chick edema disease. This problem has been most difficult and progress frustratingly slow. Despite the meager amount of new information that can be added at this time, especially since the most recent advances were reported by Dr. Artman recently, it is nevertheless worthwhile to review the several aspects of this problem and see its present status in full perspective. As you recall, during 1957 an epidemic disease caused millions of dollars in losses among broiler flocks throughout a large part of the U.S. After elimination in succession of all other possibilities, attention was focused on the fat ingredient of the feed as the etiologie agent. A series of reports in 1958 from several laboratories described the manifestations of the disease and definitely implicated a toxic fat or a toxic substance in fat as the cause. The characteristic symptoms were droopiness, ruffled feathers, labored breathing and high morbidity and mortality. Autopsy findings revealed hydropericardium, abdominal ascites (water belly), subcutaneous edema, swollen liver, swollen and pale kidneys, etc. In laying hens the toxic fat caused a rapid drop in egg production. Pullets receiving toxic fat during the full growing period did not come into production, and mortality was very high. Hydropericardiurn, the most common lesion found in young birds, was not found in birds of laying age. SYMPTOMS DIFFERENT These differences in susceptibility and symptoms in different age groups of the same species should be noted. The feeding of toxic fat to other species lias not produced such striking results as with young chicks, with the exception possibly of monkeys. However, every species that has been tested has shown evidence of deleterious effects. Very little work has been done_ with rats. Our very limited experience indicates that they are much more resistant than chicks in short-term feedings, but that when fed in sufficient dosage, extracts of tlie toxic fat produce definite deleterious effects as shown by growth depression, enlarged and fatty livers, and marked involution of the thymus. I recall few reports of the effects of toxic fat on swine, but again, in our own limited experience we have seen depressed growth . . . and have demonstrated the presence of toxic factors in the meat of hogs that had been fed toxic fat. I am indebted to Dr. Wilcke of Ralston Purina for reports of studies on guinea pigs and dogs. Guinea pigs fed 2%% toxic fat stopped growing after six weeks, and death losses occurred after eight weeks. At a level of 4%% toxic fat weight losses occurred after three weeks and deaths after four weeks. Control groups receiving non-toxic fats did not show weight loss or deaths. The only observed pathology at the conclusion of the experiment was congestion of the lungs and mottled livers. In experiments with three different breeds of dogs, using Purina Dog Chow in which 10% of toxic fat was substituted for the usual normal fat, there was poor reproduction and lactation performance. The females on the toxic fat ration whelped pups that were dead or weak, and, furthermore, the mothers <*?med to have an insufficient milk supply. When the pups were removed before n-eiining and fed a normal ration the increase in growth was immediate and dramatic. Also, the females on the toxic fat ration tended to lose hair on their sacks and shoulders. With the ration containing toxic fat, post-weaning growth tests (6-18 weeks) with five litters of pups demonstrated inferior growth performance using either weight gain or increases in body length as the criterion. WORK WITH OTHER SPECIES Iii these experiments with other species, fat that had first been proved to ;«• toxic to chicks had been used. In the case of monkeys, a sample of triolein �236 that had produced irreversible toxic symptoms for no apparent reason later was proved toxic to chicks and was the source from which we isolated a highly purified crystalline "chick edema" factor. At the present time. then, "toxic fat" that produces "chick edema disease" has heen demonstrated also to produce deleterious effects in rats, guinea pigs, swine, dogs and monkeys. It should be emphasized that the toxic fat undoubtedly contains many other substances that may have effects in these other species. Definitive information as to the effect of "chick edema factor'' (CEF) in other species must await tests with purified CEF. Triolein-fed monkeys probably received the "purest" source of CEF. However, purified CEF should be administered to monkeys to verify the implication that the severe toxic symptoms observed with triolein were due to CEF. Relatively little has been done to throw light on the mechanism of the toxic effect. Flick and Gallo in our laboratories have reported that in young chicks showing symptoms of the disease, the intra-cellular water was not changed. Neither were the total blood and plasma volumes altered so that the observed edema was primarily interstitial. Hemoglobin and hematocrit levels were low and blood glucose levels were decreased in advanced stages of the disease. Plasma sodium, potassium and chloride were not affected. In the liver, neutral fat was decreased and phospholipid increased. Preliminary experiments by Flick give some indication of increased membrane permeability, but these must be repeated under much more rigorous conditions. DISSERTATION ABSTRACT A most interesting study in chicks lias been reported by .T. R. Allen. Jr. Part of the results were presented at the federation meetings in March. 1961, and the complete study is available as the dissertation of J. R. Allen. Jr., University of Wisconsin. 1961. I will quote portions from the dissertation abstract [Dissertation Abstracts 22. [2]. 545 (Aug.. 19611 ] : "Experiments conducted to determine the effects of 'toxic fat' on mice, pigeons and turkeys demonstrated a reduction in growth without hydropericardium or ascites.'' In r-hicks. "Microscopic examination of the tissues of the test animals revealed lymphocytic foci in the epicardium and myocardium. Edematous fluid separated the myocardial fibers. Edema of the lungs was a frequent observation in the experimental birds. ". . . Blood pressure indicated the test birds had an elevated average mean pressure in the right ventricle of fi cm. water and '2 cm. water in the vena cava. Electron micrographs of the myocardium revealed shrunken, vacuolatecl mitochondria in the test animals. " 'Toxic fat' produces a reduction in growth rate of experimental animals. This reduction depends on the age of the animal and the level of 'toxic fat' added to the diet. Hydropericardium and ascites are a frequent lesion in the animals receiving from 1.0 to r-i.Qcfr 'toxic fat.' When this level was reduced to 0.25>9f- in the diet, reduced testiculnr development was a more sensitive criterion than hydropericardium. ascites or weight gain for evaluating chronic toxieity. "The mechanism by which 'toxic fat' induces hydropericardium and ascites appears to be associated with degeneration and edema of the myocardial fibers. These data would tend to eliminate the kidneys, liver and endocrines as the primary cause of edema. The early development of hydropericardium, increased venous pressure, enlarged hearts, mitochondria! changes in the myocardium and generalized edema suggest that the myocardium may lie directly inhibited' however, altered capillary permeability lias not been excluded. It is believed that cardiac decompensation and increased capillary permeability act together in producing the excessive extravasculnr fluid collection and the demise of the animal." The wide range of susceptibility within and among species and the variety of toxic effects that have already been noted would make it appear logical that some primary unit of structure and function such as the mitochondrion, may be the target of the toxic factor and that the observed differences may )>p explained by factors such as absorption, specific binding, transport, detoxication, etc.. that determine the local concentration of any substance in a specific site. 237 Another possible mechanism that had suggested itself quite early is that the toxic factor interferes with the normal regulation of electrolyte and water balance. Selye and Stone, back in 1943, described the production of edema symptoms in chicks by certain steroids and the accentuation of these symptoms by increasing the salt intake of the chicks. Alexander has shown that it is possible to produce hydropericardium in chicks by increasing the XaCl in the ration and to prevent its occurrence even with CEF by eliminating XaCl from the diet. Work in our own and other laboratories has illustrated beautifully the interaction of nutritional factors on the susceptibility of the chicks to a toxic agent. For example, the present AOAC bioassay diet for "chick edema factor" is probably four times more sensitive than the assay diet we used originally, although both have approximately the same NaCl content. A point to remember is that hydropericardium as a symptom of toxieity in chicks is not new. In addition to XaCl and certain steroid hormones, chapter 40 iu Biester and Schwarte on Poisons and Toxins indicates that: (1) Zinc phosphide, used as a rodenticide produces "various degrees of congestion with the accumulation of some serous fluid in the pericardial sac as well as in the abdominal cavity in some cases." (2) Alpha naphthyl thiourea, the rodenticide ANTU, shows in poisoned cln'cks evidence of lung edema and excessive quantity of fluid in the pericardial sac. (3) Sodium monofluoroacetate, compound 1080, another very effective rodenticide, produces in chicks distention of the pericardial sac with clear strawcolored fluid, in addition to other marked pathological changes on the heart and lungs. (4) Chlordane. "The primary lesions found in all fatal cases were in the heart. Excessive quantities of fluid were found in the pericardial sacs. . - ." The weed "corn cockle" and several species of Crotalaria produce seeds which are toxic, and in chicks the toxic symptoms include hydroperieardium. However, in each case other characteristic pathology is usually present, and in no case is the purified toxic principle of the same high order of activity as the toxic substances that have been isolated from toxic fats. It is well known now that chicks can efficiently utilize large amounts of fat in properly balanced rations. The use of fat as a standard ingredient of poultry feeds grew as the price of fat calories dropped and became competitive with calories derived from corn. The chick edema disease epidemic of 1957 was a totally unexpected consequence of this growing practice. Many of you are familiar with the story of how the toxieity was associated with a fatty byproduct of stearic and oleic acid manufacture that had been blended with feed grade fats. Very large quantifies of fatty acids are used industrially in the manufacture of lubricants, rubber, paints, asphalts, roofing, chemicals, and to a much smaller extent in foods. Relatively low grades of fat are split into fatty acids and glycerol at high temperatures and pressures, sometimes with the aid of catalysts. The glycerol is recovered and the fatty acids are distilled under vacuum. The first distillate may be used directly as the highest grade of mixed fatty acids or it may be separated by a low temperature crystallization process into stearic (saturated) and oleic (unsaturated) fractions. The residue from this distillation is resplit and redistilled. The second distillate yields a lower grade of fatty acids. The residue from the second distillation is usually suitable for use on highways or in rubber manufacturing, but occasionally it is again recycled to obtain ;i third distillate and another residue. It. was this third residue that had been blended with feed grade fat for use in feeds. Every sample of residue of this type from several manufacturers of fatty acids proved to be rich in chick edema toxieity. Our first impression, therefore, was that the toxic factor was produced during the splitting and distillation steps and that it was concentrated in the residue along with other non-volatile unsaponifiable substances. Closer study of the various stages of fatty acid production soon revealed that the toxic factor was distillable and was present to some extent in the first distillates which were used for the production of the best grades of fatty acids, and those intended for food pur|H).ses. This discovery was made independently and re|K>rted by Ames, et al., \vlu> had found several samples of oleie acid and a monoglyceride made from such an oleic acid to be contaminated. AH this happened just after the passage of the Food Additives Amendment. Tlitv fatty acid manufacturing firms were verv 1*™™*• �238 239 with information and samples. However, they sincerely believed that they were not part of the food business, that the bulk of their production went for non-food industrial purposes. In our visits and talks with their technical people we advised them to study the application of the new legislation to their industry. They realized that a substantial proportion of their highest grade of materials did enter food channels when their customers started asking for guarantees that the fatty acids met the requirements of the Federal Food and Drug Law. At that point the technical committee of the Fatty Acid Producers Council became actively engaged in a study of this problem, and we have enjoyed the whole-hearted cooperation of the fatty acid industry. While they made studies to determine what part of their processes were responsible for the production of the toxic material, studies were continuing on the isolation and chemical characterization of the active substances. Three years ago at the AOAC meeting a symposium on chick edema was held at which reports from the Purina laboratories, the Quaker Oats laboratory, the Procter & Gamble research laboratories and the Food and Drug laboratories described the progress made up to that time. The Merck group had recently entered the problem, and, although they did not report, they also had traveled a similar road. Every step and experiment had to be followed by the chick bioassay. However, during the first year all the groups had made steady progress at about the same rate. First it was demonstrated that the "chick edema" toxieity was entirely in the unsaponifiable portion of the fat. Then, in our work, the unsaponifiable was separated by chromatography on alumina into three fractions of increasing polarity by elution with petroleum ether alone, then with a mixture of ethyl and petroleum ether and finally with 100% ethyl ether. The first, or hydrocarbon, fraction contained 83% of the unsaponifiable, and its ultra-violet absorption spectrum indicated the presence of cholestadiene. The second fraction was characterized as ketonic and infrared absorption spectra suggested the presence of dipalmitone. The third fraction consisted of sterols and oxidized materials. Simultaneous bioassay of these three fractions and synthetically prepared samples of cholestadiene and dipalmitone showed that only fraction 2, but not the dipalmitone, was toxic. Several approaches were tried simultaneously to purify fraction 2. I will not take the time to describe our experiments with a 500 tube counter-current distribution between iso-octane and methanol, or the separation of earbonyls with Girard-T reagent or the consecutive rechromatography on alumina and silane treated Celite. Suffice it to say that as our fractions became more potent we relied more and more on ultra-violet spectrophotometry as an indication of concentration and purity. Two distinct types of fractions were obtained, one with the ultra-violet spectra characteristic of naphthalene compounds and the other of phenanthrene compounds. Both fractions contained the toxic factor. However, the sum of their toxicities was not comparable to that of the starting material in the final purification steps despite the fact that practically all of it was accounted for by weight in the eluates. Since only the fractions that had a significant weight had been tested biologically, the insignificant, hardly visible residue of less than 0.5 nig. in the practically "empty" beaker that represented the cut between the two major fractions was rinsed into a chick diet and to our surprise was very active at approximately 0.1 ppm., representing the most potent material we had obtained. At this stage we had practically exhausted our raw material, and we had to start from the beginning once again. During the course of these studies we had learned that Drs. Portman and Andrus in the department of nutrition at the Harvard School of Public Health had lost a number of monkeys in a nutritional study. They had used a synthetic triolein as the source of fat and had to abandon the experiment. At a Gordon Research Conference I had the opportunity to discuss this experience with Dr. Portman. Fortunately about 40 Ib. of the triolein was still available. It proved to be toxic in the chick edema assay. The triolein was of excellent quality, containing only 0.9% of unsaponifiable material. This unsaponifiable proved -to be the richest source of the toxic material we had ever examined. The toxic fraction was separated from the triglyceride by molecular distillation. The distillate from 17.6 Kg. of triolein was saponified, and the unsaponifiable was chromatographed on alumina to remove the eholestadiene -fraction. The naphthalene and phenanthrene-containing cuts were collected, yielding 670 mg. of material. This fraction was rechrornatographed on more retentive alumina with 5% ether in petroleum ether, and 240 mg. of naphthalene-phenanthrene material was separated. This was rechromatographed in the same system and three distinct types of T7.V. spectra began to emerge in the fractions: Naphthalene (235 m/i), phenanthrene (260 m/t) and a new peak at 245 m/t. These materials were combined and chromatographed on a silane treated Celite-iso-octane column, with 80% alcohol saturated with iso-octane as the mobile phase. The cuts with maximum absorbence at 245 m/t were further purified by two more cnromatographic treatments on alumina at a very high sample: adsorbent ratio (1:2,009 and 1:25,000). The final fraction was completely free of absorption peaks near 235 and 260 nyi. This fraction was evaporated to rtryness, dissolved in a small volume of boiling iso-octane and stored overnight in the refrigerator. White crystals were obtained, weighing 2.6 ing. and representing a concentration over the original triolein of three million-fold. This material at 1 ppm. in the diet killed the chicks in 12 days and produced typical hydropericardium when fed at .05 ppm. in a 21-day test. During the last stages of this work the Merck group [Harmon, et al., JACS 82, 2078 (I960)], announced the isolation of a crystalline chick edema factor from a toxic fat. On July 19, 1960. the Merck group informed us that they had found 47% chlorine in their crystalline material. The presence of chlorine in our material was quickly confirmed by the use of the Dohrmann uiicrocoulometric gas chromatograph. In this test, a few micrograms of sample is injected into a gas chromatograph and the components as they emerge are pyrolyzed at SCO" under oxidizing conditions. The halogen acid formed from halogenated materials is titrated automatically in a inicrocoulometer. Of interest also is the isolation of a crystalline material that had an ultraviolet absorption spectrum identical with that of the toxic material but shifted 3 m/i so that the major peak was at 24S m/i instead of 245 m^. It behaved like the toxic substance in the microcoulometric gas chromatograph, showing a similar retention time and halogen content. However, it was completely inactive in the chick edema test when fed at 1 ppm. in the diet. The finding of chlorine in the toxic substance was a major breakthrough. It was no longer necessary to think in terms of a naturally occurring material that had changed chemically tinder the conditions of industrial fatty acid production. The possibility of contamination with one of many familiar chlorinated hydrocarbons was obvious. Samples of tetrachloronaphthalene and liexachloronaphthalene, kindly provided by Engel and Bell of Virginia Polytechnic Institute, who had demonstrated that these compounds produced hyperkeratosis in cattle, were without effect in the chick edenia test. We have tested a long list of chlorinated compounds including aldrin, clieldrin, lindane, DDT, DDE, BHC, chlordane, toxaphene, methoxychlor, and a series of Halowaxes, without any definitely positive indication. Furthermore these have all been heated with oleic acid at 250° C for long periods and then fed to chicks, with negative results. Also, on the theory that tallow may sometimes be bleached with active chlorine materials which may chlorinate a sterol nucleus, we have chlorinated cholesterol, squalene, «stone and equilenin with negative results. To test the theory that the toxic substance is a metabolite of a chlorinated insecticide we have fed large doses of chlordane, methoxychlor, heptachlor, aldriu and dieldrin to rats for a month and are feeding the unsaponifiable extract of the rat carcass in the chick test. The results to date show that chlordane and methoxyehlor produce no response. INDUSTRIAL EXPERIMENTS Similar experiments have been carried out and are still in progress by at least two industrial laboratories under conditions of fatty acid production. The results so far have been largely negative or at best only suggestive but not clear cut. Dr. Artman has reported that chlorination of naphthalene and phenanthrene by substitution reactions has produced CEF active products. The chlorinated phenanthrene is particularly promising, since preliminary fractionation experiments indicated the possibility that a highly toxic compound was produced. Dr. Boyd O'Dell and colleagues at the University of Missouri observed lydropericardium and other symptoms of chick edema disease in chicks housed in freshly painted cages. They traced the responsible agent down to one of the paint ingredients, a chlorinated biphenyl sold under the name �240 Arochlors of different chlorine content. Some were not toxic; others produce the disease but only at relatively high feeding levels, e.g., 200 ppm. It may be that an impurity in these compounds is the toxic agent, or that they are innately toxic at the high levels fed. Of interest is the use of one of these Arochlors in some insecticide formulations. At the present time there has been no real evidence developed to implicate any product or compound. However, the circumstantial evidence is stimulating considerable speculation and activity. The feed industry has managed by careful control of ingredients to avoid a recurrence of the 1957 epidemic. The color test develoi>ed by Brew, et al., at Purina lias been very useful in screening out toxic fats. This test is not specific but is useful since the presence of large amount of steroidal compounds that give this test usually indicates still residue that may lie toxic. It is useless, however, for fatty acids and other products such as triolein, uumogylcerides, etc., that give no response in tins test, but which sometimes are quite toxic. GEORGIA OUTBREAK Last year, at this time, an outbreak of the chick edema disease occurred in Georgia. Although considerable chlordane residues were found in the feed, we do not believe they were responsible for the symptoms observed. Furthermore, the best information we have indicates that only rendered fat was used in this feed and no product of the fatty acid industry was involved. With each new development the scope of the problem increases. First we were concerned only with still residues, then also with fatty acid distillates and their derivatives. For a long time it was felt that only the fat derived from animals was involved. Recent evidence from sources in the fatty acid industry and our own studies indicates that some vegetable fat sources may yield fatty acids contaminated with CEF. Still another development lias occurred to further complicate the picture. Early this year in the course of our regulatory activities, we examined a sample of oleic acid. All the test chicks died by the end of the second week, with symptoms of severely stunted growth, ascites. jaundice and pathology of the liver and other organs, but with no hydropericardium. This sample had been tested by the manufacturer last year before the adoption of the present AOAC test procedure. The testing laboratory had used the procedure we ourselves had used in our earlier work, and had found the sample negative for chick edema. Repeat of the test in both laboratories by both procedures confirmed both findings. The sample is free of chick edema disease factor when tested on a diet of natural ingredients, and the chicks survive in apparent good health. On the casein-sucrose diet of the AOAC chick edema test, the clucks fail to grow and die early with the described symptoms but no hydropericardium. Furthermore, the dose response curve for this effect is very steep, since the ratio of the dose that gives a maximum effect to the dose that produces a minimum effect is less than two. as compared to a ratio of four to five for chick edema factor. Evidence from preliminary fractionation studies also indicates that this is an entirely different substance. There is no other information as to its characteristics at this time. There is evidence that this toxic contamination has occurred in different places from time to time and in a variety of fatty acid samples and may occasionally occur together with CEF. Here again we must anticipate that this material may occur elsewhere independently of the fatty acid industry. For whatever comfort we may derive, it should be noted that chick edema disease has been observed in England. In a letter to the editor of the Veterinary Record of June 10. 1961. C. C. Wannop of the Houghton Poultry Research Station draws attention to a condition apparently identical with that reiiorted by Sauger, et al., and Schmittle, et al.. in 1958. that has appeared in several broiler flocks. In a i>ersonal note dated Sept. 29, he says that the condition has disapi«?ared for the time in his country. At the present time fatty acids can be used in the manufacture of foods or food ingredients only if they are free from CEF. This requirement made necessary the development of a bioassay which has been accomplished by collaborative work and is now being adopted as official by the AOAC. I have tried to review the history of this troublesome problem, what little is known of its toxicology and its physiological aspects, the speculations as to the origin of the contamination, and attempts to track down its sources. I have sketched quickly our own attempts at isolation and identification of the toxic factor and have alluded to the most recent developments along this line that were reported at the AOAC Section of Fats and Oils by Dr. Artman. 241 BIBLIOGRAPHY ' Chick "Edema" Disease (Hydropericardium) : 1. Sangor, V. L., et al.—J. Am. Vet. lied. Assn., 133, 172 (1958), "Alimentary toxemia in chickens." 2. Edgar, S. A., et. al.—Poultry Sci., 37, 1300 (195S), "Effect of a toxic substance in fat on poultry." 3. Naher, E. C., et al.—Poultry Sci., 37, 1229 (1958), "Effect of certain toxic fats and their derivatives on growth, reproductive performance, embryonic development and health of chickens." 4. Schmittle, S. C. et al.—Georgia Poultry Disease Research Center, Athens, Ga. (1908), "Progress report on toxic fat disorder of chickens." 5. Schmittle, S. C., et al.—J. Am. Vet. Med. Assn., 132, 216 (1958). "A disorder of chickens probably due to a toxic feed—a preliminary report." 6. Brew, W., et al.—J. Assoe. Off. Agric. Chem., 42, 120 (1959), "Characterization of a type of unidentified compound producing edema iti chicks." T. Friedman, L., et al.—J. Assoc. Off. Agric. Chem., 42, (1959), "Studies of chicken edema disease factor." 8. Wootton, J. C., et al.—J. Assoc. Off. Agric. Chem., 42, 141 (1959), "Some chemical characteristics of the chicken edema disease factor." 9. Potter, G- C., et al.—J. Am. Oil Chem. Soc. 36, 214 (1959), "Current .status of the toxic principle causing the chick edema syndrome." 10. Anon.—Feedstuffs, 30, 24, 1 (1958), "American Feed Manufacturers Assn., issues special report on edema in chickens." 11. Wilder, O. H. M., and Dugan, L. R.—Amer. Meat Institute Foundation Special Report (1958), "Progress in certain animal fats relating to a poultry disease." 12. Machlin, L. J., et al.—Poultry Science, 38, 579 (1939), "Relationship of oxidative degradation to toxicity in certain fats." 13. Dunahoo, W. S., et al.—Poultry Science, 38, 663 (1959), "Studies on toxic fat in rations of laying hens and pullets." 14. Edgar, S. A., et al.—17th Meeting, Poultry Science Assoc.. Ithaca, N-Y. August, 1958, "The effect of a toxic substance on fat in poultry." 15. Ames, S. R-, et al.—J. Am. Oil Chem. Soe, 37, News page 10, April, 1960, "The occurrence of the chick pericardial edema factor in some oleic acids and products derived therefrom." 16. Harinan, R., et al.—J. Am. Chem. Soc., 82, 2078 (1960). "The isolation and characterization of the chick edema factor." 17. Harman, R. E., Davis, G. E., Ott, W. H., Brink, IV. G., and Kuehl, F. A.—J. Am. Chem. Soc., 32, 2078 (1960). 18. Yartzoff, A., Firestone, D., Banes, D., Horwitz, W., and Friedman, L.— J. Am. Oil Chem. Soc., 38 (2), 60-62 (1961). 19. Firestone, D., Horwitz, W., Friedman, L., and Shue, G. M.—J. Am. Oil Chem. Soc., 38, 418-422 (1961). 20. Douglass, C., and Flick, D. F.—J. Assoc. Offlc. Agri. Chemists, 44 (3), 449-56 (1961). 21. Shue, G. M., and Gallo, L.—Ibid; 456-59. 22. Flick, D. F., Gallo, I,., Winbush, J., Douglass, C. D., and Friedman, L.—J. AOAC, in press. 23. Allen, James Rex, Jr.—Dissertation Abstracts, 22 (2), 545 (1961). 24. Wootton, J. C., Alexander, J. C., Artman, N. R.—Progress Toward Identification of Chick Edema Factor—presented before the Assoc. Offic. Agri. Chemists, Oct 30, 1961. Washington, D.C. (in press). 25. JlcCune, E. L., Savage, J. E., and O'Dell, S. L.—Pre-publication report, I'niversity of Missouri, Columbia, Mo., "Hydropericardium and ascites in chicks fed a chlorinated hydrocarbon." 26. Ott, W. H., Dickinson, A. H., and Van Iderstine, A.—Poultry Sci., 40, 1016 (1961), "A chick assay procedure for the edema producing factor in toxic fat." OCCUPATIONAL INTOXICATION OCCURRING IN THE PRODUCTION- OF CHLOBOPHENOL COMPOUNDS BY H, BAUER, K. H. SHULZ AND IT. SPIEGELBERG, HAMBURG INTRODUCTION, GENERAL MAY 20, 1961 In tlie last few years, professional work in the production of chlorinated ihpnols has been the r>onca /vf -™ " • �242 243 German chemical works, the acne persisting over a long period and being associated with other effects on the health. Diseases in a group of 17 workers from a company in North Rhine/Westphalia were reported by BAADER and BAUER, as well as by BRINKMANN in 1950/51. The workers in that company were engaged in the production of pentachlorophenol. Apart from comedone acne with various degrees of secondary pustular infection and boils, most of the workers, whilst still in the first stages of the skin diseases, also experienced pain and weakness in the lower limbs, mild paraesthesia, heart complaints and indeterminate psychovegetative disturbances. Subsequent examination of the records of 17 cases1 revealed the following findings: All 17 were suffering from an acne, 4 of these being very severe, 8 fairly severe, and 5 moderately severe to mild. In almost all cases there were extensive pustular infections and boils, 4 with bursitis on the elbow. Other disturbances amongst the workers included 11 cases of bronchitis, 5 of myoeardiac damage, 2 of cirrhosis of the liver (one of which proved fatal), 9 of neuritis symptoms (severe pains in the lower extremities in 7 patients, sensibility disturbances in 4 cases, mild paresis without atrophy in 2 patients, and 2 cases of weakening of the Achilles' reflex). Seven workers complained of physical conditions such as continuous fatigue, depression, lack of vitality, nervousness, slight headaches, disturbed sleep, and decrease in libido and potency. A larger number (about 60 cases, Prof. Hergt) of similar conditions occurred in two Mid-Rhenish companies amongst workers who had been engaged for long periods, generally several years, in the production of trichlorophenol (saponification of 1,2,4,5,-tetrachlorobenzene to 2,4,5-triehlorophenol by treatment with methanolic caustic soda solution). These trichlorophenol workers, like those in a third group of affected persons from the Hamburg region who will subsequently be dealt with in more detail, suffered from further disturbances to health, these often not occurring until a fairly long time after occupational exposure had ceased. In the course of a discussion on a paper by SPIEGELBERG, who referred briefly to our Hamburg cases in a lecture at the 1960 North-West German Neurologists' and Psychiatrists' Congress in Liineberg on psychopathological delayed and chronic damage following occupational intoxication. Janzarik described largely identical disturbances amongst workers from the Mid-Rhenish companies. The third group comprised 31 workers in a Hamburg company. Those of this group who were affected were engaged in the trichlorophenol department of the company, in which the herbicide 2,4,5,-trichlorophenoxyacetic acid was manufactured from technical 2,4,5-trichlorophenol by heating triehlorophenol together with caustic soda solution and monochloroacetic acid in autoclaves. After completion of this esterification process, the end product was purified by double recrystallization. The task of the workers consisted first of all in charging the autoclaves, for which purpose the triehlorophenol in flake form had to be removed by shovel from open barrels. In this operation, a fine dust formed and dispersed throughout the room. Other operations were concerned with filling and controlling centrifuges and regulating feed and outlet pipes. Since it was the workers most exposed to contact with trichlorophenol who suffered from the severest skin conditions, it was logical from the outset to suspect the causal noxa to be present in tie trichlorophenol. The extent to which this assumption was valid is discussed later in this paper in connexion with etilogy. TABLE OP FINDINGS . ™ SOME CLINICAL OBSERVATIONS Of the 31 workers of this Hamburg company, 9 are still receiving medical attention 5 years after the termination of occupational exposure, this beinj due to residues of their acne, chronic neuromuscular weakness of the Ie? musculation, vaso-vegetative lability and, most especially, marked psychopathological disturbances. Details of the established complaints and damage to health are given in the table. The development of the skin conditions in : the patients followed, by and large, a uniform pattern. Numerous comedones formed, first on the face, especially on the cheeks above the malar bones, fore1 Our thanks are due to the Berufsscnossensehaft der Chemischen Industrie fo"placing their records and other documents at our disposal, and also for their nndfrstanding in the sometimes lengthy clinical examinations. * and the �244 Implicit paresis or atrophy, weakening of the reflexes, or absence of muscle expansion reflexes as a sign of toxic polyneuropathy was not established in any of the cases. Two of the patients examined indicated a decrease in sensibility with isolated epicritical disturbances in the lower limbs. No definite signs of neurogenic damage that could have been expected with peripheral nerve lesions were established electro-myographically; premature fatigue, which was recorded in the series stimulus test, requires further confirmation regarding both the method and the raised findings. As these findings show, the neuromuscular disturbances do not fit in with the typical picture of toxic polyneuritis or polyneuropathy. The electroencephalographic examination produced an abnormal electroencephalogram in 6 cases, with frequency lability and dysrhythmic groups of a partly asymmetrical character. In one of the patients examined, accentuated dysrhythmia and raised cerebral excitability were revealed after photostimulation. The electroencephalographic changes found were uncharacteristic and afforded no diagnostic viewpoints of any real consequence. Some of the workers examined complained of headaches, attacks of giddiness and orthostatic collapse tendency. In 5 of the 9 patients examined, there were distinct signs of vegetative hyperexcitability, fine tremor of the hands, increased perspiration on the hands and legs, axillary perspiration, raised dermagraphism and suggestions of Chvostek's sign. The blood pressure value measured during out-patient check-ups were all in the normal region, though at the lower limit of the norm in 5 of the 9 patients examined. Orthostatic collapse tendency was not established either during out-patient visits or during in-patient observation by an internist. In '2 cases, myocardiac damage was suspected. Abdominal complaints such as a feeling of fullness, pressure in the stomach and liver region, and slight pain, were reported by 5 of the 0 patients. There were 4 reports of disturbances in the gastric secretions. 3 of subacidity, one of hyperexcitability and one radiographic finding of gastritis. Very thorough investigations were conducted as part of repeated out-patient examinations and in-patient observation to ascertain any liver damage.3 Whilst the liability reactions were uncharacteristic in all cases, the bromphthalein test indicated slight delay in the dyestuff excretion in 2 instances. In 3 cases, the liver biopsy produced pathological findings, these comprising 2 cases of slight perihepatitic changes and in one case a fatty liver with inflammatory symptoms and slight fihrosis of the liver. Owing to the clinical and histologica'l findings, it was suspected that a condition following virus hepatitis existed in this instance. Deposits of ferrous and non-ferrous yellow-brown pigment were established in this case, although these did not correspond to the grey, non-ferrous pigment discovered by KALK and WILDHIRT in chlorophenol intoxication. The excretion of erythrocytes in the urine of one worker whose renal findings were otherwise normal remained unaccounted for. The psychopnthological changes in the chlorophenol workers who were all psychiatrically and psychopathologically examined were especially remarkable. In 6 cases the course could be observed over a 2-year period and there was an opportunity for objective anamnesis investigation and experimental psychological examinations.4 With a very large degree of agreement, a subjective syndrome of complaints was reported by the patients under investigation, this syndrome extending from the psyehoneuropathic complaints in the region of the extremities, cardiovascular and abdominal symptoms to the mental/ spiritual sphere, especially in the modes of behaviour associated with the vital forces (BiiRGER-PRINZ). Considered in detail, there were reports of disturbances in the vital senses such as general sense of weakness, feeling of fatigue, indisposition, sense of insecurity, inner restlessness and a feeling of illness. The basic mental mood was reported to be deteriorated and lowered towards behaviour characterized by dissatisfaction or sullenness and irritation. Xot infrequently, a mood component of fear and unease was present. Changes in affectivity in the restricted sense of the term were reported by the patients in the form of increased emotional reactions, irritability, tendency to fits of temper and also a certain hebetude. 'Prof. Dr. HORN'BOSTEL and Dr. SCHONFELDER. I. Meet. Univ.-Klinik. Hambiirt k Eppendoff. ' * Our thanks are dup at this point to Dipl. Psychologist W. von SCHUBERT, for con ducting the tests (Rorschach psychogram and " Hamburg-Wechsler intelligence test for adults). 245 General loss of strength and reduced inner vitality and impulsion were symptoms noted in eacli of the cases observed. The probands described reduction in initiative and interests, weak willpower, reduced efficiency, and more rapid exhaustion in physical and mental/spiritual matters. The subjective and objective anamnestic psychopathological picture is further complicated by a number of additional symptoms that are present with a greater or lesser degree of regularity. Disturbances of the instincts occurred in practically all cases. Thus, probands suffered from a sharp reduction in potency, and most also from decreased libido. The appetite had deteriorated ami occasionally there were substantial fluctuations in weight The sleep and sleep requirements of most probands were distributed in that there was an increase in disturbances to sleep itself and concurrently, but less often, to the process of falling asleep. In certain cases, a decrease in mental capacity, especially disturbances of the memory and perception, were mentioned. In the majority of instances, alcohol intolerance was noted. Finally, mention should be made of hyperaesthetic traits with hypersensitivity to light and noise. Marked fluctuations of the cases. psychopathological syndromes described were reported in nearly all Changes were described in the intensity of the symptoms, daily patterns of fluctuation occurring in favour of early morning, late morning or evening hours. In many probands, there were intervals of some days or weeks in which they were practically free from complaints. For completeness' sake, a further two less common phenomena are described. Two probands stated that the general symptoms and the polynetiroiwthic symptoms were relieved for a varying length of time by the use of cold media (cold showers and washing with cold water). One proband reported abnormal, constantly changing eating habits, such as a wish for nothing hut black bread, milk soup, or three litres of milk daily, there being no desire for food of other tyi>es at these times. Compared with the multifarious polysymptomatic subjective pictures, the objective psychopathological signs can be recorded at less length. In exploratory conversations, the majority of the probands displayed a distinct, slightly depressed and subdued mood, which could be brightened only slightly or not at all. So far as impulsion was concerned, the patients examined gave an impression of lifelessness; their psychomotivity was feeble and fatigued. The impression was rather one of slight cerebral organic impulsion reduction than nf inhibition. The affective modes of behaviour were occasionally notable for their reduced reactivity and oscillation capacity, though also because of lability and decoinpensability. In 2 cases, pronounced hypochondria and, in one case, slight but distinct alienation of the total i>ersonality were recorded. The psychological tests are significant for the discovery of finer intellectual performance shortcomings and psycho-organic disturbances in affectivity. In the majority of the probands tested by the Hamburg-Wechsler intelligence test for adults (HAVTIE), there was a significantly raised percentage of degeneration, this providing a certain indication of an acquired decrease in mental caiwcity. In the Rorschach psychogram, coartation of the experiential ty|K', signs of weakened emotional reactivity, poor concentration, reduction iii tempo, sluggishness of the mental processes and a tendency to perservation lioint to eerebro-organically governed changes. DISCUSSION On the basis of the findings in three independent groups of chlorophenol workers, which together included more than 100 affected by diseases, a characteristic clinical picture is provided, the most important features of this lieing the following disturbances: 1. Following initial dermatitis of the face and symptoms of irritability on the part of the conjunctiva: often together with gradually developing acne primarily in the region of the face, then the back of the neck, shojulders and upper trunk, and in severe cases on the entire body, with comedones, pustules, boils and patches of pigmentation. In several cases with severe irritation of the mucous membranes of the face and the upper respiratory tract: sometimes with continuing blepharoconjunctivitis. 2. In several cases, disturbances connected with the internal organs, especially damage to the liver, with deposits of a nonferrous pigment as a �246 247 characteristic biopsy finding. In some cases, chronic bronchitis and isolated instances of myocardiac' damage. 3. In all cases, general fatigue and weakness principally affecting the proximal muscles of the lower limbs, often with pain in the musculation and in some cases paraesthesia and slight hypaesthesia. In isolated cases only, more pronounced disturbances of sensitivity, slight paresis (implicit) and weakening of reflexes. 4. A psychovegetative syndrome with the following disturbances: Subjective: Disturbances of the vital senses, disturbances in the basic mental mood and affectivity, disturbance in impulsion, weakness of memory and concentration, hyperaesthetic traits, vegetative dysregulation, tendency to orthostasis, sleep disturbances, much increased sleep requirements, disturbances of the instinct sphere, reduction in libido and potency, and alcohol intolerance. Objective.—psychopathological: Reduction in impulsion, subdepressive traits of a type characterized by genuine vital moments of depression, disturbances in affectivity in the sense of a certain levelling-out, increased excitability; occasionally hebetude, hypochondria and personality alienation. Experimental psychological. HAWIE : Increased degeneration percentage; Itorsrharh psyrhngriiin : ('u:irt:itioM of tin 1 experiential type, signs of weakened emotional reactivity, weakness of concentration, reduction in tempo, sluggishness of mental processes, tendency to perseveration. The <lenn:ilologie:il picture of the olilorophenol intoxication described shows extensive agreement w i t h (In- disease caused J>y chlorinated aromatic hydrocarbons as first described by HEKXHE1MER and later by several authors (see BRA UN, RISSE-SUNDERMANN). On the basis of the observations that chlorinated naphthalenes were principally responsible, \VAUER, and later TELEKY. suggested the designation "perna disease" (.P&Rchlorinated A'Aphthalene). TELEKY pointed out that the ehloracne already described by HERXHEIMER in 1899 was produced not by pure chlorine but by chlorinated hydrocarbons or the simultaneous action of chlorine and tar. Further observations on perna disease made by MITTELSTADT, FLINN and JARWIK, DRINKER and collaborators, and GREENBURG and collaborators, indicated that not only the skin symptoms but also fatigue, loss of appetite, giddiness, and severe liver damage with acute yellow atrophy of the liver leading to death can result from work with chlorinated naphthalenes. BAADER mentions epidemics at American shipyards during the Second World War. In his description of the cases occurring in America and Great Britain, sometimes with a fatal outcome. TELEKY refers to the report of BROWN. President of the Halowax Co., New York (1937). that only the manufacture of the higher stages of chlorination and the combination with chlorinated diphenyls and other substances led to severe damage to the health and in some cases to fatal acute yellow atrophy of the liver. TELEKY also refers to the animal experiments by C. K. DRINKER and collaborators to support the view that only the higher chlorinated diphenylamines produce serious damage. The general symptoms in occupational chlorophenol intoxications are apparently more pronounced than those occurring with the lower chlorinated naphthalenes employed earlier. This fact was also observed by TRTJHATJT and collaborators amongst workers who had been using pentachlorophenol for wood preservation, as well as KUBOTA in .Tapan. who mentions multifarious disturbances of the autonomous nervous system and who observed several fatal cases. In all three German groups of chlorophenol intoxication, liver damage was established, the damage that was most pronounced and studied most intensely being that found amongst the cases of disease occurring in two Mid-Rhenish companies (HERGT. KALK and WILDHIRT). In all the groups, several cases of chronic emphysema bronchitis and myocardiac damage were found, although these disturbances did not occur nearly so regularly as the pronounced fatigue and neuromuseular weakness, which we observed in all our patients. The psychosyndrome described was equally regular, this being found not only by us but also by .TANZARTK and RICHEUT to a completely identical degree amongst the Mid-Rhenish workers. The psychopathological syndrome could be distinguished with a sufficient degree of certainty by differential diagnostics from endogenic psychosis, especially mild cyclothymic diseases, neurotic personality developments and organic psychosyndromes of different etiology, and somewhat presenile or cerebrosderotic processes of degeneration. Phenomonologically, the relationships to the pseudoneurasthenic syndrome, which is described in connexion with a large number of occupational intoxications such as those caused by lead, carbon monoxide, manganese, thallium, arsenic, carbon disulphide trichloroethylene, etc. (for relevant articles, see BORBELY von HATTINGS BBBG MEGGKNDORFKR, MOESCHLIN, PENTSCHEW, TEoiS-ald especially the relationships to particular endogenic mood conditions—are obvious. Lowering of the vital level, moments of depression, vegetative tymp* toms, and, not least, fluctuations in intensity can be observed predominantly in endogemc-depressive conditions. On the other hand, alcohol intolerance hyperaestheticallv excitable and polyneuropathic traits influence the differential diagnostic aspect more in the direction of an apparently exogenic condition The .somewhat older psychiatric literature should be borne in mind "n tMs connexion (MEGGENDORFER, STERTZ), this placing the neurasthenic syndrome quite definitely in the pattern of exogenic symptom complexes Not 'fc ' % ™eT ? 10Ul,d "e m-ade to the Phenomenological relationship of our obsenations to the (exogenic) hyperaesthetic-emotional conditions of weakness of BONHOFFER, winch, from the psychopathological aspect, have sigmficantly been designated by EWALI) as no longer heteronomous but homonomous in the sense propounded by KLEIST. Despite the phenomenological relationships discussed, the psychopathological di-liiyed syndrome of the chlorophenol workers scarcely corresponds completely n-itli any of the known clinical pictures. In any case, the question of a special psychic-vegetative delayed intoxication syndrome, which was discussed by SPIEGELBERG in connexion with observations on persons suffering chronic occupational damage from military poison gas, also demands consideration in view of the observations mentioned in this paper. It has been possible to rule out psychogenic-neurotic moments so far as our subjects are concerned, provided that individual neurotic conditions ie ciiaraeterogenic and experiential situative data are involved. Two of the nine probands exhibited considerable psychopathic or neurotic structural elements. However, it was easily possible to separate these two probands from the other completely or largely non-neurotic cases. Certain "collective-neurotic" factors have, in our opinion, to be taken into account as an unfortunate but practically unavoidable fact in all group investigations but especially those involving etiological evidence (SPIEGELBERG). Reactions of this type have also been observed in our cases in the sense of a superimposed psychogenic accessory with, as it were, "physiological" but not inadequate, individualneurotic (complex-determined) idemnification wishes. The psychopathological analysis of the individual case and the comparison of the findings in each instance with such independent collectives of the same etiology afford sufficient protection from authoritative and scientific false assessments. Despite the long course, the prognosis of the psychopatbological intoxication results appears favourable. Although technical aspects of the pension situation have not yet been finally clarified, there was, on the whole, a certain subjective improvement in the symptoms, or else they remained static. We have not observed any objective deteriorations, except for the momentary intensity fluctuations. The experience of the Mainz Nerve Clinic (RICHER) and the impressions of works medical staff (KNECHT) suggest a benign course of acute and chronic intoxications, provided no toxic parenchyma damage, as such, influences the prognosis unfavourably. It seemed appropriate to attribute the toxic action to the high-chlorinated chlorophenols, this view being supported by animal experiments conducted l,y MACHLE and THOMAS, H. KITZMILLER, a series of other investigators (KKHOE. DEICHMANN, GRITEBLER, BOYD, McGAVACK, TERRANOVA IMOCIONE cited aec. to von OETTINGEN) and also our own animal experiments. KniMIG and SOHFLZ were, however, able to show that the use of non-industrial, analytically pure, high-chlorinated chlorophenols (trichlorophenols, pentachlorophenols) does not lead to the characteristic symptoms of chlorophenol intoxication. Animal experiments were carried out with a view to discovering the noxae causing the symptoms. The rabbit's ear proved to be a suitable test object since it is i>ossible to produce the changes on this with the substances causing ehloracne, these changes closely resembling those of human ehloracne fHOFMANN and NEUMANN. BRAUN. LANDES, etc.). Brushing with a �; 248 substance which is active in this respect leads at first to patches of dermatitis .in conjunction with reddening, swelling and flaking: then, some days later, hyperkeratosis linked with the follicles and also small cysts occur, these being eas.v to record histologically, as well. In addition to the brushing experiments, tests were carried out on rabbits to determine the general toxicity, whilst cats, too, were used for testing a number of substances. In these tests, it was found, in corroboration of findings derived by OETTEL and also HOFMAXX from similar cases of intoxication in a large chemical works in southern Germany, that the substances producing chloracne possess marked liver toxicity in rabbits. It was possible to trace effectively the damage to the parenchyma of the liver intra vitam with the micro-modification of the bromsulphthulein test given by HOFMANX and OETTEI,. In autopsies, diffuse steatoses and extensive necrosis of the parenchyma of the liver were found. The investigations, which have already been reported (SCHULZ 1956; KIMJIIG and SCHULZ 1057) led to the following results: The effective substances must have occurred in the alkaline hydrolysis of 1,2,4,5-tetrachlorobenzene to 2,4,5-trichlorophenol, this having been carried out technically under pressure at about 180°C in the presence of methanol and caustic soda solution. However, it was not the trichlorophenol itself bat the by-products that formed in small quantities in the course of the pressurized phenol process that were regarded as the causal noxae; for it was not possible to produce any of the above-named changes on the rabbit's ear with pure, repeatedly distilled 2,4,5-trichlorophenol or with 1,2,4,5-tetrachlorobenzene,' although they did occur with the trichlorophenol used technically. Since the isolation of defined compounds from the residue occurring in the distillation of technical trichlorophenol was not possible at first, compounds were synthesized by chemical means and given to us for testing on animals where there was a certain likelihood that these substances may occur as byproducts in the saponification of tetrachlorobenzene to trichlorophenol. The substances initially available were various chlorination products of the diphenyi ether and the dibenzofuran (diphenylene oxide). Although the diphenyi ether and its IX to 4x chlorinated derivatives, and also dibenzofuran and monochlorodibenzofuran were ineffective in experiments on animals, 3X and 4X chlorinated dibenzofurans, even in concentrations as low as 0.05%, produced the symptoms mentioned on the rabbit's ear1. Single doses of 0.5 to 1 mg/kg administered orally produced severe liver damage in rabbits, this leading to the death of the animals in most instances. The clinical observation of a laboratory assistant engaged elsewhere, who fell ill with severe chloracne after exposure to tetrachlorodibenzodioxine. indicated the chlorine derivatives of the dibenzodioxine. Tetrachlorinated dibenzodioxines, especially 2,3,6.7-tetrachlorodibenzodioxine. were highly effective on the rabbit's ear. even in low concentrations. Three brushed applications with 0.01-0.005<X- sohitions (in polyglycol) were sufficient to cause severe areas of inflammation and follicularly arranged byperkeratosis. When administered orally, single doses of 0.05-0.1 mg/kg body weight led to severe liver damage and generally the death of the animals. The assumption that 2,3,6,7-tetrachlorodibenzodioxine is actually of considerable importance in causing the chloracne diseases occurring in the chemical works received further substantial support from the chemical angle. It was possible to prove that this compound is formed from two molecules of sodium trichlorophenolate in association with the cleavage of NaCl under the pressure and temperature conditions prevailing in the autoclave. It was. moreover, possible to isolate the named tetrachlorodibenzodioxine from the by-product occurring in the technical pressurized phenol process (alkaline sanoniflcation from tetrachlorobenzene to trichlorophenol). Tn prove that 2.3.6.7-tetrachlorodibenzodioxine is capable of producing alterations in the form of chloracne not only on the rabbit's ear but also on human skin, one of us (SCHULZ) carried out a test on his own body. Two brushed applj. cations of a 0.01% solution on a circumscribed skin area of the forearm leri within two days to a mild dermatitis, then some days later to a follicular hyperkeratosis and comedones, these also l>eing eas.v to record histologicalh The etiologieal significance of this substance for the diseases described her*, seems to us to be sufficiently evidenced by this experiment. However, it is nr.impossible that other chlorinated aromatic compounds with highly toxir characteristics may occur in this technical process, these possibly not havint been so far identified or tested in experiments on animals. 249 viewpoint should not be iZred H Tt ?« no^nPaV°Dal intoxi^tionS, this attached to themain proS shouW „. ir^le.^prove that impurities N ABBITS ' **"* 3 7 T EA, 2,3,7,S-TETBACHI,ORODrBENZO-p-DlOXIN* E. Linn Jones, M.D. and Helen Krizek, Ph.D comedones and cysts after exposure to ind^tr-i * appearance of papules, EXPERIMENTAL that °f -°-3 Dermatol °^ department of Medicine, University of Chicago ' e t A Division. Office of the n r ct No n by United States Public Health SeVvi?e'M^i £° t ? - »A-49-007-MD-III Presented at the Twenty-third Annual llletine $The * S f F**"* TN°" 2A-5263(Cl" T£ s tolncr. Inc.. Chicago. 111.. J un e 26 iqno °* Society for Investigative Dert With "Improved Zip." ' �250 251 the entire inner surface, and to aid in securing uniformity the dose was divided into three applications, made on successive days. The right ear was ^used as control for some rabbits in each group, and the left for others. Fourteen days after the first application, three biopsy samples, extending through the cartilage, were taken under procaine anesthesia, with a 9 mm. punch. One sample was taken from the middle, one from the posterior and one from the anterior area of the ear, at a level about 15 mm. distal to the notch of the ear. (Fig. 1) The discs were washed free of blood without delay. Under a dissecting microscope at 9x magnification, the moist samples were plucked free of any adhering hairs and clots; and the cartilage was removed with sharp #10 Bard-Parker scalpel. A thin coating of white petrolatum was applied to Hie epidermis. Each disc was floated in a 50-mni. petri dish containing 10 ml. 0.1% pepsin (Worthington Biochemicals "2X Crystallized" product) in 0.24 N HC1 and incubated 4 hours at 37°C., at the end of which period the keratin disc was gently lifted out, free of any undigested dermis by inverting and gently irrigating it, and resuspended in 10 ml. 1:1 V/V ethanol: diethyl ether mixture. A light aluminum-foil cup 9 mm. in diameter, with perforated bottom, previously washed with ether and weighed to 0.02 mg., was brought close to the disc, which was then gently transferred, with follicular projections up, with the aid of a scalpel handle, into the cup. After four hours at room temperature in the covered petri dish, the solvent was aspirated. A similar leaching with 10 ml. ether was made, after which the cup in the covered petri dish was dried in vacuo overnight. The cup and sample were weighed to the nearest 0.02 mg. Additional biopsy samples were taken on about half the animals and examined histologically after routine hematoxylin and eosin staining. The following method was found convenient for preparing a small quantity of 2,3,7,8-tetracblorodibenzo-p-dioxin.1 The sodium salt of 2,4,5-trichlorophenol was prepared by dissolving 1.6 g. metallic sodium in 25 ml. absolute ethanol in a 100-ml. round-bottomed flask, adding 20 g. of 2,4,5-trichlorophenol (Eastman "Practical Grade," recrystallized from petroleum ether) and distilling off the ethanol. The salt was cautiouly heated until the copious evolution of acid fumes which took place at 200-250° C. subsided, after which the flask was kept at 350-400° C. for 30 hours. Two zones were found in the distilling head. The lower of these, a dense, compact mass, had a melting point of 230-300° C. and was only slightly soluble in chloroform. This was the crude acnegen. The upper zone of coarse crystals soluble in chloroform was probably a tetrachlorobenzene formed in a competing reaction. Two batches of the crude acnegen were combined and recrystallized twice from anisole to yield 0.25 g. product. Its synthesis is represented in Fig. 2. Analysis. Calculated for C10H,O2C1,: C, 44.76% ; H, 1.25% ; Cl, 44.05% Found: C. 44.31%: H. 1.40% : Cl. 44.18%.= The melting point was 295-300° C.; literature values for 2,3,7,S-tetrachlorodibenzo-p-dioxin prepared by chlorination of dibenzo-p-dioxin: 295° C. (8), 320-325° C. (9). The infrared spectrum showed a doublet at 1310-1322 cm."1 in the range reported for a series of dibenzo-p-dioxin derivatives (10). The ultraviolet absorption spectrum of the compound in absolute ethanol had a maximum at 233. m/t and one at 307 m/t; the respective molecular extinction coefficients were 46,500 and 4,250. The complete removal of all non-keratinized tissue from the biopsy by pepsin digestion was confirmed by histological section of a keratin disc. (Fig. 10). In general, the keratin discs recovered from the treated animals were less fragile and thicker than the controls. At lower dosages the follicular keratin was usually observed as tree-like forms representing casts of the multiolobated sebaceous glands. (Figs. 5A, 6, and 7). These follicular projections were not present in the controls. (Fig. SB). At higher doses the keratin in the follicles appeared as a larger structure of smooth, oval shape typical of comedones. (Fig. S). Weights of keratin recovered from the biopsy samples are given in Tables 1 and 2. In Table 1 are listed average weights for the three biopsy specimens taken from each ear. In subsequent experiments biopsy specimens taken from the anterior, the middle and the posterior portion were distinguished, in an effort to assess the importance of the site of biopsy removal with respect to the weight of keratin recovered. DISCUSSION For each rabbit there was calculated a value for the relative increase in weight, (T — C)/C, where T = average weight for the three biopsy samples from the treated ear and C = the corresponding average for the control. These values and their average for each dose level are plotted (Fig. 9) against the dose. The averages vary approximately linearly with the logarithm of the dose, but individual (T — C)/C values at each dose deviate widely. The deviation may be due, in part, at least, to failure to secure uniform spreading of the aenegen, in spite of the precautions taken. Loss of keratin during manipulation of the samples may be another source of deviation; however, such an error would be expected to be most important for smallest amounts of keratin handled (i.e. the controls). An inspection of the control weights does not reveal a corresponding spread of values. Further, the probability of losing keratin was greater for the rabbits showing greatest response, because these tended to have some loosely adhering scales and large comedones which might be expressed and lost in the course of tissue removal and subsequent processing; actually, therefore, the spread of response may be larger than our results indicate. In addition, gross observation indicated that some rabbits gave a weaker response than others at the same dose level. We feel, therefore, that the values reflect a real individual variation. Inspection of Tables 1 and 2 TABLE I.—RECOVERY OF KERATIN FROM EARS OF RABBITS AFTER APPLICATION Of 0.3 MICROGRAMS 2,3,7,8TETRACHLORODIBENZO-P-DIOXIN Milligrams of keratin per 9-ram biopsy sample' Rabbit 31 49 47 Treated ear . _ 84 57 KEStTLTS Histological sections showed a characteristic hyperkeratosis of the follicular epithelium and a marked hyperplasia of the surface epidermis. At low dost level some sebaceous cells were present: at high dose level, only a few were seen, and the follicle was filled with a keratiuous mass. (Figs. 3 and 4). Thert was marked thickening of the epidermal keratin, although this was only rarely observed on the slides, presumably because it was lost in the cutting and processing. .... Control 0.98 1.07 1.91 1.88 1.58 1. 13 0.97 0.70 1 31 1.18 1.20 1.31 i Average of three samples from, respectively, anterior, middle, posterior. reveals some tendency for greater response to occur in animals for which con rrol values were high, but this tendency is by no means clear^ut^Vefoundno correlation between intensity of response and weight of the animal There appears to be no consistent difference in the three positions anterior Muddle and posterior, along the line of section, either for the Ztrol or for the treated ear, with respect to weight of keratin recovered. Average value for ker 1 Following the practice of Chemical Abstracts we describe our command as 2,37 8-tetraehlorodibeno-p-dioxin. From a comparison of its preparation with the mode r.f oriein in the Industrial process discussed by Kimmig and Schulz (7) we conclude tha> it is the same as their 2,3,6,7-tetra-chlorodibenzo-p-dioxin. - Analyses by Micro-Tech Laboratories, Skokie, 111. 45-362 O—70 �252 253 The technic developed in the present study might be used for a comparison of acnegenicity of various materials if the test substance is applied to one ear, and 2,3,7,S-tetrachlorodibenzo-p-dioxin to the other. Such a procedure might avoid the complications introduced by individual differences, and reduce the number of animals necessary. Control values could be secured independently from untreated animals. Three biopsy samples may not be necessary; an analysis of the data showed that values calculated from only one (the middle) biopsy sample were not significantly different from those based on the average for three. A fairly good correlation was found between gross observations and intensity of response as assessed by weight of keratin, but there were some deviations. Thus all the animals at 0.3 micrograms dose level were assessed grossly as showing follicular dilatations. However only four showed increase in weight of keratin. At 3.0 micrograms large comedones were observed on four rabbits, and these showed the greatest relative increase in weight of keratin; on one ing topically 0.3 micrograms in acetone, died at the end of 24 days, and two others receiving, respectively, 30 and 2 micrograms died within a week. However, there were undoubtedly other contributing factors, because in all animals treated subsequently no toxic symptoms were observed. TABLE 2.—RECOVERY OF KERATIN FROM EARS OF RABBITS AFTER APPLICATION OF 2,3,7,8-TETRACHLORODIBENZO-P-DIOXIN Anterior Rabbit Middle Control Posterior Average Anterior Middle Posterior Average Dose: 1 microgram 16 15 30 17 11 7 27 1.24 2.14 1.60 2.16 2.04 3.82 3.44 0.54 1.82 1.14 1.52 1.36 2.08 2.86 0.90 1.70 1.26 2.52 1.26 2.52 2.06 0.89 1.89 1.33 2.07 1.55 2.81 2.79 0.65 0.94 0.88 0.80 1.14 1.08 0.88 0.54 0.70 0.76 0.82 1.08 1.08 1.16 0.46 0.64 0.72 0.58 1.14 1.04 0.86 0 55 0 76 0 79 0 73 1 12 1 07 0.97 0.90 0.78 1.08 1.80 0.88 1.24 1.94 0.64 0.90 0.54 0.88 0.64 0.98 1.30 0.92 1.32 0.92 1.26 0.74 0.78 0.90 0 82 1 00 0 85 1 31 0 75 1 oo 1.38 0.88 1.20 0.78 1.32 0.98 1.18 1.08 0.96 1.08 1.16 0.80 1.08 1 28 1.06 0.80 1.20 0.78 1.08 1.16 0 52 1.08 0 88 1 15 0 91 1 07 1 07 0 99 1.07 Dose: 3.0 micrograms 18 19 20 21 22 23 24 0.86 1.76 1.80 7.00 2.82 3.00 5.86 1.22 0.94 1.08 8.54 2.04 2.88 6.14 0.96 0.84 1.58 5.64 1.38 5.86 5.56 1.01 1.18 1.49 7.06 2.08 3.91 5.85 1. Hyperkeratinazation induced on the rabbit ear by the aenegen 2,3,7,8tetrachlorodibenzo-p-dioxin is studied by a new teehnic based on weighing keratin recovered after careful pepsin digestion. 2. When applied in acetone solution to the rabbit ear, 2,3,7,8-tetrachlorodibenzo-p-dioxin is effective at microgram levels. Effect of dose and individual differences in response are discussed. 3. The new technic, using 2,3,7,8-tetrachlorodibenzo-p-dioxin, is suggested for comparing acnegenicity of various substances. REFERENCES Milligrams of keratin per 9-mm. biopsy sample Treated ear SUMMARY Oose: 10.0 micrograms \. ADAMS, E. 31., IRISH, D. D., SPENCER, H. C. AND ROWE, V. K.: The response of rabbit skin to compounds reported to have caused acneform dermatitis. Inc. Med., Ind. Hyg. Section, 2:1, 1941. 2. HOFMANN, H. T. AND NEUMANN, W.: Eine Methode zur tierexperimentellen Prufung der Hautwirkung chlorierter Naphthaline. Zbl. Arbeitsmed., 2: 169, 1952. 3. BRAUN, TV.: Chlorakne. Editio Cantor, Aulendorf i. Wurtt, 1955. 4. HAMBKICK, G. W., JR. AND BLANK, H.: A microanatonical study of the response of the pilosebaceous apparatus of the rabbit's ear canal. J. Invest. Derm., 26: 185,1956. 5. HAMBBICK, G. W., JR. : The effect of substituted naphthalenes on the pilosebaceous apparatus of rabbit and man. J. Invest Derm., 28: 89,1957;. 6. SHELLEY, W. B. AND KLIGMAN, A. M.: The experimental production of acne by pentand hexachloronaphthalenes. A.M.A. Arch. Derm., 75: 689, 1957. 7. KIMMIG, J. AND SCHVLTZ, K. H.: Berufliche-Akne (sog. chlorakne) durch chlorierte aromatische zyklische Ather. Dermatologica, 115: 540, 1957. 8. TOMITA, M., UEDA, S. AND NASISADA, M.: Dibenzo-p-dioxin derivatives. XXVI Synthesis of polyhalo-p-dioxin. Yakugaku Zasshi, 79:186, 1969, Chem. Abstr., 53: 13152, 1959 9. SANDERMANN, W., STOCKMANN, H. AND CASTEN, R.: TJber die Pyrolyse des Pentachlorphenols. Chem. Ber., 90: 690,1957. 10, NARISADA, M.: Infrared absorption spectra of aromatic ether compounds. II. Characteristic bands of dibenzo-p-dioxin derivatives. Yakugaku Zasshi, 79: 183,1959; Chem. Abstr., 53:10967,1959. DISCUSSION 32 33 34 35 36 37 38 3.28 2.68 1.26 6.56 5.14 1.64 6.28 3.30 2.48 1.62 7.12 2.90 1.34 4.02 2.02 2.60 1.18 5.84 3.84 2.18 4.90 2.87 2.59 1.35 6.51 3.96 1.72 5.07 other, comedones were observed; but the response in terms of weight recovered was less than for two which gave the gross impression of follicular papules. Attention should be called to the great toxicity of 2,3,7,8-tetrachlorodibenzop-dioxin. Kimmig and Schulz (7) reported that 0.5-1.0 mg. per kg. orally was lethal to most of their rabbits. In our preliminary experiments a rabbit receiv- DR. PETER FLESCH, Philadelphia, Pa.: Since the criterion of acnegenic activity appears to be the conversion of the sebaceous cells into keratin-forming ceils, I would like to ask, what did you see in the histologic sections? DR. E. LINN JONES, (in closing) : In the paper we will have histologic sections of treated glands and the digested keratin disc. In regard to histology there is varying response, depending on the amount applied. With lower closes there is conversion of the cells in the follicle to keriitinizing squamous cell.s, with occasional remnants of selbaceous cells in pockets here and there. With larger doses no sebaceous cells can be found. Tlie entire follicle is converted into a large keratin-filled papule. �STUDIES OF THE CHICK EDEMA DISEASE 2. PREPARATION AND BIOLOGICAL EFFECTS OF A CRYSTALLINE CHICK EDEMA FACTOR CONCENTRATE BY D. F. FLICK, D. FIRESTOXE AND J. P. MARLIAC Reprinted from 1'ori.iKV STII;.V<-E: Vol. XLIV, Xo. 5 September, I<365 (255) �256 257 3,000,000-fold from a low potency commercial triolcin produced hyclropericardium (HP) when fed to day-old chicks at SO p.p.b. in the diet, and resulted in death al 1 p.p.m. Wootton and Courchene (1964) found one fraction (designated a 3.02) which was highly toxic for the chick. These workers estimated that ingest ion of 5 jj.g. of the a 3.02 fraction was enough to kill one chick. Firestone et al. (1963) reported that signs of the disease were elicited by one fraction fed at 0.1 p.p.m., which substantiated the report by Yartzoff et al (1961). The purpose of this paper is to report our recent studies on the purification and biological effects of a concentrate of the chick edema factor (CEF) isolated from a crude toxic fatty material (TFM) known to produce the chick edema disease (Flick et al., 1962, 1963). Preliminary data are included on the effects of CEF-conlaining material on egg hatchability and develop" ment of the chick embryo. Studies of the Chick Edema Disease 2. PREPARATION AND BIOLOGICAL EFFECT? OF A CRYSTALLINE CHICK EDK.MA FACTOR CONCENTRATE D. F. FUCK, 1). l-'jKr.H-ioxr AND J. P. MAKUAC Di-t !>;,;;;> of .Vi.''; :'/o;;. Food Ciiri,i!>!;'\, and Tnr;c<.>l:>?u'.'<! Kvalu'i'.iox, I'oOil a;;tl />r,- /• Administration Wellington, D.C. 2d204 ( R n c i v i . ! for p!iV,:c.L!i. INTRODUCTION" \OLLO\Y1NG the early icports of the chick edema disease (Chi)), Sangcr it ul. d'J.i'S). Putter ct al. ( 1 9 5 0 ) , Allen (!96r> and AlU-n and Lalich t'1%2) reported toxic->U'.;ic,-t! >tudie- wluVh i-.-nhlished il::il the di.-c.i?f differed from known poultr\ di-e.s-e-. \ number of iu\e~iigutors reported </:; i';i punfica!!o;i, isnl.niun and partial Lhei,ii<.,il chararit'rix.atinn c.f toxic factor i, 1'i'ew it <•''., 195'J; I-neuman ct al., F 1959; \Vuot!<i:i and Alexai. ler. 19:59; Ames it at., 1960; Ilarnian it aT , K-60; Yartzoff ct a'., 1961; \V<>u!i<>n if a!., 1962; ami \Vootiun and'Couruiene. ?%•!). Krew ct al (1959) repr-ned that broiler chick.-- developed tl\c dis'-.'so when fed a fraction purified 3200-foH from the original ftartini; material. A fraction purified 10.000-ftild v.hirh elicited i "y.D v,a> reported by l - i i i - d m r s n ct <:!. (}''J:>'-)). Yj.rtisoff (/ <•.'. (19ol) rcpcrttd tliat p fraction purified phcnanthrene derivatives were collected, combined and concentrated. The concentrate was chromatographpd on columns of Cclite:H,SO,:fu/nii]g H,SO, ( 1 : 1 : 1 ) and eluted with CC1,. The fnregoing procedure was a modification of AOAC method 24.111 (a) (Horwitz, I960). The CCU eluates were re-chromalographed on Merck Alumina, and individual fractions were collected and checked by ut'raviolet spectrophotometry. Fractions eluting v.-ith W7C diethy! ether with absorption spectra of naphthalenes (absorption maxima in the range of 240-250 m;j..) were combined and re-chromatographed on Merck Alumina with i=ooctane as the eluant. Individual fractions were collected and bioa.-sayed, and a highly toxic crystalline fraction (CEF concentrate) weighing 79 nig. was obtained. The CEF concentrate was examined by microcoulomelric and electron capture gas chromalography. A microcoulometiic gas chromatograph (Dohrmann Manufacturing Company, Palo Alto, California) was used MATERIALS AXD METHODS at a column, temperature of 250"C. with a Preparation of CEF Concentrate. Eigh6-foot X y\ inch (i.d.) aluminum colteen pounds of unsaponifiable malerial was r isolated from ISO pounds of toxic fatty umn packed with 2Q /v Dow-Corning High Vacuum Grease on acid-washed Chromomaterial (TFM) by saponification and exsorb W. Details of this technique were detraction of the unsaponifiable fraction with scribed previously by Firestone et al. petroleum cthendiethyl ether (1:1, v./v.). (1963). For electron capture detection, an The petroleum ether was redistilled and Aerograph Hy Fi (Model 600R) gas chrothe solvent boiling between 40-60°C. was collected and used. The unsaponifiable malograph (\Vilkins Instrument and Research Inc., Walnut Creek. California) was fraction was chromalographecl on Fisher Alumina (Cat. Xo. A 540) essentially as used. This instrument was equipped with a tritium source electron captu>v detector. A described by Yartzoff ct al. (1961). Fracstainless steel column, 5 fei-f X ,'x inch, tions with ultraviolet absorption spectra r that matched those of naphthalene and packed with $ /o SE-52 silicone gum rubpenanthrene derivative? were combined and ber on 60-80 mesh Chronmsorb ^-, was chroniatographcd on Merck Alumina (Cat. u.<ed at a column temperature of 202°C. No. 71707). Foreruns were eluted with and a nitrogen (carrier gas) flow rate of petroleum ether. Additional fractions that SO ml./minute. eluted with S% diethyl ether and that exhibited the spectra of naphthalene and Feeding Procedure and fHor'-fmictif Methods. Day-old Single Comb \Viu'if Leghorn �258 TABLE 1.—Composition of basal ration Component Lcve! Corn meal, yellow \Vheat flour middlings Alfalfa meal Linseed meal J'rotein, sova (Drackett) Salts, A.O.A.C.1 Yeast Nad, iodized' Cod liver oil Cottonseed oil % (W/\V) 36 15 1 X lf> 3 3 1 1 16 100 1 For composition of A.O.A.C salt mixture, sec report by Flick ci ol. (1963). ' This addition gave a tola! of 1.99CJ. N'aCl in the diet. Hatcliability Study. For this, preliminary study, commercially available White Leghorn fertile eggs were injected with the test materials into the yolk r-ac prior to incubation according to the technique described by Mclaughlin et al. (1963). 2 Control eggs were injected with 0.1 ml. corn oil. Experimental eggs were injected with 10, 20 or 50 jil. of the unsapenifiable fraction of crude TFM from which the crystalline CEF concentrate was prepared. Eggs were candied each day after Jhe 5th day of incubation. Dead embryos were removed and examined for gross malformations. After incubation, unhatchccl eggs were opened for gross observation. Chicks which hatched were observed for 3 days. cockerel? were led ad libitum and provided with fresh water at all times. They were handled and maintained as reported preRESULTS AND DISCT-SSIOX viously (Flick et al., 1963). GLC Analysis of CEF Concentrate. Gas The composition of the basal diet used is shown in Table 1. The crystalline CEF chromatographic separation? of crystalline concentrate, dissolved in chloroform:ether, CEF concentrate were obtained by usin" was added to the cottonseed oil. and the oil both the microcoulometric and electron was \varmed with stirring to remove sol- capture techniques. A typica' electron capvent. The CEF concentrate was fed at lev- ture chromatogram of our purified preparaels of 50 and 200 p.p.b. for three weeks. Body weights and feed intake were de- to Dr. E. \V. Rice, Presbyterian Hospital, Pit'stermined and. at weekly intervals chicks burgh, Pennsylvania for determination of copper : Injection directly into the yolk has b:en were anesthetized with diethyl ether, volstudied in over 70,000 C£cs (persona/ commun'caume of hydroperfcardium (HP) was meation from Dr. Mcl.auchlin) \vi'h approxiimtclv sured according to the procedure outlined 300 chemicals. India ink and various dyes were by Douglass and Flick (1961), and post- injected in conjunction v/ith several different mortem observations were recorded. Blood studies InjecU d materials were found to continsamples were withdrawn from the right ven- ently remain in the yolk and rvo .creat difftcuUie: tricle with needles and syringe? moistened were encountered \rhuh were not amenable to correction. with heparin, and the following tests were The unfapor.iftable fraction K- a liquid residue performed: microhematocrit (Xntclson, of the t n \ i c fatty material v.'hich is a sti'I b,>: 1961). whole blood glucose (Somogyi. lorn rt-=ifli:e resulting in the corii:n?rcial produc1952) and total plasma proteins and plasma tion r of fats The un-aponifbb'e 'r.Ktion comprised protein fractions (Gornall ct al., 1949). 101 f of the crude starling material and contained hydrocarbon res ; '3ii'S an/J fatty acid degraOther blood samples were allowed to clot, dation products in addition to the chick odcm-1 and copper determinations were made on facto.-. U'hc'i centrifu::orl 100C* X ?, for IS mir,. the sera. utes at 2-!°C, the unsaponi:"vblc material cm; ' SampVs of serum \vere collected in acidrin.-cd tubes frozen under carbon dioxide and sent 259 taincd <0 59c fv.AO of p:ecipitable material Sje Brew cl al. (1950), Friedman et al. (11)50 and \Vootton and Alexander O<'5°). MINUTES are retention limes relative to aldrin. tion is shown in Figure l ( a ) . The chromatogram revealed that within a 70 minute run at Jeast eight components were present in the preparation. Woollen < / al. (1962) isolated two components, which were desi"nated ar 3.02 and 2 3.17 (gns chromatographic retention time relative to methyl arachiclale). Each of these two components contained a high-melting fraction that produced edema and a low melting fraction [hat did not produce edema. Our "toxic concentrate was chromafographically compared lo the low melting fractions of Wootton cl al.2 The peaks of (he hitter two compounds, obtained with Ihc- electron capture technique, are shown in Figure l(b) and 'Kindly supplied by Dr. X. R. Artman, Pr~c~ tcr i- Gamble Company, Cincinnati, Ohi >. - r - <« -«» * to chro^-.^apluc pt,aks l ( c ) . The chromatogram of cm CEF concentrate shows a peak fretcnliun time of 10.5 relative to aldrin) ha\:,ig the same retention time a? the z3.02 ir:'dive i^omer. The relative peak area (ri of total area of the chroma!ographic peaks) uf the 10.5 comjionent was estimated by u.-ing the retention X peak height mcth -il of Carroll (19(51). By this method the !0.5 peak obtained in Ihe niicrocoulomc-tic chromatogram represented about 20', cf total components. Chick Response.. Results obtained on weekly weight gain, food consumption, feed/gain rnlio and calculate*' consumption of the CEF are tabulated in Table 2. The weight gains among control '>. : ni> increased at a fairly constant rate eadi \.ccfc. Weight �260 TAI;I.E 2. CEIIcvel 261 If cchly u'dght gain, feed consumptii.w,_fr.ed/£ain ratio find ingcstion of crystalline Cf.F concentrate p.p.b. Diet consumed1 g. Xo. of chicks Mean weight gain1 g- 0 50 200 52 47 47 27. 4 + 1 . 12 29.8+1.4 28.0 + 1.4 0 50 200 39 36 38 36.7+1.6 29.3-1-2.0 39.0 + 2.2 0 50 200 23 20 22 58.4 + 2.1 27.5i3 0 -!0.6-2 7 Xo. of deaths 1'ced/gain CEF concentrate consumed ratio Total ;/g. ;ig./100 g. BAY. Week J 38.5 42.0 42.7 0 0 0 Week 2 82.6 80.5 72.1 tt'cck 3 139.3 112.0 133 7 1.41 1.41 1.53 0 2.1 8.5 0 0! 1 2.25 2.75 1.85 0 4.0 0 4.1 14.4 14.4 0 0 3' 2.39 4.07 3.29 26.7 0 5.6 0 3.2 13.1 0 4.4 IS. 3 value- ''cace i'S-10 chicks/cage in 5-6 cages/treatment). f t h e m^ii This cb i t k Jk-il on l l l h tlay with edema. ' Thc?c chicks were severely c-dcmatous. 3 gains among birds led 200 p.p.b. CEF were similar to those of controls for the first two weeks hut \\ere less than those of controls for the third \veek. The birds fed 50 p.p.b. CEF grew less during the second and third weeks than the controls or the birds fed 200 p.p.b. Weight sains were equivalent among all groups during the first week. There were a few deaths, all of which were associated with feeding CEF concentrate (Table 2). The chicks that died had edema. Birds with severe edema always had gross organ changes. These pathological changes were not so severe as those observed among chicks fed the crude TF.M (unpublished observations). This observation may be an indication that the purified CEF preparation has been at least partially separated from substances in the crude fat that not only enhance edema formation but also may It-ad to extensive and more severe pathological changes. From the daia in Table 2 on diet consumption, it :.- app.Ti-nt that the CEF concent rale did i-'it alter the amount of diet inji-skd during the fir.-1 wick. C'EF concuitnih: ai 200 p.p.b. (.au.-rd a moderate de-crease in dic-l consumption during the sec- ond week, and at 50 p.p.b. led to a decrease in diet consumption during the third week. The feed/gain ratio was elevated during the second week among birds frd 50 p.p.b. CEF and was further ir.creascd during the third week. The feed/gain ratio for the birds fed 200 p.p.b. CEF concentrate was elevated during the third xveek but it was lower than the ratio for (he birds fed 50 p.p.b. From the data, however, it is apparent that feeding of the concentrate was accompanied by a moderate decrr-ase in feed utilization. From determination of feed intake and knowledge of the amount of crystalline CEF concentrate incorporated into eacli diet, the amount of puriired material ingested was calculated. These ca'cuiations are given in Table 2. The lota! y.g. CEiconcentrate consumed was H.7 jig./chicl; for those fed p.p.b. and 49.6 ^.g./cbick for those fed 200 p.p.b. The amount of CF.K concentrate ingested in ng.,/100 g. of boch weight/chick is shown also in Table 2. Tintotal amount of purified concentrate ingested (-10.6 ;j.g.) was about ninefold gren!er than the amount (5 y.«.) that Woolli.;-. c/ <.'/. (19f.2) calculated could kill one \Bi.E -V Wcfkly lr\r/* c>f j-^i^nm [Tflctn: S. C. M'/iilc i,'-f;t.'>rn (tifLudxfnl CM concanr-ilc Flick ft at. (190,3) that I ho mechanism of CEF loxicity resulting in edi-iiu formation may be the effect of CEF en capillary perTula! I>rot( in meability, and perhaps moro specifically, p.p.b. on causing chemical altcniti'.r,; in cndotheI . T ' . - . l j 1 . 2 ; - 12 J. 951.10 lial intercellular cement siili.-';i:ici?, as posWot / 0 •2.03±.fl? I. (Hi-'. II J . J I i . O ' l . . . . . ( - ' . II 50 tulated by Allen (1961). liven though I . I S - ^ . I W I . 3 S . 1 2 1 . 47 - . 1 0 3 . 3 7 ± . 0 7 200 1 . 9 4 4 . 1 3 I . 2 3 - . ' J > 1 . . W - . 1 9 .5.2-U--OK there was no appreciable dc;j:v--ion of cirHVi-4 I 0 2. 21 ! ' . I ! 1.07 - . 0 1 J . O S ' . I I 3 . .'1^.11 50 culating albumin, at lca>! ->j;v:dcm to re2.5! .i'.. .. .VSi .fio 2. (•-<" - I f i J.50-t .1<J 200 2. 12 ' . 1 7 0.9; ' .OS 2. 2'' - . 2 ! sult in hypo.'ilbuininernic ed< "Vi (Host and r . 0 2 . 2 " > - . l t !.,!"-.)! I . 7 ( . - . 2 0 . l . f i S i . I l 50 Taylon, 1955; and Smith and Jones. 1 . 9 V . - . 20 l . ? T ± . . ! l 1 . I 7 - - . . S !.$&-<-. 37 200 I . ' J ( . . L . 1 6 : . r ? i , 1 2 1 . 7 7 r . . 20 3 . 1 5 . . ? t l 1961), it may be that even ^,'i;)it decreases ' S.K. of Ihe mean. in albumin could enhance .';:id (ransfer across cndotbc-lial membrane- damaged by chick. The diffcri-nce between these value- the chick edema factor. may be due to a number of rea-ons. First. The degree of development. o> IIP in reWootton cl a!. (1962) did not report the sponse to the ingest ion of the- CEF concenamount of food ingested by their birds fe 1 trate is shown in Table -1. 'Jh u>ncentrate the x 3.02 fraction. Second, they used a fed at the 200 p.p.b. level prc.-.lua-d a marheavy breed of chick, while we have used ked increase in fluid \vlvcl~- ."(cumulated a light breed. Third, though both diets uere within the pericardia! sac. Tin.- magnitude synthetic, they were not identical. Fourth, of thi? response is of the sair;.' order as retheir x 3.02 fraction w as essentially homoported previously by Flic!-; <.: •:!. (1063). geneous, while our preparation contained using 4f,o crude TF.M in v similar diet. perhaps as many a.- eight components The Ml' group scores .-how t h a r f he 50 p.p.b. (compare a, b and c, Figure J). Xemlevel of CEF did not result in abnormal thelcss, the prediction of Brew cl a!. volumes of pericardia! fluid according to (1959) that the chick edema factor may the ranges of volumes- reponod by Shtie he detectable when fed to chicks at levels and Gallo (1961). The H;- incidence of 2.0 p.p.m. or less has been confirmed shows that all the birds fed 2C-0 p.p.b. CEF by our studies and bv those of Wootton had abnormally high levels o" fluid in the • I at. ( 1 9 6 2 ) . pericardia! sac. Also shown i**. Table 4 are The results obtained for plasma albuvalues for hematoirit, serum copper and min, globulins, A/G ratio and total prowhole blood glucose. The packed cell volteins are presented in Table 3. In general, umes were decreased by the i\.'o levels of •.here were no appreciable differences in eiCEF concent rate fed. These decreased liether the plasma protein fractions or the matocrils were not ?o Itnv as ti- osc icji^rted total plasma proteins when cnnsidi-rcd on previously by Flick ct a!. '• J963) from •ho group basis. Some chicks with severe feeding 5f/c crude TF.M wit!? a practical edema had mprkcd hypoaiimminemia and type of ration. hypoproteinemin, in confirm;. I ion of the Peruni copper levels (Table 4) were eleob.-ervalions of Alexander cl a!. (1962). vated at the 50 p.p.b. level but no statisti1'rum the data it sr-tnis unlikely that the cally significant change froir. nnrm.nl oc;-:;ix<ivc edema formed was (h,, resuit of curred at the 209 p.p.b. level o'" CEF concen-i;cii modest alterations in circulating protrate. Since it had been ob.KV-. cc! that the :r",f. These data suppoit the hypothesis of liver, heart and kidney were ;i'.Ivcrstly in- �262 263 TAULE 4.—Ilydropericardium (///*) t krwaiccrit, smtn: cn'.'ficr and '.ch"Ic II(tod glucose values of S. C. \\'liiic Leghorn cockerels fed C"/-,/-" concentrate for 3 wr/rs CEF level P.Tl.b 0 30 200 Hydropericardium Mean volume ml. 0 .060 + 0.012 (S)' 0 liS + 0.018(4)' 4 -OS + 1.64(6' Score1 0 0 30+ - llematocrit Incidcnce o/s 0/5 6/6 9; cells 3-i..211 .5(5) 28..0 + 0.8(8) '21 .1 + 1 .4 (S) Serum Copper ^g. r.c 51 + b(~) 95 + 1 1 id) -11 ± 8(5) Glucose5 Unsap. injected No. eggs injected Obscrvatio ns Natch Embryo "'S- % 226+11 (8) 240+ 7(8) 217 + 10(8) 1 IIP score determined from HP volumes according to the fnllmvin^ =calc: <0.20, 0:0.21-0.40 1 + ; 041-1 00 2^ • 1 01-2.00 3 + - 2.01-3.00, 4+; 3.01-4.00, 5 + ; 4.01-5.00, C + ; 5.01-6.00, "-(-; and >O.OI, 8+. 2 These k vela ;1re considerably higher th.ln giiioo.-c lc\ i-ls among bird? fed crude Tl- M in a purified ration (Flick ft a!., 1'Xo) in another experiment \vhich resulted in lather mar!;ed liypoglycemia (controls: 146 + 5 nig. '',', is. severe disease: 110 + 8 mg. '/<-)• 1 S.K. of the niir.n (number of observations). * One sample of HP fluid lost from this £roup. volved in the chick edema disease (Allen, 1961; Alexander ct a!., 1962; and Flick ct a!., 1963), and thai changes in these organs had been involved in the mechanism of edema formation (Smith and Jones. 1961), it was thought that determination of the levels of serum copper in the birds with the disease might be of some diagnostic value. Scheinberg and Sternlieb (1963) reported that human patients with Wilson's diseasehad severe liver disease, and many patients had serum copper dyscrasias caused by excessive urinary excretion, malabsorption from the intestine, decreased protein synthesis (particularly in severe malnutrition) and severe hepatic dysfunction. It may be that the elevated serum copper among the chicks fed 50 p.p.b. CEF concentrate was associated with either decreased liver utilization of copper or decreased renal excretion. The advei>e effect on serum copper of CEF concentrate fed at SO p.p.b. is not clear, but may be indicative that CEF toxic activity is oligodynamic and more specifically oligofoxic (more toxic at low levels than at higher levels). Whole blood glucose levels were not appreciably altered by feeding the concentrate (Table -1). Crude TF.M, however, fed in a purified ration, frequently elicited a marked hypoglycemia among chicks in ad- TABLE 5 . —l:.jfcct oj CKf'-eoiilaiiiiug ttnsti/iniiijiablc fraction of TPM on embryonic dmlopmenl and egg hatcliobililf vanced stages of the disease (see footnote 2, Table 4). From what is known !<J date, the chick edema factor elicit? a number of signs of intoxication which not only accompany the feeding of crude TFM but are more severe when the crude maleiial is led. The finding that our CEF concentrate contained 8 peak? by electron capture gas chromalography (Figure 1) indicates that at least 8 compounds were present in the purified preparation fed in these studies. It may be that only one, or a few, of these compounds possess the necessary molecular configuration to produce signs of the disease equivalent to the estimated potency of the y. 3.02 fraction. Ilulchability Study. The preliminary results obtained from injection of White Leghorn eggs with the unsaponifiablc fraction of TFM, which contained CEF, are shown in Table 5. The percent hatch of control eggs was within the expected range reported by McLaughlir. ct. al. (1963). Chicks which hatched appeared to be normal in size and development by gross obs-,-rvation. When fertile ejr.;s were yolk-injected with 10 ;/.!. of the inuiilutcxi unsaponifuible fraction containing CEF, the hatch was 40'p; injection of 20 p.!. resulted in «I- 1 O 10 20 50 20 20 20 5 Chick ("Malformations of right < cerebral hcinip]>hiTes, legs [and beaks; small embryos Normal [AVeight retardation, jsparsc »nd deformed (feathers. Gf /O 932 40 20 0 • Jfo™»i wpcctaf % hatch (J;cUu R hluT™< 1963). 20% hatch and 50 ;;.!. completely inhibited hatching. Embryos which failed to hatch exhibited one or more of the following developmental anomalies: malformed beak, lack of development of the right mesencephalon, eye defects, growth retardation or leg deformities. The deformities observed were not studied further. The deformities found were common to the embryopathies which occurred and were not related to lex-el of CEF-containing fraction injected. Embryos which hatched after injection with 10 or 20 ;J.1. of unsaponifiables exhibited sparse and defective. feathering (down) and were small compared to the controls. This study revealed that CEFcontaining unsaponifiahles are capable of interfering with normal embryonic development and hatchability. SUMMARY The following studies were performed: gas-liquid chromatographic (GLC) separation of a purified crystalline concentrate containing the chick edema factor (CEF) feeding of the concentrate at SO or 200 parts per billion (p.p.b.) in a semi-synthe'ic diet to day-old S. C. White Leghorn cockerels for three weeks and determination of growth, feed intake, feed/gain ratio, mortality, total intake of CEF concentrate, plasma proteins, hydropcricardiuin (HP), hematocrit, serum copper, whole blood glucose; injection of CEF-containing unsaponifiable fraction and determination of its effects on embryonic development and on egg hatchability. The following observations were made: 1) presence of approximately eight components (GLC separation) in crystalline material, 2) moderate growth depression at 50 p.p.b. level of CEF concen5rate, 3) essentially normal feed consumption, 4) moderate!)' increased feed/gain ratio, 5) increased HP at 50 p.p.b. level of CEF concentrate, 6) severe HP at 200 p.p.b. level of CEF concentrate, 7) essentially no group changes in plasma protein?, S) moderate decrease in 'hematocrit, 9) no change or moderate elevation of serum copper (50 p.p.b. CEF concentrate), and 10) normal blood glucose levels. A preliminary hatchability study revealed that a CEF-containing material led to a decreased hatch of injected eggs and to development of embryonic deformities. ACKNOWLEDGMENT The isolation of CEF concentrate was largely carried out by Mr. .\p.c!rc'iii.-l'art^ ^Q.^.y.a;I,,;i;aiSe*ij«tt»fttograr)hy was perjfrormcd l\v.^fr._Gannelt R. )fiqgmliothiim. u R?vt?irTiT'of l^d^ C^mMy^footTamT Drug Administration. REFERENCES AlcNandtT, J. C., R. J. Voiins, C. -M. Burnett and II. O. Hal-haway, 1962. HydroiiericarA'um �264 Brink and F. A. Kuehl, Jr., I960. The isolaassay nnd Mfcty of Fats and f a t t y acid prodtion and characterization of the chick edema ucts Poultry Sci. -it : 22-32. faclor. J. Am. Chcm. Soc. 82: 207S-2079. Allen, J K., Jr, 1061. The role of "toxic fat" Horuitz, \V., Kditor, I960. Oftai.i! Methods of in the production of hydro;H':'icardium and Analysis Association of Official Agricultural incites. D ' K t o r a l Thes's I'nivcrsity of \VisconChemists Washington, D. C., 345. sin, Mcl.aushlin, J., Jr., J. P. Marline, M. J. Allen, J. R . and J L. Laiich, !«(•:. Response of Verrett, M. K. Mulchlcr and (). G. Filzhugh, chickens to p'olon^cd fecdin;: of crude ''toxic 1063. The injection of chemicals into the yolk fat." Prnc Soc. Kxptl ];io! Mcd. 109: -1S-S1. sac of fertile e?gs ]>ii»r to incubation as a Ames, S. R . \V J S\vatiMin. M I. I.udwig and loxic.ity test. Toxicol. Appl. Pharmacol. S: G. Y. Br.,k:.w, 1060. The occurrence of the 760-771. chick pcrictiri'ini edema factor in some oleic acids and products denved th..-rcr>om. J. Am. Xatelson, S., ]°61. Microtechniques of Clinical Chemistry, 2nd F.d., Charles C Thomas, SpringOil. Cheini.-ts ?oc. - 7 : 10-11. field, 111.. 76-79. Best, C. H , nnd X. B. Taylor. 1035. The rhy.-iological Ba-i* of Medical Practice. 6th I'd.. The Potter, G. C., \V. B. Brcv.% R. L. Patterson and r.. Slims, ]03o. Cm-rent status of the toxic \VillUm? St \ V i l k m > ('..nip.-.ny. li.uiimore, Md., principle causing chick edema syndrome. J. 37-40. Am. Oil Chcmi-ls Soc. 36: 214-217. Brew, \V. 15 . J B. Ilorc, J H Benedict, G. C. San^t-r. V. L . I.. Scott, A. Hamfly, C. Gale and Potter and K. Sipos, 193°. Ch:irnctfriz.ifinn of \V. 1). Pound™, 1PJS. Alimentary toxemia in a type of unidentified compound producing chickens. I. Am. Vet. Med. ASJOC. 133: 172edema in ducks. .1. A=^oc. Offic Agr. Chemists. 176 42: 120-128. Sdirinher;". I. I I . and I. Sternlieu, low. Wilson's Carrol!, K. K., 1"61. Quantitative estimation of disease and the coinent:r.tion of ccruloplasmin peak aa-.-ij in ga^-licjUid ci:ronia'c?nipl,y. X;iin se.um. The Lancet, June 20, 1420-1421. turc, 191 : 377 iTS. Shue. G., and L. Gallo. 1961. A study of the Dougbw, C. D., and D. F. Flick. 1061. Collabnoimal variation in chick pericardia! fluid. J. orative bioa$.-?.y for chick edema factor. J. Assoc. Offic. Asr Chemists, 4 4 : 456-459. Assoc. Offic. Asr. Chemi-ts. -!4: -1 |o 456. Fire-tone. D , \V. Ibrahim and \V. IIo-«;tz, I°W. S:ni!h. H. A. and T. C. Jones, 1961. Veterinary ratliol.-cy. 2nd Kd., Lc-a S: Febigcr, PhilaChick edetn:i factor ill Ap; 'ic.-tio-i of midelphia." Pa.. 121-122. crocoulometric pas cru-om?.tO'_rra;.'hy to detecSumnjyi,' M-. 1952. Noles on sugar drte.-mination of chick edrma factor in fatf or fatty tion. J. Biol. Clu-m. 105: 19-23. acids. J. Assnr. Oi'fic. Agr. Chemists 47 • 5S4-396 \Vo;>U'.n, J. C., and J. C. Ale\ander, 1059. Sonic Flick. D. F., L. C.al'.n, J \Yinbu>h. C. D. Douslass chemical chai.ictciistics of the chicken edema and L Fricdm tn. 1Q62. T'lioa^iv of the chick di-ease factor. J. Assoc. Ofiic. Agr. Chemists. edema f a c t o r . 1Q61. coll.ibur.itivc study. J 42: HI-MS. Assoc. Oific. A^r Chemists, 4v 2 '1-23°. Wootton, J- C , X. R. Artu.an and J. C. AlexFlick, n F., C n. Polish- and L. Gallo, 1963. ander, J°f>2. Isolation of three hydropericarStudies of tin chick edema cl'sease. 1. Body ciium-producing factors from a toxic fat. J. water distribution and eftt-'ct of diet. Poultry Assoc. Offic. AIT. Chemists, 45: 730-746. Sci. -4?: S'3-fft:. Friedman. L.. 1). I'instonc. \V. Horvitz, I). Bane*. Wootton. J. C.. and W. L. Courchene, 1961. A contributin:; to the knowledge of the structure M. Ansiead and G Shue. 1030. Studies of the of tv.o hydropericaidir.m-proilucinu factors from chicken edema di-t-ase factor. J . As?oc. Offic a to\ic fat. J. .V:r. Food Chcm. '2: 94-OS. A-r Chemisti. 4:: 1:0 !40. Gorn.dl, A. C, . C. J. ll.irda-.vill and M. M. David. Yai(C"ft", A.. P Fite.-t.-.nc, T>. Kanes, \V. llorwit/. L. Friedman and S. XcsKcim, 1061. Studies of 1940. Dctermii.ation of serum prutfms by the chkk edema factor. II. Isolation of a to\ir means of the biu:rt reaction. J. Biol. Chem. subita'icc. J. Am. Oil. Chemists Soc. 3S: 60177: 75I-7C.6 62. I l a r m a n , R F. . C r. n.r.'is. \V. II. Oil. N. G. 265 Reproduced by the U. 5. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARh Food and Drug Admiuistiation NUTRITIONAL ADJUNCTS Cluck Edema Factor. Iff. Application of Mierocor^omrtr-'c Gas Uiromalography to Detection of Chick Edem;> Factor in tats or laity Acids By DAVID HHKSTOXt: WALID IBRAHIM, and WILLIAM 1IOHU I'/Z (Division of 1'ooJ, 1'ood ami Drug Administration, Wellington 25, D.C.) icrt for d ct ccti,,p ,-ans,^ this ,lis,a,e have lv,n found to I,, c,,,cfc edema factor in fata ,o,lsis,s of adsorp,,,,,, ch,,,,r.a.,,prnphv of ex- A rapid W rPoni,, B r^lorinatcd .,rc,,,«a ( i(. ],vc!ro,,,!,,„.< o,,,,,,,-,,! in (o xi, f,,, in ;l,.,,iaiion v .„, ., ,',, ,„„„: ractc-l un.apon.f.ahJcs „., «l, n ,ina, followd .,y a-aly-u, of ^cific frac. lor of .vlalivolv ,io,,.oxie i i r o m n ,ir m:,,ori:,N „.;„, silni , ar (.,)rni;ra, .„,,, al (,on» by a microcoulo.^^ ^ rf.ro- ti,-s. This papc-r ,l(,,nl)0, , :c,,(,ninK )rn. matosrapl, wl,,ol, » srnshivc- only to (,,!,,ro for deletion of -,,,-h lovr L l < lu-Io^us. Il. I .ohro..nato f! ra|,!,iomp|h.i,l S,,,,.if, r f,,,,,;,,,.,, o{ ,„,.,. .,.;,;.,,,,,;' ; ; Bppean, to I» s mor,- sensitivo lh;,n the cluck I)iui!!i«ay. Toxic fill yio'.dcil {.-as clircmaloprapliic peaks ivilh r<;tciiEi>»i t'uac* r;-?:ilivc lo aiilriu tif 5 or more. A;l s; which fai!o<i to rr-^oal tlio»« d,r«matonon'oNic in Usr chick The \viili'.--].n-.i.l i,i-i-i!trrn,-f of chi,-!; i-.Icm-i ili~cM-r in !!i."i7 n - J i I i c i l in i)i ( . i| ( ..it!is of milcif voting chi'-!-:cn-. fli'- ! < > \ i r nia!C'rj:i!> ,.()la,,d f n i l l , t h c f a ,s arc .v-.^ir.c,! l,v «-in• " a niicrcic-oulfntic'tri" j;a- (•h;»';'.ia!;':;r-i|ih t an 1 in^friuiit'ii'. \v'3}>cb c:in <}c "--i .'iibnircTo^rain a i n i j i i n i ' i (if i>a!i>;.'.vn. 'J'lu % jocvin-c c i f -luwclutiii;: Mih-laiii'fs i.- an j . i ' l i - ihon of l l n H,i..k M,-!,'l:i f;i.-t(.r :,; lii'c f.,7. 'l"hr ]irc;i-t' . - t r i i c t n n 1 >.: >!K' ?uli-!ani'<cMU.-ii»;i c-liifk c'drina di-r-. r h:!\c- \'ct 1o !»• (!rl(-nniii' l d. rj'i'!;inj;;.•.•/.' vcn!). <»?! tin 1 i l c i c ' c - ' i o n , i-ola!ic:n, and '-1 i r . i i ' ! r i i / a ! i " i i of !!•!!••! sr\<r.i! laiiaraloric- (]-•'?} ill I'l.Vi 1>V �266 Harmon a m i co-workers (4) isolated a toxic substance in crystalline form from a feed grade t a l l o w . A p r i v a t e commimicalion from Ti.-hler of the same laboratory (o) <li.vclo.-erl that the crystalline substance contained about -47 r r chlorine Ynr'zoff and co-workers (6) isolated a crystalline halogen c o n t a i n i n g m a t e r i a l th.it produced cliiek edema symptom? at O.I ppm in the thoi from a sample of triolcin. Thi° triolrin w.i< to.x'.o to monkeys, producinc chaiiKe- in the I n c r , kidney, pancreas, and other orpins More recently, Wootton ami co-workers (7) isolated three compound.from o f n v c fat which prodm-ed chick edemdi-oa.-r. M";--- -pecrr.: of t v , n of t h e ro'.npounds i n d i c a t e d a molecule \ \ h i r h ha- a molecular weight of 3111 a m i contains six chlonne atoms. l " i t r a \ i o l e t spectra were consistent with t h e concept t h a t these materials are hijihlv c h l o r i n a t e d a r o m a t i c compounds Ames and co-workers (S) and Fire-tone and co-workers: (!)) reported t h e occurrence of chick edema di-e;i-c f a c t o r in olcic arid ximpfc- de.-fi'iod for Jinman emi.sumjition. A food a d d i t i v e r e g u l a t i o n ' - o f Hie Food and Dnij; A ( ! m i n : - ( r a t i o n nov, requires that food Krade f i t t v acid- lie "free of chick edema 01 other toxic i.idor." At pre-ent, the Jcteclion and a>s;;y of chick edema factor in fats icarried out by a bio.is.-ay procedure (10-12) that requires 21 days to complete. We observed t h a t unsapomfial)Ie matter from toMc f . i t - c o n t a i n e d a number of chlorinated component- winch h n d greater retention -time-- t h a n c h l o r i n a t e d pesticide^ wlien examined in a microcoulometric ga-chromatosraph;- our ob-ervadon prompted this investigation of the n.-c of inicroconl'iinetrie eas c h r o n i a ( o j r a j ) h y for rlefectinp; the presence of chick edema f a c t o r in fat=. Chick edema f a c t o r i-- prc-umed to be present if one or more ;ja-- r'hrom.-itojrrapbic peaks w i t h r e t e n t i o n tune.- r e l a t i v e to a l d r i n of /J i n 20 are f o u n d ; it" ab-enc.' i- pre-umed if analysis of the equivalent of 100 £ of a fat or ' C , , i l . - of I-V,i,.rai U c L - n h i t i o M s . T i l l r 21. S.lii.Ti 121.1070. 7: 1'ohrii'..-; 111: M::mif:ie( Hi-ing Cornpan v. I'a Ait.J. Calif. f a t t y acid fails to reveal the- presence of the-'e sas chrnmatographic peaks. METHOD Kzlinctit'ii of vusaponiftablc matter (modification of AOAC im-iiioJ 26.06.4(5).—Reflux 111 g sample w i t h 270 nil alculiol and 55 ml ,W,i («•/«•) KOJI 'for 1 hour. •Tran.-fer mixture to 2 L separator, rinsing flask with 325 ml H.O, and add rinsing to separator. Add 300 ml petroleum cihcr. A.C.S. (rcdi-iilled, rct.r.mr.e cu; w i t h b.p. -iO-SOcC), and .-'.nke vigorou-lr. Let ].-tv( r.- .-'-parafe, breaking emulsions t a u t may have formed by adding 10 ml alcohol and .-wirling gently. Draw off lower layer, and transfer upper layer to another .x-parator. Repeat extraction 3 times with 300 i:'d portions of petroleum o;her and combine extracts. U ash e x t r a c t s twice \\i\\i 60 m' portions {if H.O by swirling gently. W;. a petroleum other extracts first w i t h 60 ml H:0 and t h e n with GO ml of an alkaline dih:'.e alcohol sola (dissolve 28 g anhydrous K;C(X in 600 ml H:0 and then add 400 ml alcohol), and repeat washings in same order. Wa-h extracts w i t h 60 ml portions of H;O until i.eutral to ]iiieiiolplitl].-dein. Trar.-fer extract in a 2 I, ( rlenni'-yer and dry liy addins; 20 £ -nliydrous NajSO., s«irlniK Mporou'.-ly, ami l.-'.ting the solution stand a half hour. Deca;,: solution t h n ' U t r h a glass funnel, containing a pledget of e o i f M i i in the neek and holding 20 & i.-ihydrou,1; N:i.i?0.. i n t o a n o t h e r 2 L erleiimry.-r. ^yasll fit-t c-rlep.ineyi-r and funnel w i t h t ? . ee 10 in! portions of poiroleuin ctlier, transferring washings from the erleumcycr thiough :hc funnel and i n t o t h e filtered solulion. Kvap(.-at.o most of solvciri on slf-nm liath, and tran-:':r extract to 100 ml tared fat fla-!< eonl-iinii.;: several boiling chips. Evaporate solvent on ^ earn bath and comi'iletc drying under a genlle r:nrrent of air, or by evacuating fiask to 0.5 c:r. of merc u i v v.l.ilc s w n h n g on steam bath. D-'termhie weight of uiHapomfiable m a t t e r . Fi'acfivh'tti'fh <*/ iin*':pfinifwb'r natter by ahiffihin. clirvtnvt'iijruirfiy. — To a •-hromatopraphie rolunin. 25 mm o.d. X 300 inrn lon^, filled at llie b o t t o m w i t h a coarf porosity J r i i f e i i gl;:-.- di-k and Teflon stopco:-'-., arid rcdisiilied ] . e t r ' > ! i u n i ether, dried pri..r to nue w i t h anhydrous X.i.-SO.. u n t i l coh.'.uA is: ~<. full. Weigh 50 g aluminum oxide (Merck reagent. No. 71707}. and transfer U. column Store t h ' - a h m u n a in tightly closed !>«(|Je, and rlo-e b o i t l " as soon n < pos-iblp after wcighini: out p^rli'.ns for chroiiiatography, I^-f alumin.i si-ttl", a;,d when ;,ir bubbles slop ri-ing to the 267 surfnce of the solvent, place a disk of coarso /ig Aldrin = (peak area, in. 2 ) X [recorder filler paper on lop of the nliimiun. Cover the sensilivily (min./m.)(mv/in.)J X (35.5g/eq.) disk with 20 g anhydrous ISra:,SO,. Drain the x rt»«v./miuj X UOVfe'/gXIO'v/mv) excess petroleum ctlier so that it is level with (]0)/(sensitivity range, ohms) X (% chlorine the upper surface of the N.isSO.. in compound) X (06,500 coulorahs/eq ). Transfer unsaponifiable matter to the chrojn:itographic column, iisiug a ttjlal of 20 nd For a 0.1 mv/in. feopler scnsitivify, 2 rnin./ petroleum ether. Allow liquid I'-'vel to fall so in. chad, speed, and 12.8 OMJIS sensitivity range that it is just below the top of (he Nn2SO,. resistance, the equation above reduces to: Elute simple with 400 ml portion? of each of fig Aldrin = (area X 34.5)7% chlorine. Ihc following solvents (dried prior to use by The number of strokes of a disc chart inteshaking with anhydrous Ni^SCM: Petroleum grator enuplpd to a chronmlogniphy recorder'1 elher (fraction 1), 5'-r. clhyl e t h e r in petroleum ellicr (friftiwi 2), and 2'>% elhyl elher in pe- e q u i v a l f n t to each £<;»are inc/i of area can be determined as follows: (a;) Remove the fuse troleum elher (fraction 3). Oolkrt. cluatos in from Ih" strip chart amplifier; (fa) move the 500 ml erlemneyer flasks, add .se\'eral boiling pen upscale on the strip <harl a known dischips, and evaporate to small volume on steam tance from the baseline; t'r) run the chart a hath. Transfer residues tu fared fal extraction flasks, evaporate solve/jt, and weiitli. Transfer known disf.-uK-e; and (/•/) rj^vide the crifcalated area (In itthi. '< distance traversed by (lie pen) to 2 fi short style v i a l s having screw cap w i t h (in hner, ainl cvapnrafe solvent. by (he number of strokes obtained. 'Mic,roc<ii.il<unctii.c yas chro'nfii<>(iHiphy.—DisJ3y using the formula caf'cula(ed as described above, a rerovcry of at lea.st 70% of t h e aldrm solve 4 g sihcone grease (Do-.v Cnrrsing Hifd" inj(.'Cted should be obtaiue*?. Vacuum Crease) or Jjow ('o/ainK DC 200 i>issc)lve fractions 2 an<I 3 from alumina silicone fluid (12.WX) '-eniislokes) in 200 ml rjiloroform on steam b a t h . A<\<\ 10 g and- diromatography sejiaratcly in benzene to give 100 iil solution, and chrorn ifoirraph each soluwashed Chromosorb W (Johns-Manville C'o.). tion in the Dohrmami i n s t r u m e n t . (For analyand stir coiitmuuiisly until most of solvent evaporates (about half an hour). Let stand sis of more t h a n about GO i:i^ of each fr.-icfion, on steam bath J hour, and place in vacuum approximately 50% benzene1 .solutions of up to 250 /(I volume .should be prepared and injected. oven at 5Q°C> overnight to ir-niove residual solvent. Do not inject more than £25 mg material inio 1'ack (he i-oated Cliroraosorh W into a 3' column). First cliromatoftrujih Mo of tile fracliini, and if no cliromato,\'i-aphic peaks with length of 0.25" o.d. aluminum tubing plujr^ed tit. one end with ^Inss wool, usinjc a Uurgess /(A ~ 5 or greater are observed, chromatograph Vibnitool. (Two 3' columns may be prepared the remaining %<» of the fraction (equivalent lo 100 g starling sample).. Chromatograph a from 20 g coated Chromosorb W.) Add a plug of glass wool t" the open end of the column portion of aldrin before each sample, and calculafc /?.i value (retention time relative to and bend if. into a tight spiral, using a 3" diameier mandril. Condition (he column at aldrin) of ea<"h peak hi tlw .-ample chromato275°C for 4S-72 hours, passing nitrogen gram, u^ing a millimeter rule to measure rethrough at 20 ml per minute. tention times. Record R* of gas chromatographic peaks in the rangf lit = 5-20. Peaks Prepare .a 1.00 X iV'/r solution CIO mg/L) of aldrin in hexane or benzene and chromafo- in this range arc indicative- of the presence of p-.'iph 100 pi portions in a Dohrmani) micro- chick edema fac'for. The presence of broad coulometric pas chroma!ograj'h at 21G-2-IScC, bands with no definite pea.k.s is not indicative of the presence of chick edema factor. using a nitrogen flow rale of 50-100 ml per (Note: Types of simpler- which arc found minute so that aldrin elufe.s in 2.3-3 mitmtos.. Use the 128 ohm range setting.'1 Determine from experience to be generally free of components eliaractorisfie of toxic fats ma}- be area of nldrin peak by (riangtifafion, or with examined as described above in 100 p pori disc chart or electronic integrator installed on the strip chart, recorder, and calculate re- tion.-, the sample saponifies! by refiu.xing with 210 ml alcohol and 50 ml 50% (w/w) KOH, covery of aldrin using (he; following equation .'applicable (o chlorinated compounds): and all of each of the polsr alumina fractions gas chroma(ograplied.) '.Miniie.-ip.,lis-IIr>n,..vwel! .Model Y (M Minni':i|ioliK-]Inncy«-«'Il lie'Kiilatcir Co., delphia, I'll.), or equivalent. 3T.3X Phila- �268 269 Results and Discussions - Relative Retention Times of Chlorinated Pesticides and Chlorinated Materials from Toxic Fats.—A number of chlorinated pesticides and several chlorinated materials isolated from toxic fats were chromatographed in the Dohnnann instrument at 24S°C with the 3' column. Retention times relative to aldrin (KA) are shown in Table 1. The pesticides are representative of the whole range of retention times displayed by chlorinated pesticides. A toxic factor isolated from triolein (0), an inactive analogue, and a, concentrate prepared from a toxic fat, all yielded cliromatograms with peaks of /?A — 5 or greater whereas the pesticide peaks were all le.-s t h a n /?A = 4. The toxic fa'-un from triolein as well as the toxic fat concentrate produced chick edema when fed to young chicks at a level of 0.1 ppm in the diet. Table 3. Relative retention limes of chlorinated pesticides and material* isolated from toxic fats (3 fuot, % in. diameter column, 20^ silicon^ grease, 8Qr,", Chromosorb \V; carrier gas fl<>\\ rate, about 00 ml/min.; column tempera*'ive, 248'C; injection block temperature, 270°C) Sample Chlordane Hcptachlur Kepone Mirt-x Strobane Tcdion Toxfipfiene Toxic factor from triirlc-in Inactive analogue fioni triolein Concentrate from a tovir* f;:t Retention Tirac rs. AWrm (.R.,) at 250"C on siiicone columns. When a lowmelting inactive isoiner5 having the Fame retention time (iZ A = 6.0) as one of the toxic factors was chromatographed in the Dohrmann instrument, the following /?A values were obtained at. 2-10°, 248°, and 250° respectively: 6.0, G.4, and G.2. Preliminary Analysis of a Group oj Toxic and Nontoxic Fats. Microcoitlomctric Analysis of Unfu-por.ifia.ble Matter Without Prior Fractiojutlion on Alumina.—A group of 7 toxic and 7 nontoxic fats were examined initially. The fats are described in Tables 2 Table 2. Datn on to\ic fats Component 1. 2. 3. 4. o. G. 7. Tallow acids, still distillate Tallow acids, still distillate | Tallow acids, j still distillate Tallov.- arids. still distillate : Tallow acids, still residue TallowFat from broiler feed Organic Cl in Unsa% ITwa- I»om(iablo M»nu- i i»>iiilir.Llc! Matter. Maltrr f^cturei Ppm 1 18.3 10 1 M.8 47 2 10.1 14 3 2,9 25 1 4.5 2000 2 4 5.1 2.5 39 8000 j 1.0 0.'.) 2.2 3.G 0.5-3.5 3.4 0.6-3 K 5.0 Sl.O 2.3, 3.0, 5.-! The chirk edema-producing factor- isolated by "Wootton and co-workers (7) had retention times rc!:;tivc to methyl ararhidate of 1.17. 302, and 3.17 when chromatograplieii :<t 2,'0'C on a 20^, siiicone column. Since a l d n n ehitcs twice as fa.-t as m e t h y l arachidate under the-" ronditiims, it would be expected t h a t these toxic factors would T have Rf, vahies of about 2.4, fiO, and G.J and 3, respectively. Presence of chick cdcnr-. disease was determined by bioasssy (2) usini a special basal ration. Organic halogen in the unsaponifiablc matter (assumed to b*ohlorine) V.MS determined by microeoiiiomr. trie gas clnomatographic analysis of 50 inz portions of unsaponifiable matter withouprior fraciiona'ior: by ahimijia fhromato;raphy. Both toxic and nor.toxir fats cor.taiued widely variable amounts of unsapo::ifiabie matter and organic chlorine. Soun-<refer to individual manufacturers. Gas cliromatograms of unsaponifiable matter fro::, two of the toxic fats (Xos. 1 and 2) tho-,-. peaks with 7?A values of 14. Gas chromat<grams of unsaponi liable matter from th< * Supplied liy Dr. X. U. Arttnan, Procter a:v' Gnnih!c Co., ("inoinnati, Ohio. Tnhlc 3. Data on noiilovir fa O'mijxjucnt 1. 9 3. j 1. 5. i I n. 7 Cottonseed oil ((SO) Cottonseed oil (ISO foots. still residue OHO fatty ucids Vegetable oil foots Tallow f a t t y 1 acids, .still j residue ! f'orn nil ( 1 Orcjinic Cl ,11 I'n.snP'jiii'nhl.-Wi.ttor, Pi..,. Manufnclui.-r "/ Van,. ponll'atble Mi.tt,.r 8 O.C, 80 8 (, 0.5 12.1 13 5 2 5 0.4 2.2 31 28 G 2.2 7 7 O.C 1 i 1 ISO nlhi-r toxic fats and from noj.tcvic fa(s obt.iimd without prior frartior i i i n n by aluniiin rlii'iiirit.»Y,.,],], v „,,,,. n ,.,,| ar; ,,,oyt ,,f tin- ortvniK haloprn i.]n;-.-.l in 2-1 minutes, •tii'l lli'.-n- wi-re IK, p,.,!:-. w i t l i ft, vahies greater l i u i u '.]. .Since- ci<iiiponcnts w i t h /f x values of ^ or more :>r<« iiMi.iJ|_ v pre.-c'jjf in the unsaptuiifiabU- inatlcr of toxic fats at very low levels, a conc'omt.-.tion -top bv .ilnmin-i rliriiii!;ii.i;.|-.i;.|iv ,. „,.,.,,, .,, v j,,;,,,. to microcotilometric analysis. Frnc.tionntiuii of l;>,s/:pmiififtblc Matter mi Aliim;>i,i I'rii,,- f,, .WirwrHitirwtrir Anntifii. —The 'i toxic an.l 7 nonloxic fats were ,'hcii anai\zi-d by prnmhm-s essentially as ..Vscribed al-nvi'. .At f:i.,|, r,-h*«, :MI\ J!(((T. ,i-foot i-liroinatograpliii- ((,,'1,11111. were ti-t-d in the niicKicotiloinclric «,-,.-. chiomalograph. U'llh shorter ruluiiius, j.-^lrr dt/lioa permitted auaiv.-is ( ,f 11> xunples each woikin;; i l i v , and H'.-iills were coni]iaral,le to those .ililaiiKvl with the coniraiional (i-foot co!itmns. I'.ec.iux- I O - K K I m- p n r l i n n s of f.tm. p'p were rcpeati-dly mjc-.-i. d, all coiiijioiieiits oi the niicroco:iloii,ft;-ic t,s± rliro;ii.i!csi:iph v;cre cleaii"d evi-ry 2-4 weeks as required. 0:is chroma!o::rapliic coliunns wi-rp rr-plurd CMC!) ]-2 mouths. Jn c.-i.-i-s ivhciv s.-inij.'es rnnlained 1 ,rge amounts of '.in.-aponifi.iblc matter, larger alumina columns were u-ed for (lie cohiiim cljrojii;>ln)ri-;!ii)iv .~o that the nlio of ahiinim to i;n.--ip!inifialilc m - t t o r v.:is at least 20 to ]. In thi--i- rax-s /ipjiro- prialcly larr"r .'imounts of eluting solvonls were ;d'i' i!s*"'i{. Fractions obtained by ad-orplion clironiatography on alumina of »ni-al)onifiablf' m a t ter wen1 analyzed bv n^'Tocnuloinetnc gas rlironiM'nc.tniiliy G.'i- <-(!roni/irofjr:ims of polar Itaetions from tuxtf fats (rluted ivith 5f<i and '2^>'/,, ethyl ether z:i petroleum ether) all showed peaks with /V.s values greater than 5. .\ci peaks with JfA values greater than 5 v-erc found in ga~ cliromatograms of these fnictimi* from th" nonloxif fats. Portions of alumina Evictions 2 and 3 cqnivaiViif f •> only about ."";) g of the nontoxic still re.i-due-, (Ta|,ic j_ -,, m ples 3 and 6) were ciirniii.-lijgraplif'd m the Dohrinann ins(niiiK.n . i> .-aiiM.' of t h e j.-e.-enec of a large amount. >ji i i y ~ t a l h n e n r i i < rial in the.'-e frarlions. '1 lie infrared sport s u m of this material (irii-t'cd by recrvstaii.zing from petroleum -ciherl resembled t h •' "f di]i:i!:uitnne. An a d d i t i o n a l cleanup jirr.'. - l i n e mu-:t he developed for routine in.i!y-i-- of U'D g samples of such sril! residues. A - M i t i o n a l examination of un.-apnnifiabios frti:,i 2000 g portions of three nontoxic vegetable .-..'';•- (Table 3, sample.-; 1, -, and 7) failed to reveal chromatographir peak-' with /f"A valu-'-.- greater t h a n 3. TabTe -i. Slow-elutinp p'.-ulis in jniorocouloiiiftru' cas criroTj>.-«i'>,'rrams from «»naljsi«* *»f tt»vi*- f:>t« No. Toslr Tat 1 2 3 4 o 0 i Ta'low acids, still distillate Tallow acids, stil' residue Talitiw acids Tnllmv fallow ncids, still distillateTallow acids Mi -.11far T»r Ifj §5 1 G, 9, 12, 21 ! G, 9, 10, 14 t G, 9, 10, IS G, 9, 12 G, 9, 10, 18 ? 0 -•{ G, 10 Table 4 lists tho slow-eh.i' ing peaks found in gas chrriinafograms from (3 of the 7 toxic fats. Figure 1 shows chron: tograms of alumina fraction 3 from thro.' still distillates, ear-fj of which was obtained :"rom a different m a n u f a c t u r e r of cominercia! fatty acids. A similar p a t t e r n of slow-ehi:ing peaks sug- �270 gests that a common complex contaminant may be responsible for the presence of chick edema factor in fats. Each peak probably represents a complex mixture of closely related compounds. In fact, when polar alumina fractious from several toxic fats were further fractionated by additional column chromatography on alumina, such purification often resulted in partial resolution of the corresponding peaks into at least 2 components. The fat from a chick edema-producing sample of broiler feed (toxic sample 7) contained over 400 ppm chlordane, and this sample required special treatment because large quantities of chlordane clutcd in alumina fractions 2 and 3. Although the 7?A of chlordane = 1, the large amounts present produced overlo.ided chroma tograms which interfered with gas chromatographic detection of other components. A portion of combined fractions 2 and 3 was molecularly distilled in a ''cold finger" pot still for 2 hours at So°C and 50 fj. pressure. The chlordane was volatile under these conditions and 271 w:is eliminated from the residue which was analyzed in the Dohrmarn instrument. Peaks with flA values greater than 5 were then found in the chromatejirams. These results indicated that alumina chromatography of unsaponifiaKe matter followed by microcoulometric gas chromatography of appropriate fractions might be used as a screening procedure to detect chick edema factor in fats ajnd fatty acids. Additional work on cleanup procedures is required before this technique can be applied routinely to examination of 100 g samples of low-grade fats, such as stiil residues containing large amounts of material that elute in alumina fractions 2 and 3. Effect of Alumina Activity and Column Dimensions on Adsorption Cliroinatography. —Alumina activity was found to affect the rate of elution of substances from toxic fats which are responsible for the pas chromatographic peaks of # A = 5 and! greater which are characteristics of toxic samples. Various batches of Merck alumina used for this work were found to van' in activity from (C) RA=6 RA-.0 (b) R =18 A (a) RA=I8 10 20 30 minutes . i ;r..«lnc tj':-- chrorrialc';rens of olun'r.a fraction 3 isolated from tor.ic b^'cn s'ill distillate! (b) 5 g, and tc) 20 9 cf fot. Brockmaim activity I to activity II. ActivBecause of the design of the mic.rocouloities were determined by observing the rate metric gas chromatigr,..p!) u^-pd for thts work, of travel of solutions of specific, pairs of azo whereby oven temperatures are controlled dyes (13, 14). Using activity I alumina, un- only by a variable transformer, line voltage siponifiable matter from toxic fat or from fluctuations result in continuous variation of toxic fat added to USP cottonseed oil was column trmpcra'un-. A variation of it 1°C eluted from the columns so that thu charac- within a 1-2 hour period is the best stability teristic slow-eluting peaks were found in gas to be expected. Even with these variations, chromat'ograms of alumina fraction 3. With however, the instrument is suitable for deactivity II alumina, these peaks were, found tection of .slow-eluting mat/'rials in toxic in chroni.-Uor;rams of alumina fraction 2. fats hp'-mise of the large difference in reGenerally, the alumina ik-ed in this work tention Hints between tl;e.=e slow-eluting was not standardized; it. required gas chro- materials and the chlorinated pesticides and matographic analysis of alumina fractions 2 other fast-eluting chlorinated materials and 3. Standardization of the alumina found in all fats examined. should permit elution of the slow-eluiing Analysis of Dioassay Collaborative Samcompounds in oue fraction, reducing the ples.—A recent collaborative study of the number of samples required for gas chroma- AOAC bio.iisay method for detection of tography. For example. Mcrr-h alumina chick edema disease (12) indicated that the 'heated 48 hours at 200 °C had a Brockmann lower linit of sensitivity was obtained with activity I, and all the slov-eluting peaks a test cample containing 3.56% of a toxic from several toxic fats examined were found fat in liSF cottonseed oil. One hundred g in chromaUigrams of fraction 3. Work is portions of this sample, the toxic fat, and continuing on a procedure for standardizing the original cottonseed oil wore analyzed. alumina in a simple and reproducible man- Extracted unsaponifiablcs were chromatoner. graphed on 50 g alumina as described above, Column dimensions also %u-re found to and the 5% and 25% ethyl ether eluates affect the elution of characteristic substances (fractions 2 and 3) were gas chromatofrom toxic fats. When activity I alumina eraphed. Chromatograms of fraction 2 from was used, these substances wrre eluted in the cottonseed oil without and with added fraction 3 from a 25 X 300 mm colurhn, toxic fat are1 shown in Figs. 2 and 3, respecwhereas they eluted in fraction 2 from a tively. Peaks with 7? ' values greater than A 30 X 300 mm column. .1.5, including the 7BA = 6 and RA = 10 Effect of Column Temperature and Flow peaks, are due to the toxic fat. A chromaRate on Gas Chromaiography of Mmc-cluting togram of alumina fraction 2 from 100 g of Components of Toxic Fats.--Studying the test sample containing 0.78% toxic fat in gas chrom.itographic behavior ot chlorinated pesticides, Burke and Johnson (15) found Table 5. Analysis of USP cottonseed oil that varying the column temperature and/or containing various levels of added toxic fat; the carrier gas flow rate resulted in varia- microcoulometric gas chromalography of tions of relative retention times of the pesti. alumina fraction 2 cides. Similar variation' in relative retention Disc Integrator Response limes were observed with ihe slow-eluting (No. of Pen Strokes) components of toxic fats. 7?A values, howToxic Fnt Added. % ever, were affected more by variations in RA. -6 KA = 10 temperature than in flow rate. A toxic sub0.00 stance isolated from triolein (6) had the 0 0 0.78 6 following IIA values at 2-iG", 250° and 2 1.56 14 6 252°C: 5.0, 4.<>, and 4.7. An inactive' ana3.12 32 13 logue isolated from the sample had the fol4.08 32 12 lowing 7?A values at these temperatures: 9.7, G.24 47 28 SO, and 7.S. �272 273 RA.|.Z5 (o) to minutes Fig. 2—Microcoulometric 'j;as chromatogroms of (o) aldrin standord (1 719), ond (b) alunvr.c* fraction 2 frc USP cotlonseed oil without added toxic fat. •.lie USI' cottonseed oil is shown in Fig. 4. Tht- KA - (i ami /r\ - 10 peaks can still -.- definitely Detected ;i( tins level of toxic ; , i , which was half of that, found to lie at •lie lower limit of detection of the AOAC :,uussay. Sample* of I'SP cotlon.-ced oil coiitamin;; Lirioits levels of added toxic fat up to (>2-l'i •.-.ere analyzed. The disc integrator rc.-pon.-e ;-;' the HA = (i and 1<\ - ]() peaks are. shown ;n Table 5. The integrator response (ninn\-r of pen strokes) is :t]ipr<i\iniatcly propor::onal to the level of added toxic fat in the .•jtlonsml «il- L'acli t-troku is equivalent to ijuiit 0.05 /tg of orsiamc halogen. Analysis of ('uiiiMuraul O.'sit: tnul Niarir .Iri'/s mul iJirirnlii'ts. -Tuelvt- loin] gr.id.ivic acids were examined liy the elirom.ito ;:.i]ihic method. 'J'lie c-xaiuinaiion ,,f <i ,,f :':0.-T samples for chick edema 1o.\icity \va.~ rvpurted earlier (',)). C,as ehi-,niatograii;i of Emilia /raction-- from .-, lio'itoxic and a toxic -,riil are shown in I-'ig.-. ;i and (i, resjioclively. Chiumatograjihic i.,-.,k.- of # A ~ (j alK i ;:i-ater from the toxic s-aniplc are shown in t K . (i. These peak.- \\ere present in alumina fMotions 2 ami 3. 'i'he "iH-ak" with A\ ^ f, in the chroniatogr.-mi from alumina fraction 1 is believed tu be an artifact due to overloading of the coiilomctcr. He-suits of analysis of the 1'2 oleic acids, comiiarcd with I In- i-liiek bioas,-ay usinj; a special ba.s-d ration il'j. are shown in Table (i. li"tiroppricardiimi aciivily, the jiriniary index of the jirescnc.- of chick edema factor, was estimated as described by Firestone anil co-workers (!l). Six samples wore posiiive by both the liio-.-ay and rhromatogra;>))i'' meihods. One faiujjic (\o. (!) was negative by the regular bioassay wlien fed at ttie u.-ua! level of Ui'r in the diet. However, a positive response was obtained when extracted ijiisajioiiifiai-'e m a t t e r was fed at a Icu'i "'(luivalenl to (• limes that pre.-cnt in the normal test diet. Sample 10 was m g a l i v c by the chick bioassav, but gave a weak positive rc.-ponse by the cliromatograpiiic method. A small fhrom.'itographic peak with K_\ = 10 was obtained from alumina fraction '2. In comparing bioassay a c t i v i f v and peak areas of slow-cluting component of the other toxic oleic acids, it would !>e exported t h a t the level of chick edema factor in sample 10 (a) R =10 A 10 m inuies 10 minutes Fig. 3—Microcoulometric gas cnromaiogrotn of alumina {faction 2 from USP cottonseed* oif containing 1.56V toxic fot. *• " ^ ^:r^.i,-h±Lrd^4:0^^icSs.lJSP — " �274 275 7"al»/i: 6. Analysis of <-onn»iiT*'iaI oh*ic acids ic) Sai-.ij'le N'o. P' i B'a (b) 3 4 5 G 7 S (a) !) 10 11 12 Hio.-ts^tv: llv<lro|>-Mcimliiiiii Activity (VI + 0.2 +3 +0.3 +0.1 +0.2 ClironiKtoitiaiiliii: An aly.sis, Kdaliv*- I'cik An-a llntc-RrntCT Sirjkt :-J j HA = T>- f, /;,. - 9-;o 74 ISO :{40 98 118 40 20 30 40 minutes f-tg. 5—Microcauiometric gas diromatograms of (a) alumina fraction 1, (b) alumina fraction 2, and (c) aly. mina fraction 3 from a non-toxic commercial oleic acid. (c) R =10 A (b) (a) 20 30 40 minutes nina fracfTo" 3 fror.-. a !ox J, (b) alumina fraction 2, ond (c) ileic ocid. R* = j.->-is 9 8 104 0 14 !X) 196 ISO QO OO — . +0.1" 100 105 0.0 _ — 1 I » Knell stn.k'- i-1 fl'i'vjtnt '.o ..l-f.ut (1.0.> w. ur-tlut- h.-.!- ;•- :•b Unsuj'omftable matter fpd ^t a I'-ve! ftjuivalcnt to 0 tiir' s that present in the normal tost (.':••! 10 j 25 i might be below that detectable by the. bio'1 and .'•; to varying extoil.-, depending upon assay. the aiuinina activity. 'J'lie total integrator In comparing hydropencardium activity response from both almnina fractions was and gas cinematographic response (relative fairly constant, however, varying from 130 peak area), no parallel relationship was to 109 integrator pen sirckes. found between the bioatsay response of the In addition to the o'tk- acids, a number toxic oleic acids and the chromatographic of derivatives of oleic :i;id (triolein, fjlycerol responsc. It should be emphasized, however, monoolcafe, etc.) were an.ilj-zed. Xine of that the sfow-elutmg compounds in toxic ten Rimpl.-s were negative by both the biosamples represent both toxic and relatively assay and chromatographie analysis. One nontoxie materials, and most of the g:l~s toxic .-ample, a triolein, pive gas chromatochromafographic response is probably due grams'wilh peaks of /? — 6, 10, and 19. to relatively nonfoxic substances. Neverthe- - ']>" stearie acids andA <>rivtti\es examined less, niieroroulometrie gas chromalographie were neea(ivc by the bio.i -.-iiy and ehromatodetection of slow-elating compounds appear.- graphic nx'tliod. to be an effective scH;cning 1»ol for segregatAnal'isii* (>! Animal <;nd Vegetable Fats ing (|iie.-.tionable samples an ! div.-rting them mid ('iinuht n'ia/ Vf(jt:!tili •: Oil 1'atty Acids — to nonedible uses or foi iudh'jr testing by Fifhon animal and vegetable fats were exhioassay. amined, 'i'liree toxic .Miimal tallows were The gas chroni.-ilofrivijjhic response of the positive by the chroni-itographic method. tf,v = 9 Peak pre.-ent, at widely different levels in 2 oleie acids was- cheeked by analyT;,l)Ic 7. Response- nf KA = 9 peak sis of individual samples on different days from 2 ctloi' iioi<Is ..sing different gas clmm.afonraphie columns' Pit': ,.!.- ! Sample 2 fnr each run. Kesults are shown irT'fable 7 Sample I contained a low level of material RlK, Fr:i<-Frac!"m<Fr.ocNti lion 2 tion :i tloi, :; tiun 3 responsible for the # A _. 9 peak. The dilierTotal .-•nccs in integrator rc.-p,,i,-e are due both to 1 0.2 0 20 ]30 variation in coulometer response and to in•> 0.8 0 ;i» •18 144 ;icniracies of disc inlegrafor response at low 3 0.0 0 JOS ion :..il»Ren levels. The larger amounts nf niate-1 5.0 0 JOO 5(1 150 ri:il in sample 2 respoii,-,ble for the /> t -.- 9 " peak were distributed in alumina fractions , L-« Dis.- intf: f i:>f(»r n'.s|if>n.sc m ii:tt;i'»fi' »f 'iisc iii'ci;r:tt<)[' ^ttoki-*. �276 277 (G) Yarl/.<.:T. A., Firr-miie. D.. Ban,-;, D., Hoi-u]!,:, W., Fiii-dinaii. J . , :.nd Xc-ln-mi. .S., J. Am. (til CV-i M/..// ,S<«-.. 38, (i'J (b) 112) Fli.-k. D. F., Call... I,., Wmbu-h. J , DougI;,.-*. C. D.. and Fncdmar-, I/., ibij., 45, 231 (11102) (1901) (7) Woolion. J. C'., A i t m a n , X. K., and Alexa m l i r , J. C:., J'l,i« Ji.itm-'l. 45, 739 (1902). (8) Ames, S I!.. .Suan-on, A\' ,1., !.ud«i(i. M. I., and linkaw, G Y., J. Am. Oil C!u->nixtx' &,<•., 37, 4 (10) (19i«). (9) Viri-si'.in , I)., Hurwitz. IV. Friedni.-in, J... ., . and Sluie,, (i.. M., HiiiJ... 3'<j.. -IIS (1901). (10) ]JoiicJa.-<, C. D., and Khrk. U. K, Th Jtinn,-i "C'lia»! (1961).i: (Ki) Ucfiinaim, K., Chn,iniitoyfOi,n>j, Ueinhold l j ttl..li.-.liiiip Corj'.. New York, 1061, p. 3-1. ( 1 ! ) Hrockiiiaiin. ))., ;nnl Si-It',Mrr, JT., Her., 74, 73 (19-11). (15) JJiirhc, J., and Johnson, J,., Tim Journal, 45, SIS (1902). -0, lo 77'/.v Journal for pultiication !»• Scvi-uty(ID ;c-i,a,, ,-, i,, M,.,h.,-is,- n,i.i, 44, MO -iv-'n^rM^inr^r'A^,,';;,,,;'^ ! Il ( mniw (a) 10 minutes Fig. 7—Microcoulometric gas chronntogroms of (a) alumina fraction 2, and (b) alumina frc-cf'on 3 from a growth-depressing colinmecd oi! whith did not produce chick edema disease. Ten vegetable iVs dricludip.;: cottonseed oil, corn oi!, peanut oil, saiflower oil, and toybean oil) were vx-:mu:.-d, and were ctgative by the bioassay and chromatographic method. Chromatograms from several of the vegetable oils showed broad bands \vitli no definite jjr-aks. Chromato:;r;;ms of alumina fraction- 2 and 3 from one of these sample.-, a USP eotton;C"'l oil, ."re shown in FIR. 7. The broad band in chruina'.ofcram (a) is typical of that found in the othor oils. This sample depressed the growth of young chickens, but did not produce symptoms of ch:;-k edema di>cr,se. The broad band in chromatogram (a) with a maximum at S.4 is not characteristic of a toxic f a t . No additional work has been done to identify the substances causing these bro.id bands. Twelve ccinnicrcial ve-ietable oil f a t t y acids, no;ii'j\ic by the chirk edema bioa.-s.iy, were examined. Ti)f-o samples incliulcd f.-.tty acids from coconut, cottonseed, corn, palm, soybean, and tall oils. F.leven of the sample.- wore negative by the chromatographic ir.othod, but one of the samples (a tall oil fatty acid) gave a sm;.-U chromatoEraphic ;>eak with R^ — 5, in* iicalive of a toxic sample. Acknowledgments The authors wish to express f'>.cir appreciation to Andrew Yarfzoff who .'• insisted tl.initiation'of this work; to Glen Sliue, Dw.ald Flick, and Lind.i Galiy vi>-o conduct!-: the biuass.iys; and to Benjamin Y,~el>b, Alv;:. Frei'Iand, and Peter LeXaid whci carried oi:p most of the analytical work. REKKKKXCES (1) Brew. W. B., Don-, J. B.. IV-",odict, 3. Potter, G. O., and Sipos, E., ";'his Jon; 42. 120 (1039). (2) 1'rirJnnn, L., Firestone, D., ? iorwitz, Banes, P., Anstead, M., f:n= 1 .S!mc, ihiW, 42, lL'r' (1059). (3) \Vooiton. J. C, and Alex;.- der, J. i/.iV/., 42. Ml (!9.v9). (-1) Harmon, K. K., Davis, G. K , Oil, W. Brink, X. G, and Kuelil, F. A., }. , Ci.cm. fuc... 82, 207S (T.lOO). (ri) Ti-'lili-r. M.. Merck and Com;-any. pnv communication, July 19, 1WO. «i 'wS , t;;™i, «y »- <*•*• '^>. - �278 Reprinted from TOXICOLOGY AND APPLIED PHARMACOLOGY, Volume 5, Number 6, November 1963 Copyright © 1963 by Academic Press Inc. Printed in U. S. A. TOXICOLOGY AND APPLIED PHARMACOLOGY 5, 760-771 (1963) The injection of Chemicals into the Yolk Sac of Fertile Eggs prior to Incubation as a Toxicity Test JOSEPH MCLAUGHLIN, JR., JEAN-PIERRE MARLIAC, M. JACQUELINE VERRETT, MARY K. MUTCHLER, AND O. GARTH FITZHUGH Division of Pharmacology, Food and Drug Administration, Department of Health, Education, and Welfare, Washington 25, D. C. Received January 24, 1963 The increasingly large number of food additive chemicals introduced into the market each year has necessitated the development of rapid and reliable methods for the evaluation of their toxicity. Toxicologic studies of all these chemicals by the usual methods using animals are very difficult, and such studies sometimes give inconclusive results. The toxicity of some chemicals, and especially of food additives, may be determined by injection of the chemical into the yolk sac of fertile eggs prior to incubation and subsequent observation of the effects of the chemical on the embryonic development of the chick. This appears to be a promising method in that it may be carried out much more economically in terms of money and space than would be possible with larger animals. Hundreds of chicken embryos may be observed in a minimum of space, and over a comparatively short period of time. The feasibility of using such large numbers is valuable also in the statistical evaluation of toxicity data. A review of the literature shows how little work has been done in this field except in a fragmentary way on isolated cases. Most of the reports refer to injections of chemicals made after the fourth or eighth day of incubation and examination of the embryos killed before they hatch. The earliest work that we have found in the literature was by Fere (1893). During ten years after this date he published about sixty-seven papers; a review article (Fere, 1899) contains a summary oi many of his studies. His work consisted mainly oi injection before incubation, but the eze; 279 Since 1900 other articles have appeared in the literature, but most of them concern the effects of one or two chemicals on a small number of embryos. One of the most informative papers is that of Ridgway and Karnofsky (1952) in which they report experiments on compounds containing fifty-five of the elements, mainly the metallic ones, generally injected at either the fourth or eighth day of embryonic development to study toxicity and teratogenic effects. Hamburger and Hamilton (1951) have described an elegant method for determining the stages of development of the chick embryo. More recent investigations employing the chick embryo to study the toxicity of various chemicals have been reported by Kemper (1962), Platt et al. (1962), and Goerttler (1962). Finally, the classic books of Romanoff and Romanoff (1949) and Romanoff (1960) contain a wealth of information on the avian embryo. Preliminary reports of this investigation have been presented fay Marliac (1962) and McLaughlin and Mutchler (1962). EXPERIMENTAL The fertility and hatchability of eggs and the livability of chicks are dependent on a complex interrelationship of ecological factors, among which are the genetic background and the age of the mated birds, the nutritional status and general management of the flock, and seasonal variations. In view of this the initial phase of our work, which started in 1959, was devoted to a study of the hatchability of our supply of White Leghorn eggs1 under conditions existing in our laboratories. The data accumulated during two years for control eggs showed that the hatchability of these eggs was, in fact, consistent, reproducible, arid very high. The possibility of a seasonal variation occurring in responses to compounds introduced into the eggs was also examined by repeated testing of several chemicals at all seasons of the year. No important variation was detected. Selection of eggs. Eggs to be injected are first candled in order to discard those that are defective and to outline with a pencil the exact location of the air cell. In our laboratory 9% of 5000 eggs had to be discarded: 2% cracked, 4% with improperly calcified shells, 1.5% with a tremulous air cell, 1% with the air cell in the wrong place, and 0.5% with blood clots. After the elimination of such defective eggs, the hatch of control eggs averages 95%. A further restriction is based on the weight of the eggs: all those weighing less than 52 g or more than 63 g are rejected. After l Truslow Eggs. Chestertown, Maryland. �280 candling, the eggs are randomized in order to avoid series of infertile eggs in any one experiment. The initial experiment with a given chemical is for range-finding and is performed at two or more concentrations of the chemical with 10 eggs per level. On the basis of this information, 20 or more eggs are injected with the appropriate amount of the chemical. If the chemical proves to be nontoxic, the experiment is repeated with the minimum number of eggs that will give a reliable and reproducible value for the hatchability. In the case of a toxic chemical, additional eggs are injected to determine the specific effects of the chemical. The total number of eggs used for a chemical depends upon the data obtained initially and upon the kind of information desired. Hence, data for some chemicals are based on less than one hundred eggs, whereas data for others are based on several hundred eggs. Technique of injection. The injections of pure chemicals, chemical solutions, or suspensions are made at volumes up to 0.10ml. When necessary, dilutions are made with solvents such as water, propylene glycol, corn oil, peanut oil, or other nontoxic solvents. In order to avoid contamination, the injections are carried out in an Isolator Box2 with a sterile atmosphere created by using formaldehyde vapors (produced by mixing 2 g of potassium permanganate and SO ml of 37% formalin). During the period of a year, more than fifteen hundred noninjected eggs were exposed to formaldehyde vapors; no toxic effect was noticed. After exposure to these vapors for 30 minutes, the eggs are ready for injection. The large end of the egg is wiped with a sterile gauze pad moistened with a 70% alcohol solution, and a hole is drilled in the shell in the center of the surface over the air cell (Fig. 1). Care must be taken not to damage the shell membrane with the point of the drill3; this is to avoid, if possible, contact of the air with the egg membrane. Fine particles of shell are removed with an aspirator to prevent the needle from carrying them into the yolk. Immediately before the injection, each egg is shaken with a quick twist of the wrist. Since the germinal disc occasionally sticks to the air cell and it is possible to damage it with the needle, this movement will allow the disc to float free in the egg. 2 3 Kewaunee Scientific Company, Adrian, Michigan. Burgess Vibrocrafters Inc., Grayslake, Illinois. 281 The needle (hypodermic, 1 inch long, either no. 22 or no 27 on the vaoojmy of the liquid to be injected) is inserted hori tne Tt "^he ^ (Fig" !)- CarC mUSt be taken * withd needle to avmd dancing the vitelline membrane since such damage cause the yolk to spread out in the albumen. If the end of tbe^Ste some yolk on it the injection is not satisfactory. The needle should wlped »ith a stole gauze pad between each injection. As soon i tSe has been ejected the hole in the shell is covered with a smanTe e Scotch tape, care being taken not to cover the entire air cell GERMINAL DISC ALBUMEN SHELL MEMBRANE \ / / VITELLINE MEMBRANE / CHALAZA FIG. I. Diagram of egg and position for injection. Incubation and hatching. The injected eggs are put into the incubator trays with the large end up; the trays are placed in the incubator,4 which automatically rotates hourly and is maintained at an optimum temperature of 38°C and a relative humidity of 60%. The eggs are candled on the fifth day of incubation and every day thereafter. Clear eggs and dead embryos are removed for examination. On the seventeenth day of incubation the fertile eggs are transferred to the hatcher5 and kept at a temperature of 37°C until they hatch. EVALUATION OF DATA The injection of the chemical into the egg may produce one of four possible results: (1) the chemical is highly toxic at the level injected, and * Humidaire Incubator Co., New Madison, Ohio; model no. SO, capacity 450 eggs. 5 Brower Manufacturing Co., Quincy, Illinois. �282 283 all the embryos are killed during the first 20 hours of incubation (before the two-somite stage); (2) the chemical is toxic but allows a number of embryos to develop only up to a certain point, and some possibly even to hatch; (3) the chemical has little effect on the hatch; and (4) the chemical has no effect on the hatch or on the posthatch development of the chick ("no effect" level). If the chemical appears to be highly toxic and all the embryos are killed, the experiment is repeated with smaller doses of the chemical until some hatch is obtained. If the chemical is toxic, but allows the embryos to develop for a longer period of time, dead embryos are examined pathologically and the chicks that do hatch are examined for eye damage, color of the feathers, weight, length of the legs, form of the beak and of the rump, hematologic changes, and condition of the internal organs (liver, kidneys, heart, gall bladder, and spleen). It is advisable in all cases to observe the chicks for a period of a fenweeks in order to detect any delayed effects. Since as much as 30% of the yolk remains at the time of hatching and is absorbed during the first 7 days thereafter, effects of a toxic chemical may first be observed at this time. There may be weight retardation, death during the first week, or the appearance of nerve damage occurring as late as 2-6 weeks after hatching. The toxicity of a chemical is evaluated mainly from the percentage of hatch at varying dosages of the chemical as compared to noninjected (control) eggs, from a study of the embryonic development of the eggs that fail to hatch, and from a study of the appearance and development of the chicks that do hatch. However, several other factors must also be considered in this evaluation: these are specific gravity, solubility, coagulating effect, pH, and the ionic concentration of the chemical tested. If the chemical has a high specific gravity, there is the possibility of its settling out in the bottom of the egg and thereby giving a value of apparently low toxicity. The solubility is quite important since the availability of the chemical for utilization in the chick embryo is partially dependent on its solubility in the egg. However, since egg yolk is an emulsion, solubility problems are somewhat minimized. In the case of insoluble chemicals that are injected as suspensions, it is also necessary to consider particle size in the evaluation of toxicity data. In order to have a toxic effect, the chemical must come in contact with the embryo either directly or indirectly through the bloodstream. Chemicals such as the lower aliphatic alcohols have a coagulating effect on the pro- tein, and this coagulation may decrease the availability of the chemical as well as some yolk nutrients, and thereby alter the response of the embryo. If the pH is highly acidic or basic, a pH effect differing from the true toxic effect of the chemical may be obtained due to interference with the normal acid-base equilibrium in the egg. The ionic concentration is also important for a similar reason. The observation of increasing toxicity with increasing concentration of a chemical should be interpreted cautiously, since highly concentrated solutions may upset the physical equilibrium of the yolk by causing osmotic effects. Finally, the introduction of a chemical into the yolk may cause a special type of toxicity because it destroys, alters, or combines with essential nutrients such as vitamins and minerals. EXPERIMENTAL DATA Twenty-five thousand eggs have been used in our laboratory during the past three years to test more than 100 chemicals with the following TABLE 1 NONTOXIC CHEMICALS Solution injected Dextrose Undiluted Undiluted Undiluted 0.9% in water 5.0% in water 5.0% in water 10.0% in water in water O.OS 0.05 0.05 0.05 0.05 0.05 0.10 0.05 0.05 produced effects at dose K *» in feeding often provided Geologic information venuonal methods ; bicoW „« tfn „ 95 95 90 90 90 70 30 20 90 ""* Chemlcak by con- �285 284 ,', .2 £J ** -a v glflf o a > I &1 c -2 •§ 1 tt •S c ^J '-• I O I £ XI <u V C j<" tJ •M 8 -a 0 <u § fe-a 0. rt " 3 a-^ -a c o f « * isiiHj ijUfisii II is e I;" .! 5 |1 ej^iji §^ (S S3i ? g | a 8l S » < 53 <~> " .3 ' S « «3 Cu Q -1 O i2 e CA O •g 60 V • §5 g -a T= •a E S 03 *r* ^d O tf _1 vn irj pa E- CD O Q O Q °. 0 0 o o O O S S S/ g C . o O o d o o o o C! "S CJ ai g g c i o c 5 S ^> .S .S .5 5 ^S 5 i; ^" 1 "c j: c c . O) I 11 -2 3-g • C u •o c D o d d 3O a •a c 3 O I i3 •o 2S 5 T1 -C ^ .§! These data, supplemented by an examination of- the nonviable eggs and an autopsy of the chicks that hatched, have not indicated any hazard from their use in food. However, it must be pointed out that any chemical, added at a sufficiently high concentration, may have some toxic effect. Since our data and those reported in the literature indicated safety, this phase of the study was not carried beyond a preliminary examination to ascertain that nontoxicity also was shown for these chemicals by the chick embryo technique. This is of theoretical as well as practical importance for the evaluation of any new technique to be used in toxicologic studies. Table 2 lists our results with several chemicals which have been shown to be highly toxic to animals. In each case there is not only the low percentage of hatch at a low level of the chemical tested, but there are also congenital abnormalities and other responses that raised extremely serious questions as to the safety to the consumers of any food contaminated with these chemicals. Lead acetate resulted in no hatch at a level of 1 mg per egg. Autopsy of the dead embryos showed extensive brain damage, as has been reported by de Franciscis and Bocalatte (1962) and by Karnofsky and Ridgway (1952). Mercuric chloride showed no hatch even at a level of 0.5 mg per egg. Thiourea is a known carcinogen wi'h a basic effect on the thyroid gland. The hatch time wat> delayed with im ; .-asing amounts of this chemical. At a level of 5 mg per egg no chicks hnV'tied and the embryo required 35 days to develop to the stage normally aUsir.ed at 20 days. At 2.5 mg per egg some chicks hatched, but most of th^rn had to be helped out of the shell. This effect has been reported also by Yushok (1950). Cavanagh (1954) reported that triorthocresyl phosphate (TOCP), a well-known plasticizer for nonfood use. causes paralysis when fed to adult chickens. We observed this paralysis in some chicks which hatched from eggs inj'ected with 10 mg of the undiluted chemical. The compound ^,/>'-diaminodiphenv]methane has been reported by Zylberszac (1951) to cause cirrhosis of the liver in rats. We found this chemical to be extremely teratogenic at a level of 5 mg per egg; more than 90% of the chicks had a short mandible and leg damage consisting of a severe bending of the tibia and a general shortening of the bones of the leg. Sodium selenite proved to be highly toxic. At a level of 0.1 mg per egg, no embryos developed to more than the 5-day stage. Dibutyl-tin-dilaurate showed no hatch at a level of 10 mg per egg. The majority of the embryos, which did not live more than 15 days at this �287 286 level, had a short and/or flexed mandible; in addition, some embryos showed subcutaneous edema, Aroclor 1242 gave no hatch at a level of 25 mg per egg. At a level of 10 mg per egg, one chick hatched out of 20 injected eggs, but died 2 days later. Some embryos, which were examined after they died, showed beak deformities (often a short upper beak), edema, and growth retardation. Table 3 lists preliminary data on the toxicity of some chemicals which have been shown to be toxic in some degree to animals, and which may be found in some processed foods. Included in this group are various solvents, plasticizers, and insecticides. Some of the chemicals on this list require further study, including observation of the chicks until they reach maturity, before they may be classified as to low or high toxicity. !£* V V ^* .. ~ c •3 -- S I 2^ """ rt £# "" c jj O I I hi 3 " Q §3 I- DISCUSSION p O <" 5 3 32323232SS23323 ? ^ ° d o d 6 d o' o o o o o o o o o o o o o f Q p O O O O O O 2 u£ J ? a •8 -^ u S p •a c P •c _ -o pP 2. "8 2 |i3 l _ U . > •o c p p§ 3 I. 9 2 75 o j= ">. • U 1 1 -^ -t 2. -g & Q. K 3 I >. o ax . J3 s C P The injection of chemicals into the yolk of fertile eggs prior to incubation is a method which can be advantageously used as an element in the evaluation of the safety of food additive chemicals and drugs, and which could be used to screen new pioducts and eventually to correlate their toxicity with that of similar products already tested. If one considers that a chemical injected into the yolk may be compared to a substance which has the power to cross the placental barrier, this technique, in addition to being an embryonic feeding study, assumes further importance in that it is also a reproduction study. The unfortunate experiences recently suffered with chemicals that have teratogenic effects in humans, and the failure of conventional testing methods to produce this effect in animals, emphasize the urgent necessity for new methods of analysis. Preliminary work that we have done in this area has given satisfactory results (Verrett and McLaughlin, 1963). Since this represents a system in which the chemical is in direct contact with the embryo throughout development, it is more than likely that any toxic or teratogenic effects would be readily observed. However, there is always the possibility that the chicken will not be a species susceptible to a particular compound, just as it has been shown that the other commonly used species of animals do not respond to all chemicals in a similar manner. It is also possible for the chicken to be more sensitive to a chemical than other species. Finally, this technique may be applied also to the study of the synergistic effects of chemicals. Results of experiments in our laboratory on �288 289 the potentiation of a few pesticides have been very encouraging (Marliac and Mutchler, 1963). SUMMARY ROMANOFF, A. L. (1960). The Avian Embryo: Structural and Functional Development, 1st ed. Macmillan, New York. ROMANOFF, A. L., and ROMANOFF, A. J. (1949). The Avian Egg. Wiley, New York. VERRETT, M. J., and MCLAUGHLIN, J., JR. (1963). Use of the chick embryo technique in the evaluation of the toxicity of drugs. Federation Proc. 22, 188. YUSHOK, W. D. (1950). The relationship of thyroid activity to the growth and the cytochrome content of the chick embryos and their organs. Inaugural dissertation, Cornell Univ., Ithaca, New York. ZYLBERSZAC, S. (1951). Cirrhosis-provoking action of insoluble diamino-diphenyl compounds on the rat liver. Compt. Rend. Soc. Biol. 145, 136-138. An evaluation of toxicity by injection of the chemical into the yolk sac of fertile eggs prior to incubation gave the following results: Water, propylene glycol, corn oil, peanut oil, isotonic saline solution, and isotonic glucose solution showed no toxicity or a very low order of toxicity. Mercuric chloride, lead acetate, selenium, triorthocresyl phosphate, #,#'-diaminodiphenylmethane, thiourea, Aroclor 1242, and dibutyl-tin-dilaurate showed a high order of toxicity and/or teratogenic effects at certain levels. Acetone, methanol, ethanol, n-butanol, diethylene glycol, ethylene glycol, isopropanol, di-2-ethylhexyl phthalate, hydrochloric acid, carbon tetrachloride, ethyl acetate, malathion, heptachlor, and styrene showed an intermediate order of toxicity. REFERENCES CAVANAGH, J. B. (19S4). The toxic effects of tri-ortho-cresyl phosphate on the nervous system: an experimental study in hens. /. Neural. Neurosurg. Psyckiat. 17, 163-172. DE FRANCISCIS, P., and BOCALATTE, F. (1962). Lead acetate and development of the chick embryo. Nature 193, 989-990. FERE, C. (1893). Note sur 1'influence, sur 1'incubation de 1'oeuf de poule, d'injections prealables dans I'albumen, de solutions de sel, de glucose, de glycerine. Compt. Rend. Soc. Biol., 45, 831. FERE, C. (1899). Teratogenie experimentale et pathologic generate. Cinquantenaire de la Societe de Biologie, Vol. jubilaire, pp. 360-369. GOERTTLER, K. (1962). Der 'teratologische Grundversuch' am bebruteten Huhnchenkeim seine Moglichkeiten und Grenzen. Klin. Wochschr. 40, 809. HAMBURGER, V., and HAMILTON, H. L. (1951). A series of normal stages in the development of the chick embryo. /. Morphol. 88, 49-92. KARNOFSKY, D. A., and RIDGWAY, L. P. (1952). Production of injury to the central nervous system of the chick embryo by lead salts. J. Pkarmacol. Exptl. Therap. 104, 176-186. KEMPER, F. (1962). Thalidomid und Entwicklung von Hiihnerembryonen. Arzneimittel-Forsch. 12, 640. MCLAUGHLIN, J., JR., and MUTCHLER, M. K. (1962). Toxicity of some* chemicals measured by injection into chicken eggs. Federation Proc. 21, 450. MARLIAC, J. P. (1962). Injection of chemicals into chicken eggs as a toxicity test. Federation Proc. 21, 450. MARLIAC, J. P., and MUTCHLER, M. K. (1963). Use of the chick embryo technique for detecting potentiating effects of chemicals. Federation Proc. 22, 188. PLATT, B. S., STEWART, R. J. C., and GUPTA, S. R. (1962). The chick embryo as a test organism for toxic substances in food. Proc. Nutr. Soc. (Engl. Scot.) 21, XXX. RTDGWAY, L. P., and KARNOFSKY, D. A. (1952). The effects of metals on the chick embryo: toxicity and production of abnormalities in development. Ann. N.Y. Acad. Set. 55, 203-21S. �290 291 ucts resulted in a toxic response that correlated well with that obtained by injection of pure aflatoxin B, solutions at the same dose levels, and in most instances the chemical analysis was confirmed. The presence of aflatoxins G,, Bz, and G. had no apparent effect on the toxicily due to aflatoxin Bj, at the levels at which they occurred in the particular samples tested. The separation of aflatoxin B, from contaminated extracts by thin-layer chromatography, and its subsemicul elutioii from the plates and injection into the eggs, confirmed that the toxicity of these extracts was due primarily to their aflatoxin B t content. Use of the Chicken Embryo in the Assay of Aflatoxin Toxicitr By M. JACQUELINE VERRETT, JEAN-PIERRE MARLIAC, and JOSEI'H McLAUGHLIN, JR. (Division of Toxicological Evaluation, Food and Drug Administration, ton, D.C. 20204) The possibility of using the chicken embryo as a test organism for the assay of aflatoxin toxicity has been investigated and found to be feasible. The injections of test solutions were made before incubation, in fertile White Leghorn eggs, by either of two routes: yolk or air cell. The development of the embryos was observed for the full 21 day incubation period. The vehicle for all injections was propylene glycol. The injection of solutions of pure aflatoxins B, and G,, and of extracts of aflatoxin-producing mold culture* ». dicated that the chicken embryo w» sensitive to these compounds. A dow. response was exhibited in that the toticily of the samples was related to tW mortality at the time of hatching. Extracts of aflatoxin-free pcaa* products were found to I>e nontoxie to the chicken embryo. The addition «f aflatoxin B, to such nncontaminatc4 extracts produced the expected toxicjrt in the embryos. The injection of tj, tracts from contaminated peanut pr«4. The sensitivity of the chicken embryo to aflatoxins was reported in 1962 by Platt, et d. (1) who observed that preparations of "groundnut toxin" injected into the yolk of 5-day old chicken embryos caused deaths, and that the quantities required were about i/200th of those required for a positive result in the day-old duckling. An investigation of the feasibility of using ihe chicken embryo for a bioassay of aflatoxin toxicity is currently underway in our laboratories. The preliminary results reported here consider only the general svstemic toxicity of the aflatoxins to the chicken tmbryo. At the present time, the studies ire not sufficiently complete to verify whether the aflatoxins produce any specific pathological lesion in the embryos. injected at two or more levels when there was sufficient material available. More than 8,000 eggs have been used in these studies to date. The injections into the eggs were made by one of two routes: into the yolk, or into the air cell. The technique for injection into the yolk has been described previously (2). The volume injected into the yolk was 0.05 ml or less in all cases. For injections into the air cell, a hole of about 5 mm diameter was drilled in the shell over the air cell. The solution was then deposited on the egg membrane by a syringe, and the hole was sealed with adhesive cellophane tape. The eggs were allowed to rtmain undisturbed in a vertical position (air cell up) for about an hour to let the material disperse. The volume injected into the air cell was restricted to 0.04 ml or less. The solvent used for all injections was propylene glycol, which was known, from previous investigations (3), to be nontoxic in the eggs at the levels used. However, eggs were periodically injected with this solvent in appropriate amounts and incubated with the afiatoxin-injceted eggs. Noninjected controls and drilled-only controls were also included in the experiments to provide the necessary data on the background mortality. The e*gs were incubated (2) and candled daily from the fourth intubation day on, at which times all nonviablc embryos were removed and examined grossly. Results and Discussion Purified Aflatoxin Solutions and Mold Culture Extracts.—The toxicity of solutio?is of Experimental crystalline aflatoxin B, and aflatoxin G, 'to The sensitivity of the chicken embryo to the chicken embryo was first determined. ifbtoxins was. studied by injecting the followThe toxicity of aflatoxin B, to the chicken ing: (1) solutions of pure aflatoxin B, and embryo was greater when injected via the pore aflatoxin G,; (3) extracts of aflatoxin- air cell route than when injected into the producing mold cultures; (3) extracts of raw yolk. Figure 1 contains plots of the morpeanuts, roasted peanuts, peanut meal, and tality at 21 days due to the injection of peanut butter; and finally (4) aflatoxin B, several dose levels of aflrttoxin B, by both obtained from crude extracts by elution from the yolk and the air cell routes. The LD s 90 thin-layer chromatographic plates. obtained were 0.04S and 0.025 fig for the The solutions were injected into fertile yolk and air cell routes, respectively. niitlc Leghorn rggs before incubation, Aflatoxin (!,, which was injected into the f.ruups of at least 20 eggs were used at each yolk only, showed .1 lower toxicity to the ,j*c level of a sample, and each sample was chicken embryo than that obtained with �292 aflatoxin B,. The injection of 1.0 /ig of aflatoxin G, produced a mortality of 00% (at 21 days), while 2.0 /ig was required to produce a mortality of 90%. * « 0.1 £ * 3 0.05 Fig. I—Toxicity of otiatoxin Bj in the chicken embryo: mortality at 21 days. LDso: yolk, 0.048 fig; air cell, 0.025 /ig. Open squares: yolk injection. Closed squares: air cell injection. The toxicities of these extracts to the «bryos showed a good correlation with lie standard solution of aflatoxin B,. There w* no apparent alteration in the toxicity of afe, toxin Bt due to the presence of aflat«B» B2, G,, and G2 at the levels at which they occurred in these extracts. Aflatoxin-free Extracts of Peanut ft^g. ucts.—The usefulness of the chicken cmbn* in a bioassay of aflatoxins in peanut pn4. ucts depends on whether the constituent* rf aflatoxin-free peanut product extracts at inherent!}' toxic to the embryos. To determine the toxicity of these m^fcrials, extracts of raw peanuts, roasted panuts, peanut meal, and peanut butter, wfcxf, were found to be free of afiatoxins by ebon, cal analysis, were injected into eggs by brtfc the yolk and air cell routes. In most of that experiments the equivalent of original p^. nut product injected ranged from 1 to J • per egg. The toxicity was low for aj ^ these extracts; in general, it was equal to or only slightly higher tlian that of Ife background, which ranged from 0 to Jjjj mortality. One experiment was carried out with a extract of aflatoxin-free raw peanuts, m. jected by both the yolk and air cell in quantities corresponding to a raw equivalent of 1, 2, 4, and 8 g per egg. toxicity observed for the S g level was M significantly higher than that observed fe}, the 1 g level or that of the background. In the same experiment, known of the pure aflatoxin B, were added to raw peanut extract, and injected at the i levels and by both routes. This was determine whether the aflatoxin B, could be masked or enhanced by the cnce of increasing amounts of peanut rial. Separate groups of eggs were it with corresponding amounts of the stand*! aflatoxin B, solution for comparison. H« results of this experiment indicated thjt» little a* 25 ppb of aflatoxhi B, in the ofe. nal raw peanuts could be easily detailadministration of the extract by cif " Examination of nonsurviving embryos from eggs injected with aflatoxin Bt by either route revealed a severe growth retardation in most cases. In addition, edema, hemorrhage, underdevelopment of the mesencephalon (in embryos that died before the seventh day), mottled and granular liver surface, short legs, and slight clubbing of the down were also observed in many of these embryos. • Extracts of several cultures of aflatoxinproducing molds were used to determine the sensitivity of the chicken embryo to combinations of aflntoxins B,, G,, B.,, and G.,. The concentrations of these four constituents in the extracts were known from prior chemical analysis. The extracts were injected into eggs by both routes, in amounts designed to contain specific level? of aflatoxin B,, to compare their toxicities to (how of soluThis iiaper \vtis prosi'nh-d at tions of the pure aflatoxin B, at the same i-iKlith Annual Mwtiiis: of the Official Agricultural Cho:nists, Oot 19-*** dose levels. at Washington, IXC".. —• 293 route, and that the toxicity was not and air cell injections were made in embryos jniificantly different from that obtained from 1 to 18 days old. cjth the solution of pure aflatoxin B, of the These experiments revealed that, with both 0pic concentration. injection routes, the embryonic sensitivity to Aflatoxin-contaminated Peanut Products. the aflatoxin decreased rapidly as the em.-Extracts from raw or roasted peanuts, bryo age increased, and the maximum toxic panut meal, and peanut butter, which were effect was obtained with pre-incubation in£otrn to contain aflatoxins by chemical anal- jections. ps, were injected into eggs by both routes. Evaluation of Samjile Toxicity.—In the ffce amount of extract to be injected in the course of these studies it was also observed Mg was calculated on the basis of the afla- that the toxicity of aflatoxin B, injected at unn B, content determined chemically, irre- the higher dose levels was apparent early in jpedive of the amounts of aflatoxins B,, Gj, the incubation period, since most of the emini G. present. In most instances the results bryos did not survive beyond the eighth to aiiToborated the chemical analysis, since the tenth daj'. With lower dose levels, it is necgocity was comparable to that obtained essary to continue observations for the duranth Hie injection of equivalent amounts of tion of the 21 day period, since many em&e pure aflatoxin Bt solution. bryos survive longer than the tenth day but Afatoxin B, Obtained by Thin-Layer fail to hatch. In these instances an evaluaQffomatography (TLC).—In order to con- tion of the toxicity of a sample on the basis fcm that the toxieity of the contaminated of survivors at 8 or 10 days would be preOtracts was primarily due to their aflatoxin mature, and the true toxicity of a sample ^ content,. separate portions of some cx- might be underestimated. tuets were subjected to TLG and the reAcknowledgments »J(uig aflatoxin B, spots were removed from Ac plates, eluted, dissolved in propylene We wish to thank Man' K. Mutchler and fad, and injected into eggs. William F. Scott for their technical assistance Separate experiments were performed to in this work. itofirm that no toxic materials were derived The Division of Food Chemistry, Food and ton the TLC process itself. TLC "blanks" Drug Administration, supplied the pure aflagpcted into eggs in a similar manner had toxin Bj and Gt and the extracts of mold I wry l°w toxicity and .were comparable to cultures and peanut products, and performed the chemical analyses of these ex|*iground. Tlje toxicities of these eluted aflatoxin B, tracts. aots from more than 20 extracts of a variREFERENCES 4j of peanut products correlated very well ygli that of standard aflatoxin B, solution (1) Platt, B. S., Stewart, R. J. C., and Gupta, R, Pros. Nulr. Roc., 30, 21 (1062). «ded at the same dose level, and verified fet the toxicity of these extracts was, in (2) McLauKhlin, J., Jr., Mnrliac, J. P., Verrett, M. Jacqueline, Mutchlcr, Mary K., fael, primarily due to aflatoxin B,. and FilEhugh, 0. G., Toxicol. Appl. Phargmbryo Age and Aflaloxin Toxicity.—The . macol., S, 760 (1963). of aflatoxin B, to the chicken eni- (3) McLaiigliliu, J., Jr., Marliac, J. P., Verrett. various stages of incubation was also M. Jacqueline. Mulchlcr, Mary K, and d. Single injections into the yolk Fitzhugh, O. G., Am. Jnd. Ilyg. Assnc. J., made up to the fourth incubation day, 25,282 (10C4). Reprinted from the Journal of the Association of Official Agricultural Chemists, Vol. 47, Dwmbcr 1964. �294 295 THE ROLE OF "Toxic FAT" IN THE PBODTJCTION OF HYPBOPEBICABDHJM ASD ASCITES iw CHICKENS James B. Allen, D.V.M., Ph.D. SUMMARY blood counts were conducted on a portion of the sample, and the remainder was allowed to dot After centrifugation the serum was saved for total protein,* electrolyte concentration,7 and nonprotein nitrogen determinations' and electrophoretic studies.11 In experiment 1, surviving birds were used for vascular perfusion studies. Immediately after they were killed, the thoracic aorta was perfused with 5.0% dextrose to remove blood, and 1.0% silver nitrate was injected to outline the •cement substance" between the endothelial cells of the mesenteric capillaries. Slides were prepared of the mesenteric vessels and examined for alteration of interendotbelial silver precipitate in birds consuming toxic fat Five control and 5 test birds given the ration containing 1.0% toxic fat were selected for hemodynamic studies after 150 days on trial. Venous pressures were determined by making an incision through the skin over the jugular vein. The vein was dissected away from the surrounding tissue, and 2 ligatures were placed around the vessel. The superior legature was made secure, and a snail incision was made into the vessel. A No. 5 cardiac catheter was placed In the vein and made secure by the inferior ligature. The birds were placed ander a fluoroscope and the exact position of the catheter determined. The catheter was attached to a 3-mm. water manometer filled with physiologic Kline solution, thus enabling pressures to be determined in the venae cavae and right ventricles. At the termination of experiment 2, tissues were obtained for electron microscopy by severing the cervical spinal cord and exposing- the heart by removal of the sternum. Sections were taken from the left ventricle and cut into blocks of approximately 1 mm. with a razor blade. The tissues were Immediately jxed in 1.0% osmium tetroxide, dehydrated, and embedded as outlined by palade." Thin sections were cut on a cicrotome.1 Tissues were examined with an electron microscope ' operating at 50 kv. When "toxic fat" was added to the diet of experimental birds at concentrations from 0.25 to 5.0% for 35 to 150 days, edema of the myocardium, skeletal musculature, and lungs; hydropericardiurn; ascites; and foci of lymphoytes la the myocardium and epicardium were observed. Appreciable changes were not observed in the total serum protein levels, albumin: globulin ratio, electrolyte balance, or in the nonprotein nitrogen levels of the blood. There was dilation, edema, and lymphocytic infiltration of the heart. The myocardial mitochondria were vacuolated and shrunken. An increase in venous pressure was also noticed. The fluid imbalance observed in birds that consumed toxic fat' did not result from a decrease in total blood proteins or from an alteration in the albumin: globulin ratio, but was associated with cardiac decompensation and increased capillary permeability. Sehmittle et al.™ were the first to incriminate some fats as responsible for the production of hydropericardium and ascites in young chickens. Subsequent studies3'12 have demonstrated a reduction in growth rate, retarded sexual development, and increased mortality in pullets that have consumed toxic tat A marked reduction was observed in the hatchability of eggs from hens fed toxic fat" Turkeys and ducks appeared to be less susceptible than chickens to the detrimental effects of this fat in the diet* Fat-soluble tissue extracts from chickens fed toxic fat were capable of producing hydroperieardium and ascite* when added to the diet of other birds." Associated with the transudate, Sanger et al.™ observed necrosis of the liver, subepicardial hemorrhage, and lymphocytic infiltration of muscle fibers in chickens consuming toxic fat. In addition to these changes, Simpson et al.x also noticed bile duct hyperplasia and proliferation of the endothelium in the parenehymal tissues. The unsaponifiable portion of some batches of fat was found responsible for the toxicity.2'16-17-2* Repeated passages of this fraction through alumina and silica gel columns led to the isolation of 8a crystalline factor of unknown structure which was able to produce anasarca. The primary aim of this study was to investigate the mechanism by which toxic fat produces anasarca. Preliminary studies suggested that a vascular or cardiac injury may be responsible for the transudation of fluid into tissues. In the hope that some clarification of the mechanism might be obtained, arterial perfusions, recordings of hemodynamic changes, and ultrastructural myocardial studies were undertaken. MATERIALS AND METHODS In experiment 1 (acute), 100 White Leghorn cockerels, 1 day old, were separated into 5 groups of 20, placed in heated batteries, and fed rations containing 0, 0.5, 1.0, 3.0, and 5.0% toxic fat* for 35 days. Because the exact chemical nature of toxic fat is not known, the amount of the toxic material undoubtedly varies in fats from different sources. The fat used in these experiments was from the same source, identical shipments, and had the samd LDM wbea fed to day-old chickens. In experiment 2 (chronic), forty-eight 4-week-old cockerels of comparable weight were given, for 150 days, a ration** containing 0, 0.25, and 1.0% toxie fat The birds were fed and watered daily. Throughout the trial period the general appearance, food consumptions, and mortality were spleen, pancreaa, kidneys, adrenal glands, skeletal muscle, brain, and bone marrow were fixed in 10.0% neutral formalin, embedded in paraffin, sectioned at 7 ft, and stained with hematoxylin and eosin. Frozen sections were prepared from the liven and kidneys and stained with Sudan IV to demonstrate neutral fats. At the termination of the trials, blood was obtained from the cephalic vein, complete Received for publication Jan. 6, 1964. From the Department of Patholocy, University of Wisconsin Medical School, Madbv. This Investigation was supported In part by Public Health Service research er*M HP-8985 from the National Heart Institute, Public Health Service. *w The author is indebted to Mr. Homer Montague for the photography. • Emery Industries. Cincinnati. Ohio. •« McMlllen Feed Mills, Fort Wayne, Ind. BEST7LTS In experiment 1, birds given rations containing 0 to 1.0% toxic fat survived, whereas those consuming rations with 3.0 and 5.0% toxic fat had a 25.0 to E.0% mortality, respectively (Table 1). As the level of toxic fat in the diet TABLE 1.—EFFECTS OF TOXIC FAT CONSUMPTION ON CHICKENS Toxic fat in diet (percent) No. birds « y m 9 • g g I I . , 0 0.5 1.0 3.0 5.0 0 0.25 0.5 1.0 Days on trial No. died 35 35 35 35 35 150 150 150 150 Hydropericardium 0 0 0 11 0 1 4 6 Ascites 20 20 0 0 2 20 20 0 4 8 11 increased, there was a corresponding decrease in growth rate. The control averaged 348 Gm., whereas those given rations with toxic fat averaged J78 Gm., at the end of 35 days. Hydroperieardium and ascites were not Served in the chickens fed rations containing 0 and 0.5% toxic fat Pericgrdial fluid volumes ranged from 0.5 to 5.0 ml. in the birds given rations with 10 to 0.5% fat Ascites was also observed in these groups, but the volume was «t determined due to the partial coagulation of the transudate. In experiment 2, 6 birds fed 1.0%, 4 fed 0.5%, and 1 fed 0.25% toxic fat in the ration died. The mean survival time for the groups was 80 ± 20, 100 ± It, and 135 days, respectively. The control group gained an average of 16.5 6*., whereas the survivors of the group fed the ration containing 1.0% toxic fct gained 14.9 Gm. per day. The weight gain of the surviving birds of the > Porter-Blum mlcro+ome. Ivan Sorvall. Inc., Norwalk. Conn. »BCA, EMU-3G, Radio Corporation of America, Camden, N.J. �296 297 groups fed rations containing 0.5 and 0.25% toxic fat was almost comparable to that of the controls. Hydropericardium was observed in 11 and ascites in 2 Jbirds fed rations containing 1.0% toxic fat Eight of 12 birds on the 0.5% toxic fat ration had hydropericardium but were free of ascites. The average volume of pericardial transudate in the test birds was 13.5 ml. No appreciable difference was noticed in the organ weights of the test and control birds, with the exception of the testes and hearts of the birds in experiment 2. The average weight per testis of birds in the control group was 14.9 Gm., whereas that of the 0.25% group was 6.6 Gm., the 0.5% group was 4.1 Gm., and the 1.0% group was 3.9 Gm. The hearts of birds of the control group averaged 15.0 Gm., and those of birds in the test groups averaged 23.0 Gm. and were markedly dilated in most cases (Fig. 8). Tissue sections of the various organs were examined microscopically. The birds with hydropericardium had fibrinous deposits and foci of lymphocytes on the visseral pericardium. The muscle fibers were separated by edematous fluid. A number of the small myocardial artieries were surrounded by lymphocytes. There were also foci of lymphocytes between the myocardial fibers (Fig. 12). The lungs were congested and had a moderate amount of peribronchial lymphoid hyperplasia. Pulmonary edema was found in those birds which died during the experiments. Livers from birds with marked ascites frequently had thickened capsules. In many cases there was coagulation of the transudate on the convex surface of the liver which formed a false capsule (Fig. 3). There was moderate fatty infiltration of the hepatic cells in birds given toxic fat in experiments 1 and 2 (Fig. 4). Extensive lymphoid hyperplasia around the periportal areas was found consistently (Fig. 5). Five of the test birds in experiment 2 also had myeloid hyperplasia scattered throughout the parenchymal tissue of the liver. Microscopic examination of the testes of the birds given toxic fat in experiment 2 revealed a reduction in size of the seminiferous tubules. The Sertoli cells and spermatogonia appeared normal, but there was a reduction of the primary and secondary spermatocytes. These maturing cells were reduced to the point where no spermatids and spermatozoa were observed. Cells in the testes of the test birds appeared normal. The major difference between testes and 143 mEq./liter in the control birds. The potassium levels were 5.3 mEq./liter in the test birds and 5.5 mEq./liter in the control birds. When 1.0% silver nitrate solution was perfused in the thoracic aorta and the mesenteric capillaries examined microscopically, a difference was found in the cement substance located between the endothelial cells of the test and control birds in experiment 1. The control group had uniform diamond-shaped bands of silver precipitate between the capillary endothelial cells, whereas the bands id the test birds were irregular and indistinct (Fig. 9,10). Catheterization of the heart revealed the mean right ventricular pressure of the control birds to have an average value of 15.3 Cm. of water, whereas the test birds'had an average of 21.3 Cm. of water. The pressures in the inferior rena cava averaged 5.5 cm. of water in the control birds and 7.1 cm. of water In the test birds (Table 3). TABLE 2.—SERUM PROTEIN STUDIES ON CHICKENS FED TOXIC FAT Toxic fat in diet (percent) 0 ... 0.5 1.0 3.0 5.0 . Globulin Scrum protein Albumin A/G (Gm./100cc.) (Gm./lOO ec.) (Gm./100cc.) 3.36 3.24 3.14 3.08 3.06 Nonprotein nitrogen (mg./100 cc.) Sodium (mEq./liter) Potassim (mEq./litei) 2.30 46 21.7 143 5.5 0.94 2.12 44 20.7 146 i'j 1.06 of birds in test and control groups was a lack of spermatogenesis and a marked reduction in tubular size (Fig. 6,7) in testes of test birds. Total blood proteins were determined by the micro-Kjeldahl method. The control group of experiment 1 had an average value of 3.36 Gm./lOO cc. The 0.5, 1.0, 3.0 and 5.0% groups had values of 3.24, 3.14, 3.08, and 3.08 Gm./100 cc. of protein, respectively (Table 2). In experiment 2, it was found that the groups given rations containing 1.0% toxic fat had an average of 44 Gm./lOO cc., the 0.25% group averaged 4.5 Gm./100 cc., and the controls had an average value of 4.8 Gm./100 cc. of serum protein. The protein level of the aseitic fluid of the experimental birds was 1.7 Gm./100 ec., with no appreciable difference in the levels of the various groups. The albumin: globulin ratio wag determined by paper electrophoresis, and little difference was found in the protein ratio of the various groups. Sodium, potassium, and nonprotein nitrogen levels were determined on senna samples from 6 birds given rations containing 5.0% toxic fat and 6 control birds. No appreciable difference was found in the nonprotein nitrogen levels <tf the 2 groups, with the controls averaging 21.5 mg./100 cc. and the test bird* • 20.7 mg./lOO cc. The sodium values were 146 mEq./liter in the experimental TABLE 3.-HEMODYNAMIC ALTERATIONS DUE TO TOXIC FAT CONSUMPTION IN CHICKENS Mean right ventricular pressure (cm. Hrf) Toxic fat in diet (percent) Mean superior vena cava pressure (cm. HjO) Mean inferior vena cava pressure (cm. HjO) 0 15.4 5.6 5.2 0 0 15.2 15.0 5.2 5.8 10 5.0 5.0 22.2 20.0 21.0 7.4 7.0 7.2 5.6 5.8 5.7 7.0 71 6.9 Preliminary data obtained from electron micographs of heart muscle from 10 test and 10 control birds of experiment 2 indicted that the major changes occurred in the mitochondria of the cardiac muscle cells. In some cases the mitochondria were markedly shrunken and raeuolated with indistinct cristae. In the birds with more advanced cases, many of the mitochondria had disappeared, leaving large areas devoid of any organelle. DISCUSSION There was normal development of all organs with the exception of the testes of birds in experiment 2, which were markedly reduced in size. Kumaran and Turner10 outlined the tubular and spermatogenic development in chickens at various states of maturity. It would appear that the consumption of toxic fat in these experiments retarded testicular development by approximately 2 months in the 6-month-old roosters. Despite the 'reduction in testicular size, secondary sex characteristics such as comb size and body conformation were unaffected. In earlier studies,1 young birds were killed every other day for 3 weeks to determine when hydropericardium or ascites developed. Hydropericardium was usually noticed about the 16th day after feeding toxic fat in the ration was Initiated, without ascites or pulmonary edema. Fluid accumulation in the lungs and peritoneal cavity invariably developed later. When the concentration of toxic fat in experiment 2 was reduced to a low level, the birds failed to develop ascites, but hydropericardium was a consistent observation. The most prominent lesions at necropsy were hydropericardium, ascites, and pulmonary edema. A number of procedures were conducted to resolve what caused the anasarca. Birds with ascites and hydropericardium in many cases bad higher serum protein levels than those in the control group. The albumin : globulin ratios of experimental and control birds were approximately comparable. The nonprotein nitrogen and electrolyte levels of the blood were within normal range in the test groups. There was a reduction in the cement substance in the experimental birds with hydropericardium and ascites. Since a large quantity of fluid escapes between the endothelial cells rather than through them" any alteration in the cementum would affect the permeability of the vessels.1' Whether this alteration in the cement substance of the capil- �298 299 laries wajs associated with hypoxia of the endothelial cells resulting from cardiac decompensation or from a direct effect of toxic fat on the capillary manbrane remains to be clarified. " Hemodynamic studies revealed an increase in right ventricular and vena cava pressures in the experimental birds. Associated with these pressure changes were dilatation and hypertrophy of the right side of the heart of birds given toxic fat Pulmonary edema, cardiac dilatation, and increased venous pressure are indicative of cardiac decompensation. What do these data reveal in regard to the causes for the excessive ertravascular fluids? The lack of any appreciable change in the total blood protete and albumin: globulin ratio would indicate that the liver was producing adequate amounts of albumin. There were no marked alterations in kidney function, because the serum albumin and nonprotein nitrogen levels were conparable in the 2 groups, which suggests that the kidney tubules and adrenal cortex were not affected by toxic fat ingestion. The microscopic observation fortify the biochemical data, because the alterations in the liver, kidneys, and adrenal glands of test birds did not appear of sufficient magnitude to account for the anasarca. Since myocardial fibers are very active, there is a constant demand on the respiratory enzymes located in the numerous mitochondria between the myofibrils. A rather high metabolic demand is implied, because the ratio at mitochondria in cardiac muscle to skeletal muscle is approximately 500:1* Any significant alteration in the mitochondria would affect the respiratory enzymes and sooner or later lead to cardiac failure. Certainly this wooU appear to be a logical explanation for what occurs in birds consuming toxfc fat. As the heart becomes less efficient, there will be an increase in hydrostatfc pressure in the veins and capillaries, with a predisposition for the extravaattion of fluid into the tissues and body cavities. 14. Naber, E. C., Bletner, J. K., and Touchburn, S. P.: Effect of Certain Toxic Fats and Their Derivatives on Growth, Reproduction, Embryonic Development, and Health of Chickens. Poult Sci, 37, (1958): 1229. 15. Palade, G. E.: Study of Fixation for Electron Microscopy. J. Exptl. Med., 95, (1952) : 285-288. 1& Pappenheimer, J. R.: Passage of Molecules Through Capillary Walls. Physiol. Rev., 33, (1953): 387-423. 17. Potter, G. C., Brew, W. B., Patterson, P. L., and Sipos, E.: Current Status of the Toxic Fat Principle Causing the Chick Edema Syndrome. J. Am. (HI Chem. Soc., 36, (1959) : 214r-217. la Sanger, V. L., Scott, I*, Hamdy, A., Gale C., and Pounden, W. D.: Alimentary Toxemia in Chickens. J.A.V.M.A., 133, (Aug. 1,1958) : 172-176. 19. Schmittle, S. C., Edwards, H. M., and Morris, D.: A Disorder of Chickens probably Due to a Toxic Feed—Preliminary Report J.A.V.M.A., 132, (March 1, 1958): 216-219. 20. Simpson, C. F., Pritchard, W. R., and Harms, R. H.: An Endotheliosis in Chickens and Turkeys Caused by an Unidentified Dietary Factor. J.A.V.M.A., 134, (May 1, 1959): 410-41ft 21 Williams, F. G., Pickets, E. G., and Curren, E. L.: Improved Hanging-strip paper Electrophoresis Technique Science, 121, (1955) : 829-830. 22. Wooton, J. C., and Alexander, J. C.: Some Chemical Characteristics of the Chick Edema Disease Factor. J. Assoc. Off. Agric. Chem., 42, (1959) : 141M& REFEBENCES 1 Allen, J. R.: The Role of "Toxic Fat" in the Production of HydroperiMn. . dium and Ascites. Ph.D. Dissertation, University of Wisconsin, Madison, Jane, 1961. Dissertation Abst, 22, (1961) : 545. 2. Brew, W. B., Dare, J. B., Benedict, J. H., Potter, G. C., and Sipos, R : Characterization of a Type of Unidentified Compound Producing Edema hi Chickens. J. Assoc. Off. Agric. Chem., 42, (1959) : 120-128. 2. Dunahoo, W. S., Edwards, H. M., Schmittle, S. C., and Fuller, H. JJLStudies on Toxic Fat in the Ration of Laying Hens and Pullets. Poult, Sci, M (1959) : 663-667. 4. Edgar, S. A., Bond, D. S., Melius, P., and Ingram, G. R.: The Effect of » Toxic Substance in Fat on Poultry. Poult Sci., 37, (1958) : 1200. 5. Folin, O., and Wu, H.: Revised Colorimetric Method for Determination «f Uric Acid in Urine. J. Biol, Chem., 38, (1919) : 469-460. 6. Friedman, L., Firestone, D., Horwitz, W., Barnes D., Anstead, M., aai Shue, G. : Studies of the Chick Edema Factor. J. Assoc. Off. Agric. Chem. 4* (1959) : 129-140. 7. Hald, P. M.: The Flame Photometer for the Measurement of Sodium «H Potassium in Biological Material. J. Biol. Chem., 167, (1947) : 499-510. 8. Hannan, R. E., Davis, G. E., Ott, W. H., Brink, N. G., and Kuehl, F. A.; The Isolation and Characterization of the Chicken Edema Factor. J. *n Chem. Soc., 82, (1960) : 2078-2079. 9. Kingsley, G. R.: The Direct Biuret Method for the Determination 4 Serum Protein as Applied to Photoelectric and Visual Colorimetry. J. Lab. 4 Clin. Med., 27, (1942) : 840-S45. 10. Kumaran, J. D. S., and Turner, C.: The Normal Development of TeMH in Plymouth Rocks. Poult, ScL, 28, (1949) : 511-520. 11. Kuschner, M., and Lordell, D. H.: The Pathology of Congestive Heuf Failure. J. Chron. Dis., 9, (1959) : 424-441. 12. Machlin, L. J., Gordon, B. S., Meisky, K. A., and Maddy, K. H.: Bel*. tionship of Exudative Degradation to Toxicity in Certain Fats. Poult ScL, M. (1959) : 579-585. ^ 13. Marches!, V. T.: The Passage of Colloidal Carbon Through Endothelium. Proc. Roy. Soc. Biol., 156, (1962) : 550-552. STTMMABIO IN INTEBLINGUA Le Rolo de "Grassia Toxic" in le Production de Hydropericardio e Ascites in Gallinas Qnandro "grassia toxic" esseva addite al dieta de aves experimental a concentrationes de 0,25 a 6,0 pro cento durante 35 a 150 dies, le sequente alteraUones esseva observate: (1) Edema del myocardio, del musculatura skeletic, e del pulmones; (2) hydropericardio; (3) ascites; e (4) focos de lymphocytes in le myocardio e le epicardio. Appreciable alterationes non esseva observate in It nivellos de proteina total in le sero, in le proportion albumina a globulina, b le balancia electrolytic, e in le nivellos de nitrogeno non ligate a proteina in le sanguine. Esseva notate dilatation, edema, e infiltration lymphocytic del forde. Le mitochondrios myocardial esseva vacuolate e contrahite. Un augaento del tension venose esseva etiam notate. Le imbalancia de liquido obsernte in aves que consumeva grassia toxic non resultava ab un decline in total •oteinas de sanguine o ab un alteration in le proportion albumina a globuina led esseva associate con discernpensation cardiac e un augmento del permeabilHate capillar. INDUSTEIAIXT ACQUIRED POBPKTBIA Twenty-nine patients working in a chemical factory engaged in the manufacnre of 2,4-dichlorophenol (2,4-D) and 2,4,5-trichlorophenol (2,4,5-T) exhibiting features of chloracne were studied for the presence of porphyria cutanea Hida. In 11 cases urinary uroporphyrins were elevated. Two of these patients who showed evidence of acquired porphyria with tiloracne were hospitalized. The features of chloracne as well as the clinical tad laboratory features of acquired porphyria have been discussed. There appeared to be an etiologic but not quantitative relationship between the chloraeoe in workers 'engaged in the manufacture of 2,4-D and 2,4,5-T and porphyife cntanea tarda of the acquired type. It is our feeling that either the feigned chemicals or some intermediate are responsible for both diseases. Since Waldenstrom first implied that porphyria cutanea tarda might be acquired, a growing number of chemials have been implicated in the pathogeuris of this disease. These chemicals have included alcohol, sedatives, fungidfes, etc.1"" While treating a severe outbreak of chloracne in a factory which otnofactures 2,4-D and 2,4,5-T, a number of workers were noted to have Ipperpigmentation, hirsutism, fragility of the skin and vesiculobullous ernpIKos on exposed areas of skin, together with cutaneous findings of chloracne. Instigation revealed evidence of porphyria cutanea tarda of varying degrees «f aeverlty in 11 out of 29 workers investigated. Porphyria cutanea tarda has gerer before been described as related to chloracne, nor has it been ascribed to 45-362 O—70 20 �301 300 Industrial exposure in the United States. This outbreak is therefore of inters* in adding more evidence to the growing concept that porphyria cutanea tarda may be an acquired disease occurring after various insults to the liver. Three cases were studied in detail. From the Departments of Dermatology and Medicine, Newark Beth Israel Hospital. Chief of Dermatology, Newark Beth Israel Hospital (Dr. Bleiberg); Senior Resident Physician in Medicine, Newark Beth Israel Hospital (Dr. Wallen)Assistant in Dermatology, Newark Beth Israel Hospital (Dr. Brodkin) ; Director of Medical Education and Consultant in Medicine, Newark Beth Israel Hospital (Dr. Applebaum). BEPOBT OF CASES CASE 1.—A 48-year-old white male who was employed at the factory tat I three years as a chemical operator. His work brought him into Intimate con. tact with the suspected chemicals. His past history included two attacks of bfl. iary colic prior to 1953. He was never a heavy user of alcohol. A diagnosis of cholecystitis had been made and a cholecystectomy was performed early fe 1953. After this he came to work at the factory In question. In 1956 he begu to notice some darkening of his skin and suffered right upper quadrant pain. A diagnosis of common duct obstruction was made, and this patient was operated on again in January of 1956. An unsuccessful attempt was made to probe the common duct, and no further operative procedure was done. During the postoperative course, this man received 2 gm of barbiturates. The patient stated that his urine had turned "the color of Coca-Cola" at least one year prior to the second operation. That spring, an eruption of bullae appeared <• the face, ears, and hands. These lesions could be produced either by exposure to the sun or by pressure In addition to the vesicular eruption, the patient noted progressive darkening of his skin and marked hirsutism, especially over the temples. Inspection of the urine revealed a Coca-Cola coloration and, under the Wood's Light, a brilliant red fluorescence. The exact laboratory data OB this patient are no longer available except for the presence of quantitatively markedly increased excretion of urinary porphyrins Including uroporphyrina, coproporphyrins, and porphobilinogen. This man is now alive and well and apparently is suffering minimal if anj symptoms of porphyrim cutanea tarda. His present job does not entail the oat of any chemicals. He has failed to present himself for further testing. CASE 2.—This is a 60-year-old white male who has been employed in the factory for seven years as a welder. In the course of his work, which consisted of welding tanks and pipes, he was brought into frequent and prolonged contact with chemicals. He was admitted to the Newark Beth Israel Hospital tat investigation. He stated that three months prior to admission, he had noted u increased darkening of the skin, thickening of the eyebrows, and a darkening and reddening of his urine. His family history and his past medical history were unrevealing, except for moderately heavy alcohol intake for some yean. Physical examination revealed numerous comedones and small epidermoid cysts and furuncles the face, chest, and shoulders; There was intense grayishbrown hyperpigmentation with a purplish tint on the exposed surfaces of the face, neck, chest, and hands and moderate hypertriehosis of the temples. The scalp hair showed a lusterless, dull silver color change. The liver edge wa» palpable about 3 cm below the right costal margin and was smooth and noatender. The remainder of the physical examination was within normal limita, A casual urine specimen revealed a strong tea color with a deep fluorescence, reddish, under the Wood's light Laboratory studies revealed increased urinary uroporphyin, coproporphyrin, and urobilinogen excretion. There was no demonstrable porphobilinogen. The feces showed increased uroporphyrins and coproporphyrins. All porphyrhi determinations were qualitative and done by the Watson-Schwartz method. Other significant findings included an elevated serum glutamic oxaloacetie transaminase ranging between 41 and 51 units on five different days. Sena glutamic pyruvic transaminase on corresponding days ranged between 53 and 64 units. The sulfobromophthalein retention was 6% in 30 minutes. The erythrocyte sedimentation rate (Westergren) was 94 mm in the first hour. AB other studies, which included complete blood count, bleeding and clotting tune, urinalysis, glucose tolerance test, serum bilirnbin, blood urea nitrogen, total proteins, albumin-globulin ratio, serum cholestrol, alkaline phosphatase, cephalin locculation, and thymol turbidity, serum electrolytes, and CO, combining power, as well as serological tests for syphilis were all within normal limits. Electrocardiograms and chest x-rays were normal. A liver biopsy was performed and the specimen immersed in isotonic saline. It fluoresced intensely nider the Wood's light The red pigment diffused out Into the saline, so that the entire tube fluoresced. The microscopic examination of the liver specimen revealed parenchymal cell regeneration and hemofuscin deposition. A skin Ifcpsy from clinically hyperpigmented postauricular skin showed a normal epidermis except for the dermoepidermal border, where there was a striking depo4Uon of brown granular pigment In addition, there was mild infiltrate of gpaii round cells in the dermis. No sebaceous glands were visible in the sections. Shortly after discharge from the hospital in June, 1963, the patient was Heated for a chromic trichophytosis of the feet with griseofulvin 0.5 gm twice filly. About four days after the onset of this treatment, a severe vesiculobulloos eruption on the dorsal surface of the hands appeared. The griseofulvin ms stopped, but the eruption progressed for another two weeka Healing time fta very prolonged, and at present, residual atrophic scarring is visible on both kinds (Fig 1). In the scars occasional milia are seen. SUMMARY OF DATA FOR 26 WORKERS WHOSE URINE WAS TESTED FOR PORPHYRINS Hyperpigmentation Chloracne' Flfcnt . Mild : ..... do Severe lld " -,; *- - -*> Severe Mild..... . Moderate. Severe . None Mild do None do . Mild. Moderate... Mild Moderate... do None do do jjT . Moderate B_: ..... Severe m_ Mild... Mild. Moderate Mild Moderate... Marked-Mild. None at::::. Nonedo ""::: None"-":::::::::::::do::::: do do ' do; do Hirsutism . Moderate... None do Mild do . None ..do. Severe . Mild . Moderate do . None do do . Moderate... . Marked.... . None . Marked . None . Moderate... . None do do do do..... do Urine Uroporphyrins Pos. None do... do... Pos. . None Pos.. None. do... do do... do... do... . Pos None.... do... do... —do— Pos. Pos.. None.... do... do— do... Pos Pos Chemical > Contact . Moderate.. . Severe do.... dodo "Moderate^: . Moderate.. do.... do . Severe.—. . Moderate.. . Severe do . Moderate... do I... . Severe . Mild do do— do do do do do do Skin Fragility Pos. Net Do. Pos. "DO. Neg. Do. Do. Do. Do. Do. Do. Do. Do. Pos. 'Do. Do. Do. Do. Do. Do. Do. Do. Do. Do. •Severity of chloracne is judged on the presence of comedones, epidermoid cists, and furuncles and pustules. Hit extent of exposure is difficult to truly judge because of such variables as personal hygiene and work habits. > Brief period of employment OH 2,4-Dichbrophenol OH 2,4-5-TrichlorophenoI Hexachlorobenzene 3.—Comparison of structural formulae of weed killer manufactured and fungicide responsible for acquired cutanea tarda. Note the similarities CASE 3-—This is a 48-year-old white male employed at the factory for eight jeers mixing batches of chemicals. During the past two years, he had developed hyperpigmentation of the exposed skin of the face and hands. There was �303 302 of ingestion of hexachlorbenzene,1- * chemically a closely related comaoond (Fig 3). This would lend support to the concept that porphyria cutanea is not necessarily genetically produced, unless the genetic defect is an dtremely common one. An analysis of the table of the 26 surveyed workers (Table 4) reveals: the —rerity of chloracne does not usually correspond to the degree of exposure to Itonicals (patients \, 8, 11, 19 or patients 2, 3, 5, 10, 12, 13, 16). The severity STnrophyria does not usually correspond to the degree of chemical exposure rtatients 7, 20, 25, 26 or patients 2, 3, 4, 10, 12, 13, 16). The severity of chloracBe does not usually correspond to the presence of porphyria (patients 1, 4, 8, iTor patients 5, 20, 25, 26). Therefore it would appear that there is some ^dividual susceptibility to these disease. It has been observed in general that: (1) Patients with adolescent acne tend to get worse chloracne: (2) Possibly •revious liver damage (alcoholism, etc) predisposes to porphyria. Also there .nst be in these cases some etiologic relationship between chloracne and por•fcyria since a relatively large number of both diseases began to appear and hare persisted at the same time. On the basis of the elevated transaminase levels and the histological signs of Hrer cell regeneration in the liver biopsies, it may be assumed that the basis •f the distrubed porphyrin metabolism is a hepatotoxic effect of one or more ef the chemicals in this factory environment The synergistie roles of other biown liver toxins such as alcohol and barbiturates, or griscofulvin ( 1 case), cannot be overlooked. ^nre would like to express appreciation for the help offered by Dr. Donald J* Birmingham of the division of Industrial Dermatology of the U.S. Public Health Service, to Dr. Marcus Key, and to Dr. Norman Olivier. Jacob Bleiberg, MD, 40 Union Ave, Irvington, NJ. marked Mrsutism which involved the temples. The dull silvery tint of the hahr was visible. He stated that in the past he had had episodes of blistering of the .exposed skin. He also had noticed that his urine was dark on voiding. Hto family history was noneontributory. The physical examination revealed aa intense hyperpigmentation of the face, neck, and hands. There was severe Mrsutism involving the eyelids, eyebrows, and lateral aspects of the foreheal (Fig 2). Comdones and small epidermoid cysts were very prominent, and then were numerous furuncles scattered over the entire body. The remainder of the physical examination was within normal limits except for prolapsed hemor. rhoids. The following laboratory studies were within normal limits: complete blood cell count, urinalysis, bleeding and clotting tune, prothrombin tint blood glucose tolerance test, urea nitrogen, cholestrol, bilirubin, alkaline pho» phatase, total protein and albumin-globulin ratio, cephalin flocculation, thymol turbidity, serum electrolytes including sodium potassium, chlorides, COi COB. bining power, calcium, and phosphorus. The serum glutamic oxaloacetic trana*. minase on five successive days ranged between 39 and 56 units while the serum glutamic pyruvic transaminase on corresponding days ranged betweet 47 and 72 units. The sulfobromophthalein retention was S% in 30 minutes. The electrocardiogram was normal. The chest x-ray revealed a diffuse nodular infiltration of both lungs due to pneumoconiosis. This was consistent with the patient's history of having worked a number of years as a coal miner. The plain film of the abdomen was negative. The urine revealed a negattte Watson-Schwartz test The urine failed to fluoresce under the "Wood's light The erythrocyte sedimentation rate (Westergren) was 24 mm in the first hour. A liver biopsy was performed and the specimen immersed in saline. Under the Wood's light the specimen and saline in which it was immersed fluoreseet faintly. On microscopic examination, the liver biopsy showed evidence of lit^ cell regeneration and hemofuscin deposition. A skin biopsy showed brown granular pigmentation as the basal margin of the epidermis. There was a mflj chronic inflammatory infiltrate scattered through the dermis. No sebaceow glands were visible in the sections. Since the man's chloracne has been so severe, he had been removed frcw contact with chemicals two years prior to his admission to the hospital Thfc probably was responsible for the failure to prove qualitative chemical evidence of porphyrins in the urine. It also may indicate that acquired porphyria cut*. nea tarda is reversible. EEFEBENCES • Waldenstrom, J. : The Porphyrias as Inborn Errors of Metabolism, Amer J Med ^iCw^C.': Cutaneous Porphyria Related to Intoxication, Dlrfn (Istanbul) 34:11-15, 'fschmld, K. : Cutaneous Porphyria In Turkey, New Eng J Med 283 :397-398, 1960. «Brunsting, L. A. : Observations on Porphyria Cutanea Tarda, Arch Derm Syph "•'Watson,' C. I.': The Porphyrias, Advances In Internal Medicine, Chicago : Tear Book Mdlshers, Inc., 1954, vol. 6. «B»rnes, H. D. : Porphyria In Bantu Races on the Wltwatersrand, S Afr Med J »:T81-784, T>1955. j *"_•_« __ i jr An SCREENING TESTS Twenty-six additional men working at this chemical factory were studied «• an ambulatory basis. In addition to routine urinalysis, each urine spetimeB was tested for uroporphyrin by the Watson-Schwartz method. Eight out of the 26 manifested significantly increased excretion of urinary uroporphyrins by the Watson-Schwartz method. If the three cases described in the case reports above are added, this is a total of 11 cases of porphyria cutanea tarda at varying degrees of severity out of 29 patients tested, or 37+% (Table) . me Roy Soc (Blol) 143:257-279, 1955. • Goldberg, A., and Eimlngton, C.: Diseases of Porphyrin Metabolism: American Lec-D* Series, Springfield, 111.: Charles C Thomas, Publisher, 1962, p 194. • Solomon, H. M., and Elgge, F. H. J.: Disturbance in Porphyrin Metabolism Caused to Feeding Methyl l,4-Dihydro-2,4,6-Trimethypyridtne-3,5-Decarboxylate, Proc Soc Exp SLlMed 100:583^586, 1959. n De Mattels, F.; Prior, B. E.; and Rimlngton, C.: Nervous and Biochemical Dlsturbtftt* Following Hexachlorobenzene Intoxication, Nature (London) 191:363—366, 1961. B Watson, C. J.: The Problem of Porphyria—-Some Facts and Questions, New Eng J MM 263:1205-1215, 1960. •TIo, T. H., et al.: Acquired Porphyria From Liver Tumor, Clln Scl 16:517-527, 1957. COMMENT Hyperpigmentation in these workers was limited to the sun-exposed area* of the head, neck, and hands. It was more frequently observed in the Nea* patients involved. The degree of hyperpigmentation was roughly proportional to the severity of the chloracne The hyperpigmentation varies from mild redness in extremely fair individuals to dark gray intense dusky bronzing of the skin. The degree of hirsutism was also proportional to the severity of the chloracne. This too was quite variable in degree but always involved the tea. pies between the lateral half of the eyebrow and the temporal hair of the scalp. The hirsutism in a few cases, notably case 3, extended beyond this airi involved both the upper and lower eyelids. The hair was of approximately the same texture and density as that of the eyebrows. The occupational environment of these men consists of a group of baait . chemicals including acetic acid, phenol, monochloracetic acid and sodium hydroxide, plus the finished products 2,4-D and 2,4,5-T as well as maw unknown intermediary products. It is known that one of the intermediaries]* a highly volatile chlorinated phenolic ether which contains six chlorine atom*. -This particular compound, because of its volatility, is strongly suspected at being a possible causal agent Porphyria has been described in many cases as a EIECTBON MICROSCOPIC ALTERATIONS IN THE LIVER OF CHICKENS FED Toxic FAT* Toxic fat is the name applied to certain fats that produce hydropericardium, fcydrothorax, and ascites when added to the diet of chickens.*- "•" It has been fcmonstrated that the toxic fraction & associated with the unsaponifiable portion of the fat10-" The toxicity can be increased by repeated passages silica and alumina gel columns.'-* Crystalline preparations of the fraction have been prepared, but the chemical nature of the compound Binains to be identified." Accepted tor publication March 21, 1966. Prom tlie Department of Pathology and the Beglonal Primate Research Center, Unl•ifiJtr of Wisconsin. Madison. Wisconsin. • This research was supported In part by grants HE-OS681 and FR-0167 from the j»tkmal Institutes of Health. I �, 304 305 The following alterations that resulted from the consumption of toxic fat hare been reported. Sanger, Scott, Hamdy, Gale, and Pounden1" obserred foctl necrosis of the liver, epicardial hemorrhage, and lymphocytic infiltration of the cardiac muscle. Simpson, Pritchard, and Harms" reported bile duct hyperpU. sia and proliferation of the endothelial lining of the smaller blood vessel*. Allen and Lalieh* observed testicular hypoplasia. Marked dilation and edemt of the myocardium and altered right ventricular and vena cava pressures haw been demonstrated.* The present experiment was initiated to determine the morphologic effect «f toxic fat upon the liver of chickens and to correlate these findings with tW development of anasarca. EXPERIMENTAL PROCEDURE . One hundred sixty-eight 1-day-old White Leghorn-New Hampshire chicken were divided into groups of 48 and 120 chickens. One-half of the chickens fron each group were given a commercial diet containing 3.0 per cent toxic fat (Emery Industries, Inc., Cincinnati, Ohio), while the remaining chickeiw received a comparable diet which contained 3.0 per cent corn oil. In the group of 48 chickens, three control and three experimental chickens were killed every other day for 16 days and sections of the liver were obtained for electro* microscopic evaluation. The group of 120 chickens remained on the toxic fat diet until the LDi, wa, established. From the survivors, blood was obtained for hematocrit," hemoglobin,8 total serum protein," serum electrophoretic pattern,23 and electrolyte studio." The ascitic fluid was also collected for total protein and electrophoretic patten determinations. All chickens that died during the course of the experiment as well as those sacrificed at its completion were necropsied. Portions of the liter were fixed in 10 per cent buffered formalin for 24 hours, dehydrated, embeddei in paraffin, sectioned and stained with hematoxylin and eosin. Other portion of the liver were sectioned on a cryostat and stained with Sudan IV for ne»tral fats. Additional paraffin-embedded tissues were stained with aniline bltt and phosphotungstic aeid-hematoxylin.B Small sections of liver from the 48 chickens of the first group plus 20 control and 20 experimental chickens of the second group were7 obtained at tkc tune of death, cut in small cubes, and fixed in Caulfield's and MillonlA fixatives.™ These tissues were dehydrated through a graded series of ethtarf and embedded in an Araldite-Epon mixture." Thin sections of approximately 0.5 p were cut on an ultramicrotome for light microscopy and stained by tkc toluidine blue method." TTltrathin sections were placed directly on 400-meak nncoated copper grids, stained with uranyl acetate, and examined withta RCA EMU-3G electron microscope. TABLE 1.—ALTERATIONS IN THE PERIPHERAL BLOOD OF CHICKENS CONSUMING TOXIC FAT No. of chicks Toxic fat in diet Hematocrit 30 30 Percent 00 30 Percent gm. per 100 mEg.ftiier mEg./liltr gm. per 100 ml. ml. 31 0 10 0 160 54 34 ISO 60 162 59 23 Hemoglobin Na K Serum protein Serum i Penaa t. S RESULTS When 3.0 per cent toxic fat was added to the diet of young chickens SO per cent died within 15 days. Approximately 24 hours prior to death, the chtckev became listless and moved only when agitated. Because of the marked abdoarinal distention the chickens assumed a ducklike gait when forced to wijfc. Moist rales were present in the lungs and considerable amounts of clear t» bloodrtinged fluid were observed in the oral cavity of the experimental chick. '- ens at the time of death. On the 15th day, blood studies were made on the surviving chickens. The* fed toxic fat showed a reduction in several of the blood components (Table 1) Hemoglobin levels decreased from 10.0 gm. per 100 mL to 6.0 gm. per 100 H" "and hematocrits were reduced from 31.0 per cent to 18.0 per cent There insm decrease in total serum protein from 3.4 gm. per 100 ml. to 2.3 gm. per 100 a& and a shift in the albumin-to-globulin ratio of the treated chickens. Similar electrophoretic patterns were obtained on the ascitic and pericardial fluids although the total protein of these fluids was only 1.3 gm. per 100 ml Blood electrolytes were not appreciably altered. OBOSS AND MICROSCOPIC OBSERVATIONS A large quantity of colorless, semiclotted fluid accumulated in the subcutaneous tissue and the pectoral^ thigh, and lower leg muscles were pale and edematous. The abdominal cavity of each experimental chicken contained approximately 40 ml. of ascihc fluid. The liver was somewhat mottled with rounded margins and a thick gelatinous material resembling coagulated plasma was fc ly t ta ed tlle S / , ^, ^ ^P8"16 ta 25 *** ceat of m* chickens that had received toxic fat The kidneys were pale and swollen. All of the organs witMnthe abdominal cavity appeared edematous. There was marked distention of the pericardial sac with a clear, slightly yellow fluid. Five to 10 ml. of pericardial a "L^f^J^J^*^.™™™ ^ce^le ^tion and The major microscopic alterations were observed in the liver. Its cansular Mrface was covered by a thick layer of homogeneous, eosinophilic material eontaining numerous fibrin strands. Although the general architecture of the Hrer parenchyma was maintained, numerous foci of necrosis were observed (Ffe. 1) These lesions varied in size, from only two or thra"cdto to a portion of the lobule. There was a fairly sharp line of demarcation bet U* viable and dead cells. Immediately adjacent to these necroticTf ocL the parenehymal Krkable ^ was «uite ™cuolated, but otherwise not THIN SECTIONS FOE LIGHT MICROSCOPY A . alterations w a s obtained from tissues embedded in the Araldite-Epon mixture than was possible using the conventional Paraffin embedded preparations. The parenehymal liver cefis of the control chickens stained uniformly with toluidine blue. Their nuclei were dark Moe and the cytoplasm was much lighter and granular. The degenerating parenehymal liver cells of the toxic fat chickens were of two types on the bafis^f **?„*?£ £ % tol™dine blue- One <*u type had a decided affinity for the dye „ that both the cytoplasm and nucleus were very dark (Fig 2) Manv of fese cells were shrunken, distorted and contained large morphologic featares «f the organelles were fairly well maintained in the cells obtained during the 6 ** ,» yL I?*1*?1*?6"*- However, the organelles were in very dose apposition as a result of what appeared to be shrinkage of the cells and loss * tte rtf81^ m£trix: ^though the external mitochondrial membranes were quite irregular, the shape and distribution of the cristae remained unaltered. The membranes of the endoplasmic reticulum were in close apposition *2ft!S?*t "bosomes a!ong the outer surface. Free ribosomes ^dispersed throughout the cytoplasm. In the dark cells, the Golgi complex lyso«nes and microbodies were usually sparse and only vaguely discernible. Only " ^SLTII * C cy*°Plasm contained glycogen granules. Numerous structures appeared to arise from the cytoplasmic membranous sysf? t^e T1? °f these ^^ ^^ had ^approximateirthrsfme density as the cytoplasm. The paired nuclear membranes appeared as wide, irregular dense band containing few discernible pores. Exceptforthe From the 7th through the 15th day of the experiment, the dark parenehymal l«r cells became increasingly electron-dense, the intercellular s'pacJs wider nd myelin-like bodies more abundant along the plasmalemmaa The eyt£ nic organelles became more difficult to visualize. Numerous oval to oblone vacuoles of variable size developed in the cytoplasm. Myelin-like bodies, If th.«T £g the 5*"m«1««««». ^ere present throughout the cytc^ . At this time it was difficult to visualize the existence of a distinct •Khar and cytoplasmic separation because of the electron density and the �306 absence of any distinct nuclear membrane. Eventually these cells developed into dark, shrunken, fairly homogeneous masses which exhibited little morphologic resemblance to liver parenchymal cells (Fig. 6). The second type of parenchymal cell in the degenerative areas was large and very electron-lucent. An isolated cell of this variety was seen most frequently within a group of the previously described dark cells. The plasmalemmae of these cells were in close apposition to the adjacent dark cells. The microvuU projecting into the space of Disse were shorter and less abundant than tho* of the dark cells (Fig. 7). Many of the organelles were markedly distorted Small segments of the endoplasmic reticulum were scattered throughout the cytoplasm. The majority of these membranes contained ribosomes along their outer borders. Numerous free ribosomes were also scattered throughout the cytoplasm. The mitochondria were large and disorted. Their external membranes were quite irregular and formed many bulblike projections from the surface of the mitochondria. Small vesicles were apparent between the external and internal, mitochondrial membranes. The enlarged cristae pracHcally filled the interior of the mitochondria (Fig. 6). Large, clear vacnoles and cytoplasmic degenerative bodies were abundant throughout the cytoplasm while glycogen granules were very sparse in the light cells. The Golgi complex consisted of collapsed vesicular sacs dispersed in small groups throughout the cytoplasm. The mitochondria were large and distorted. Their external chromatin material quite evenly dispersed throughout the nucleoplasm. During the terminal portion of the experiment, the large electron-lucent cetti developed openings in their plasmalemmae. Their cytoplasmic organelles were observed in the sinusoids, space of Disse, and in the Intercellular spaces. The endothelial cells which line the sinusoids showed similar changes to those previously described in the dark parenchymal cells. The endothelial changes were most frequently observed adjacent to a large focus of dark cells. The long, slender cytoplasmic processes of the endothelial cells became extremely dark with the cytoplasmic and nuclear structures becoming indistinguishable. Large fragments of these endothelial cells were frequently seen in the sinusoidal spaces. Similar changes were observed in the bile duct epithelium. It was not uncommon to see two or three extremely electron-dense ceU» which had lost all their internal structure adjacent to a number of seeming? normal epithelial cells (Fig. 8). DISCUSSION The gross lesions produced in chickens fed toxic fat have been well doeo. mented, yet the mechanism by which this fat produces ascites, hydropericardium, hydrothorax, and edema has not been clarified. Vascular lesions and cardiac decompensation have been mentioned as possible causes of the massive accumulation of extravascular fluid in chickens that have eaten toxic fat1 la this experiment the liver was investigated to determine its association with, the accumulation of extravascular fluid in chickens whose diets contained toxle fat There was a decided reduction in the total blood protein of the toxic fat chickens. The electrophoretic patterns of their sera showed the drop in proteb resulted from a decrease in the serum albumin. "When the liver tissues were examined microscopically the explanation for the decrease in serum albuoua became apparent. There were numerous areas of focal necrosis throughout the liver of the experimental chickens. There is no clear explanation of why some parenchymal liver cells shrink and become extremely electron-dense, while others appear to swell and become very electron-lucent Undoubtedly each of these changes is a degenerative proceo which eventually terminates in the death of the cell. One could postulate that these two cell types perform different functions and because of this functional difference, the reactions of these cells to this toxic material were manifested differently. Such changes in size and lucency, however, are not unique for toxic fat but resemble those observed in hepatic lesions produced by monocrotaline and x-irradiation.* Further studies will be necessary to clarify the differing responses of various parenchymal cells to this toxic substance. The accumulation of large quantities of extravascular fluid in the tissue* and body cavities of chickens that have consumed toxic fat can be attributed at least in part to an alteration in the permeability of the vascular bed. The high level of protein in the ascitic and pericardial fluid and the electrophorette 1 307 pattern which resembled that of blood serum would be substantiating data for the above statement. Extensive necrosis of the endothelial cells of the liver would also be related to altered capillary permeability. These observations would confirm a previous report1 regarding the toxic effect of this fat upon the vasculature of chickens. STTMMABT Three per cent toxic fat was added to the diet of 1-day-old chickens. Fifty per cent of the chickens died within 15 days. There was a reduction in the hemoglobin, hematocrit, serum protein, and a shift in the albnmin-to-globulin ratio of these chickens. The gross lesions in these chickens included hydroperi(ftidlum, dilation of the heart, pulmonary edema, ascites, and subcutaneous edema. When the tissues were examined microscopically the most distinct lesions were necrosis of the parenchymal cells, bile duct epithelium and endothelial cells of the liver. The liver lesions were evaluated electron microscopically and the morphologic changes reported. The reduction in total serum protein and shift In the albumin-to-globulin ratio of the chickens that consumed toxic fat were attributed to liver necrosis. The decrease in serum albumin may have predisposed to a portion of the eitravasated fluid. However, altered permeability of the vascular bed following toxic fat consumption appeared to be the major predisposing factor for the accumulation of large quantities of extravascular fluid. Acknowledgments. The authors are indebted to Dr. Karl T. 'Zilch, Emery Industries, Inc., Cincinnati, Ohio, for supplying the toxic fat EEFEBENCES > Allen, J. B. The role of toxic fat In the production of hydroperlcardinm and ascites. pb.D. dissertation, University of Wisconsin, Madison, June, 1961. Dissertation Abstr. tl: 545, 1961. 'Alien, J. B. The role of toxic fat In the production of hydropericardinm and ascites tn chickens. Amer. J. Vet. Res. tS: 1210, 1964. 'Allen, J. R-, and Carstens, L. A. Unpublished data. < Allen, 3. R., and Lallch, J. J. Response of chickens to prolonged feeding of crude toxic fat. Proc. Sac. Exp. Biol. lied. 109: 48, 1962. 'Armed Forces Institute of Pathology. Manual of Histologic and Special Staining Techniques. Ed. 2. New York, McGraw-Hill Book Co., Inc., 1960. •Brew, W. B., Dore, J. B., Benedict, J. H-, Potter, G. C., and Slpos, B. Characterization of a type of unidentified compound producing edema in chicks. /. Assoc. Off. Agric. 'Canlfield, J. B. Effects of varying the vehicle for OsO« in tissue fixation. J. Biophyii. Hochem. Cytol. S: 827, 1957. • Crosby, W. H., Munn, J. I., and Furth, P. W. Standardizing a method for clinical fcemoglobinometry. U.S. Armed force* lied. J. S: 693, 1954. •Danahoo, W. S., Edwards, H. M_ Schmittle, S. C., and Puller, H. S. Studies on toxic tot in the ration of laying hens and pullets. Poult. Sci. 38: 663, 1959. • Friedman, L., Firestone, D., Horwttz, W., Barnes, D., Anstead, M., and Schue G Studies of the chicken edema—diseases factor. J. Assoc. Off. Agric. Chem. it: 129, 1959. u Hald, P. M. The flame photometer for the measurement of sodium and potassium in ilolodcal material. J. Biol. Chem. 1ST: 490, 1947. , . a Barman, R. B., Davis, G. B., Ott, W. H., Brink, N. G., and Euehl, P. A. The Isolation and characterization of the chicken edema factor. J. Amer. Oil Chem. 8oc. 82; 2078, "Ki Kingsley, G. R. The direct binret method for the determination of serum piroteins as applied to photoelectric and visual colortmetry. J. Lab. Clin. Jled. tl: 840, 1942 «Macklln, L. J., Gordon, B. S., Meisky, K. A., and Maddy, K. H. Relationship of exudative degradation to toxidty in certain fats. Poult Set. 38: 579, 1959. "McGovern, J. J., Jones, A R., and Steinbelg, A. G. The hematocrit of capillary Hood. Sev> Eng. J. lied. 253: 308, 1955. • Millonlg, G. Further observations on a phosphate buffer for osmium solutions In fattfon. In Proceedings of the Fifth International Congress on Electron Microscopy, rUladelpMa, 19S1, Vol. 2, P-8, New York, Academic Press, Inc., 1962. a Uollenhaner, H. H. Plastic embedding mixtures for use In electron microscopy. J. ttaitt. Techn. 39: 112, 1964. "' • Potter, G. C., Brew, W. B., Patterson, P. L., and Sipos, E. Current status of the Uric fat principle causing the chick edema syndrome. /. Amer. Oil Chem. Soc. 36: 214, •ganger, V. L., Scott, L., Hamdy, A., Gale, C., and Ponnden, W. D. Alimentary toxemia In chickens. J. Amer. Vet. lied. Assoc. 133: 172, 1958. • Schmittle, S. C., Edwards, H. M., and Morris, D. A disorder of chickens probably "Simpson, C. P., Prltchard, W. R., and Harm's, R.' H. An endotheliosis in chickens ud turkeys caused by an unidentifiable dietary factor. J. Amer. Vet. Med. Assoc. 13iUO, 1959. "Tramp, B. P., Smuckler, E. A., and Bendltt, E. P. A method for staining epoxy actions for light microscopy. J. Vltrastruct. Res. S: 343, 1961. • Williams, F. G., PIckels, E. G., and Durrum, B. L. Improved hanging-strip paper rfertrophoresls technique. Science Ml: 829, 1955. • Wootton, J. C., and Alexander, J. C. Some chemical characteristics of the chicken disease factor. J. Assoc. Off. Agric. Chem. 4*.- 141, 1959. �309 308 CHICK EDEMA FACTOR : SOME TISSTJE DISTRIBUTION DATA AND TOXICOWXHC EFFECTS IN THE RAT AND CHICK.* (31029) The chick edema factor (CEF), responsible for a large number of deaths ia the broiler industry in the fall of 1957, was traced to the unsaponiOabli matter (unsap) of the fat used in the broiler rations. It since has been crj»tallized (1,2,3) and its structure proposed as that of a hexachlorohexahydiophenathrene (2). It is known that in the toxic fat, a mixture of related compounds can be found, some toxic and some relatively nontoxic. To lay the groundwork for a study of the specific physiological effects <g pure CEF compounds, a few short studies have been completed to learn the distribution of the toxic material in the body and which organs were primarily affected. Experimental. Adult rats and day-old White-Rock chicks were used. Became the pure material was not available in sufficient quantity, we used, from the toxic fat, the unsap which represented 38% of the original toxic fat and was estimated to contain at least 10 ppm CEF. The unsap was force fed because the animals' food intake was drastically curtailed when it was mixed in the diet All animals were offered water and commercial feed ad libitum. Table I shows the experimental plans and dosage levels employed for all S Trials. Feed consumption and fecal and urinary excretion were measured for the rats. Body weights were recorded in all experiments. All animals, upon sacrifice, were examined grossly for pathology and selected organs wen weighed and frozen. Hydropericardial fluid (HPF) volume was measured n the chicks. In addition to the examination for gross effects, the presence of CEF material in various tissue was determined. Adrenals, kidneys, and livers were assayed in the rats; livers only have been analyzed in the chicks. To obtain « picture of the amount of material being absorbed and perhaps excreted, analyses of the feces and urine of rats and of the combined chick excreta wen made. TABLE 1—EXPERIMENTAL PLAN OF TRIALS 1% SE-30 on ANAKEOM ABS (Analabs). The operating conditions were: temperatures, column oven 180°C, EC detector 200°C, flash heater 240°C; gas flows, helium carrier gas, 60 cc/min; argon, 10% methane purge gas 180 cc/min. The limit of detection by this method is approximately 0.2 ppb, assuming a jam-tie size of f> sr and a sensitivity of the EC detector of 5 X 10"11 g. The so-called CEF components can be seen as a gas chromatographic pattern 0{ peaks shown in Fig. 1. This pattern is that of a highly concentrated material isolated from the toxic unsap. It is similar to a material isolated by TartHff et al and kindly supplied to us by Firestone (3). We have numbered the peaks 1 through 8 as shown. Peak 4a was not seen in the Firestone preparation.! Results and discussion. Table II shows the effects upon body weights, feed conversion, and feed consumption for the rats. The results for the vital organ Heights and HPF volume are shown in Table III. There was no apparent poss pathology in either species with the exception of HPF and some ascites tsd subcutaneous edema in the chicks. While only the chicks develop hydropericardium, apparently both are equally Kiisitive to liver weight increase, as shown in Table III. In the case where etch species was maintained on CEF for 6 days, (Trials II and III), the rats TABLE II.—GROSS EFFECTS IN RATS Trill Trial 1 H days. 11 6 days III 6 days Species Group 2 High Rat do Chick . 4 4 4 6 6 6 High Low Control High Low Control" Is ^ 6.6 y Feed intake depression i (percent) —38 29 lcraitrol.- Dry matter diz.coeff. (percent) 76 78 72 75 77 i Depression below control animals. Dosage (per kj) Number of animals Group _, (High LMdWLow (High *•' Percent bodywt change TABLE III.-ORGAN WEIGHTS EXPRESSED AS PERCENT OF BODY WEIGHTS Unsap CEFtefi(ml/day) mated 04Mq) 2.0 10 2.0 1.0 0 5.4 1.1 » M i M "5 > « M 1 ' Control animals were not inhibated. Trial « tjK n •• • Chicks "• ^^ Group Heart Liver Spleen .34 .32 .31 <4 08 »4.08 3.36 20 .20 .19 .86 .70 .78 13.85 3.15 2.73 61 54 .65 {High.. (Control fHi£h -{Low (Control. .{LOW Kidney 90 .82 .76 HPF(ml> J 65 i 12 .04 Adrenals * 21 .18 .14 iMF_hydropericardial fluid (ml). tSoiifiant statistically at I percent probability level. >A?5 percent probability level (Hogben L testXS). The assay method for CEF is being reported in detail elsewhere.J In short, the sample is homogenized with water and saponified with alcoholic KOH. Th» unsaponifiable portion is extracted with petroleum ether and chromatographcdj first on an alumina column. The eluate obtained with 25% ethyl ether ia petroleum ether, following prior elution with petroleum ether and 5% ethyl ether in petroleum ether, is concentrated and chromatographed on 500 p thi» layer silica gel plates with 3% ethyl ether in petroleum ether. The silica gel it the area of Rr 0.80-1.00 is removed and eluted with ethyl ether, the solve* removed, and the residue is redissolved in isooctane for gas chromatography. We used an F & M Model 400 gas chromatograph with an electron captmi detector and a TJ-tube column, 3 ft X 6 mm (o.d.) X 4 mm (id.), packed wltfc • Supported by TJSPHS Grant EF 00305. Contribution 858 from Dept. of and Pood Science. Massachusetts Inst. of Technology. t T. C. Campbell and L. Friedman in press, J. AOAC, "Chemical Assay and of Chick Edema Factor." (9 jg/kg/day) showed an increase over controls of 21% while the chicks (10 .^/kg/day) showed an increase of 15%. (The increase in liver weight in the dicks was not due to moisture or fat.) Also, in Table III, neither heart nor spleen weights in either species are significantly affected. In rats, there appears to be a slight increase in kidney nights, though not statistically significant, and a highly significant increase •f 50% in adrenal weights for animals on the high level. Whether this adrenal ceight increase is simply a non-specific stress effect from intubation is not taown, although it would seem that this cannot be entirely responsible, since (be high level group showed an increase of nearly twice that of the low level troop. Some other observations which have been made on previously studied birds to this laboratory are of interest Hematocrit values are depressed. Of a total 4 43 birds, we have observed that, with an average of 0.08 ml HPF in control I The chromatogram for the Firestone preparation is presented elsewhere. �311 310 FIGTJBE 1.—Toxic CEF components found in unsap used in this study. FIGURE 2.—Eat liver extract showing 2 CEF peaks. FIGURE 3.—Hat feces extract showing CEF peaks, with altered Nos. 4a and 7. FIGURE 4.—Chick liver extract showing 2 CEF peaks. (Large peak just before No. 4 found to be contaminant leached from liner of sample vial.) birds (considered normal), there was a packed cell volume of 33.0%, while for diseased birds having 0.73 ml HPF, the packed cells volume was 27.9%. Control animals not receiving CEF did not show these peaks. These chromatograms show that only peaks 4 (or 4a?) and 7 were present in the liver. In the fecal extracts all peaks were found, with the exception that instead of peaks 4a and 7 showing their original retention times, these were slightly increased in each case and have been designated 4a' and 7'. These retention times increases in the fecal components were measured by noting their retention times in relation to their neighbors, as shown in Table IV. Whether the 2 components of the liver are the products of components 4 and 7 or the original unaltered substances cannot be accurately determined from these chromatograms, since the rest of the CEF chromatographic pattern Is missing. It may be concluded, however, that there is a selective absorption of peaks 4 (or 4a) and 7, with metabolism by the liver and excretion into the intestine. It appears, on the basis of the tissues analyzed, that the liver is the target organ. Furthermore, the pattern of CEF chromatographic peaks presented here, Nos. 4 and 7 are the key components. We have succeeded in partially separating the CEF components such that 4a and 7 crystallized together, indicating a certain chemical as well as physiological similarity.^ There is no way of differentiating 4 from 4a with regard to its absorption and metabolism on the basis of the available evidence. Summary. TJnsaponifiable matter isolated from a toxic fat and containing an estimated 10 ppm chick edema factor (CEF) was force-fed daily to adult rats at levels of 2.0 cc and 1.0 cc/kg body weight/day in 2 studies of 14 and 6 days, respectively. Feed consumption, body weight, and digestibility were depressed. Heart and spleen weights were unaffected, kidney weights seemed to be slightly increased, and adrenal and liver weights were significantly increased. In the chick, typical hydropericardium, ascites, and subcutaneous edema were observed. There were no significant changes in heart or spleen freights. Liver weights were significantly increased. The rat was as sensitive as the chick to CEF according to increase in liver weight Of the 8 or 9 CEF components shown to be in this unsap, 2 (Nos. 4 and 7) were found to be absorbed and located in the liver, while the other components were not TABLE IV.—CHANGES IN CHROMATOGRAM RETENTION TIMES OF CEF PEAK NOS. 4a AND 7 Relative retention times No. of runs Material analyzed Distone Standard _ RT:. R,K. l!l4±.03 4a not found 1.22±.01 1.27±.M 31 2 13 > Includes feces of 11 individual rats and the composite feces of 2 groups of 6 chicks each. < The deviation includes the total range of values. detected. In place of Nos. 4 and 7, there were 2 new peaks in the feces with jlightly increased retention times. This suggests that the 2 active CEF components are metabolized in the liver and excreted into the intestine via the bile, both in the chick and the rat No CEF-like material was found in kidneys, adrenals, or urine. FIGURE 5—Chick feces extract showing CEF peaks, with altered Nos. 4a and 7. Also, we observed that whereas control birds will show disappearance of « given amount of an I.V. injected dose of T-1824 dye (Evans Blue) from their vascular system of approximately 1%/min, poisoned birds will show a disap. pearance of approximately 2%/min, supporting the observation of Allen (4), that the permeability of the vascular wall appears to be increased. Distribution of the CEF in animal 'body.—Of the rat tissues and sample* examined, CEF was detected only in the liver and feces. Fig. 2 through 5 ahcnr typical chromatograms of purified extracts of feces and liver of both specie*. STUDIES ON TEE METABOLISM OF CHICK EDEMA FACTOR : DISTRIBUTION IN CHICK TISSUES D. Firestone, G-. R. Higginbotham, D. F. Flick and J. Bess The distribution of chick edema factor In chick tissues following consumption of rations containing toxic fat has been of considerable interest Chick i Barman, R. E., Davis, 6. E., Ott, W. H., Brink. N. G., Kuehl, F. A., J. Am. Oil Cbcm Soc.. 1960. Y82, 2078. • Wootton, J. C.. Courchene. W. L., J. Agric. Food Chem., 1964, v!2, 94. • Tartzoff, A.. Firestone, D., Banes, D., Horwitz, W., Friedman, L., Nesheim, S., J. im Oil Chem. Soc., 19«1. T38, BO. «Allen, J. K., Ph. D. Thesis, 1961, TJnlv. of Wisconsin. •Hogben, C. A. M.. J. Lab. Clln. Med., 1964, v64. 815. • Received January 13,1966. P.SJ3.B.M., 1966, v!21. �313 312 factor was detected in chicken tissues by electron capture gas chromaa^d^n eSatfon of the concentration in various tissues was made ^ tissues and parts from ttree « submitted by the Division of Nutrition, were examined for factor by a recently developed method.1 One group had been fed a 3% ofTreference toxic fat Another group had been fed a ration t unWonifiables equivalent to 3% of the toxic reference fat The ?as?£wp received a ration free of toxic fat The weights of tissue samS SSgld foomT* to 93.4 grams. The samples were received in glass stoppered Brlenmeyer flasks, in ethanolDETERMINATION (1) Extraction of unsaponifiable matter from animal tissue (Modification of e solution of the homogenized tissues to a bottom flask, add ethyl alcohol to give a final volume of 4 but not less than 50 ml. Add 2 ml KOH solution (3 + 2) per y boiling with occasional swirling on a steam bath for one hoar aircradenser. Transfer alcohol soap solution whUe stall warm to using water (equivalent to twice the volume of ethyl alcohol) Rim* nonmc«Si flask with the same volume of ethyl ether and transfer to separttor StateVigorously, let layers separate and clarify, breaking any emulsion bv Iddtalup to 1/20 volvimes of alcohol and swirling gently. Drain lower layer and MU?etter layer through top into a second separator containing water (2 mLtetiSue) but not less than 20 ml. Make two more extractions of soap sol* Son w^lthy^ ether (8 ml/g tissue). Rinse pouring edge with ethyl ether ami R o c m b n r « r gently with the H,O (violent shaking at thi, stage may ^.^troublesome emulsions). Let layers separate and drain aauwuTlayeT Wash with two additional portions of ttO (2 ml/g tissue), drtto? vigorously Then wash ether solution two times with alternate portions f2 m!% tissue) of K,CO, and H,O. If emulsion forms during washing, drata as much of aqueous layer as possible, leaving emulsion in separator witt ether layer and proceed with next washing. Wash final solution with Hfl until washings are neutral to phenothalein. Transfer ether solution to erlenmeyer, rinsing separator and ite pouring edge witt S adding rinsings to main solution. Dry ether solution by adding rnhydr™%a,S07(l g/g tissue) and swirling vigorously ca 1 nun. Let solotion rtand 10 min Decant ether solution through glass funnel containing Dtedget of pre-rinsed cotton in neck and holding 25 gr anhydrous Na,SO, into another erlenmeyer containing boiling chips. Wash first erlenmeyer with ether "SSSMTtf SSH? steam bath and transfer to 100 ml (preweighed) extraction flask. Evaporate residual solvent on steam bath under K. Sryflask to constant weight and obtain weight of unsaponifiable matter. Proceed with fractionation of unsaponifiable matter on alumina and cleanup of alumina fraction 3 as directed in method (1), analyzing the residue by el«tron capture gas chromatography as directed. RESULTS AND DISCUSSION All residues except for liver samples were initially taken up in 100 ul of ISO-octane and 5 ul injected. The liver samples were taken up in 250 ul of i*. octone and 1 ul injected. Examination of the chromatograms indicated th» Presence of a small contaminant (Ra 11.1) in the reagents used in the cleannp °f S^SfloSL^^oxic" from each group of chieksfed toxic fat* material. The liver sample in each group was the most "toxic , containing rWhly 83% of the total "toxicity" as indicated in Table I. The terms "toxfeor "toxicity" as used here to describe results of gas chromatographic analyse, refers to the amount of material in an individual tissue which produces chtracteristic peaks at Ra 12 and 22 resulting from feeding the toxic fat Of the S adrenals, bone, brain, heart, intestine, kidney liver.skeletal nraafe, skin, and testes that were . examined, only the bone, heart, intestine, kidney. liver, and skin showed peaks characteristic of the presence of chick edema factor. No characteristic peaks were found in the adrenals, brain, skeletal muscle, and testes; perhaps the concentration was too low to be detected in these tissues. The toxic fat as well as the toxic unsaponifiables used for this study exhibited 4 characteristic GLC peaks of Ra = 10.6, 12.4, 18.6, and 21.6 (See Table 2). The intestine; chromatograms exhibited greatly diminished 10.6, 18.6 and 21.6 peaks, and a peak appeared at 12.0 in place of the 12.4 peak of the toxic fat and toxic unsaponifiablea The 10.6 and 18.6 peaks were not evident in chromatograms from the other tissues exhibiting characteristic peaks, and the 12.0 peak was the major characteristic peak. In addition, the bone, heart, kidney and skin extracts from the chicks fed toxic fat (Group A) exhibited a 12.4 peak which occurred as a shoulder on the major 12.0 peak. These results suggest a selective absorption of the chlorinated components of the toxic fat; this is most clearly indicated by the diminished 10.6 peak in the intestine extract which is completely absent in the other tissue extracts. The appearance of peaks at 12.0 or 12.0 and 12.4 in place of the 12.4 peaks of the toxic fat can also be explained by selective absorption and deposition of individual components if we recognize that the GLC peaks observed each represent more than a single component Campbell and Friedman (2) have also observed selective absorption of chick edema factor components in rats as well as in chicks. However, these authors only detected chick edema factor in the liver and feces. REFERENCES »Hlgglnbotham, G. R., Ress, 3., and Firestone, D., JAOAC, in press. » Campbell, T. C., and Friedman, L., JAOAC, 49 82^-828 (1966); Proc. Soc. Exp. BIol. Med. lil, 1283-1287 (1966). TABLE 1.-RELATIVE % Of CHICK EDEMA FACTOR IN POSITIVE TISSUES" Group A (3 percent toxic fat) Tissue 7.1 DAQI u-irt Intestine Group B (unsap.~ 3 percent toxic fat) - 0.6 2.4 2.4 - KMflRV IMMT ev:« 82.8 4.7 _ . 4.8 1.2 2.4 18 83.6 7.2 i The relative percent of CEF in each positive tissue was estimated by adjusting the volume of each sample so that a i a injection of each would yield a chromatogram exhibiting the major peak (Ra 12) with a height of ca 6 cm. The following formula was used to calculate percentages: %CEF< volume of sample )xlOO \Total volume of all samples in group, TABU ((.-RETENTION TIMES <RA)« OF CHARACTERISTIC OF CHICK EDEMA FACTOR COMPONENTS IN TOXIC FAT ADDED TO DIET AND ISOLATED TISSUES IBitber Cdman Model 5360 gas chromatography; 7 foot 1/4 id glass column packed with 2M% SE52 silkone gum rubber on 60-80 mesh Gas Chrom Q; 3 X10-8 amperes full scale; detector voltage, 30; injector temp., 240° C; column temp. 200° C; detector temp., 210° C) Sample Group A (Fed 3% toxic fat) Tuicfat ................................. Bauponifiables (from toxic fat) Unnals ................................. tnin tat. uterine y .................................... ............................... ................................ ................................... 10.6, 12.4,18.6,21.6. ................................. .. ............ None. 12.0,12.4,21.6 ...... None.. ............ 12.0,12.4,21.6 ...... 10.6,12.0.18.6,21.6, 12.0, 12.4, 21.6 ...... . Skeletal Muscle ........................... None 5S .............................. 12.0,12.4.21.6 ^'" ............................... None.... Bi- itetention time relative to aldrin. Group, 9 (fed unsap. st (3% 10.6,12.4,18.6,21.6 None. 12.0,21.6 None 12.0 10.6.12.0,21.6 12.0,21.6 12.0,21.6 None 12.0,21.6 None. �L , 314 315 LIGHT AND ELECTRON MICROSCOPIC OBSERVATIONS IN Macaco, mulatto MONKEYS FED Toxic FAT J. R. Allen, D.V.M., Ph.D., and L. A. Carstens, B.S. 8T7MMABY Thirty-six Macaco, mulatto, monkeys were given a diet that contained 0.125 to 10.0% of a fat capable of producing hydropericardium, ascites, and death in chickens. There was an inverse relationship between the concentration of toxic fat in the diet and the survival time of the monkeys. The monkeys given the greatest level of toxic fat had the mean survival time of 91 days, and the monkeys given the lowest level and the mean survival time of 445 days. During the last 30 days of life, the monkeys developed generalized subcutaneous edema, ascites, hydrothorax, and hydropericardium. There were decreases in erythrocytes, leukocytes, total serum protein values, and altered albumin :globulin ratios. There was also cardiac dilatation and myocardial hypertrophy and edema. Experimental monkeys had reduced hematopoiesis and spermatogenesis, degeneration of the blood vessels, focal necrosis of the liver, and gastric ulcers. It was proposed that toxic fat exerted its injurious effects upon the parenchymal cells of the liver, endothelium, and myocardium with subsequent development of generalized anasarca. Fats from plant and animal sources have been used to increase the caloric level of diets for animals. As a result of increased demands by feed manufacturers for low-cost fats, almost every available source of these products has been utilized. Certain fats were found to be extremely toxic to poultry, and hundred's of thousands of chickens died or were killed after they were fed diets containing these fats. Besults of experiments indicated that young chickens developed hydrothorax, pulmonary edema, ascites, and1 subcutaneous edema in 1 or 2 weeks when there was toxic fat in their diet. -1'""" The accumulation of large quantities of extravascular fluid in chickens seemed to result from altered permeability of the vascular bed, cardiac decompensation, and liver neerosisi1-3 Chemical studies on toxic fat indicated that the toxic fraction was located in the unsaponifiable portion.'-30 By repeated passages through alumina and silica gel columns, crystalline preparations of the toxic fraction were prepared'; however, the chemical composition of the compound was not determined. Before the cause of this intoxication of poultry was established, many chickens that had been fed toxic fat were processed for human consumption. Since that time, the clinical, histologic, and electron microscopic 151 changes that occurred in the intoxicated chickens have been enumerated.1'*- " Data are not available, however, concerning the effects of toxic fat on primates. Since the chemical composition of this fat is unknown, the possibility exists that it may once again adulterate various edible fats. This is a report on experiments undertaken to determine the effect of toxic fat on lower primates. The result* of the experiments may be helpful in postulating the effects that toxic fat might have in man. MATERIALS AND METHODS Since the chemical nature of toxic fat was unknown and chemical procedures were not available to determine the toxicity, a biological assay was performed on the fat used in the experiments. When the diet of 1-day-old chickens contained 3.0% toxic fat, 50.0% died within 15 days. In the initial experiment, 16 Macaca mulatto, monkeys (av. weight, 4.2 kg.) were allotted to 4 groups and fed diets containing 0 (control), 1.0, 5.0, and 10.0%, respectively, of toxic fat In the 2nd experiment, 20 if. mulatto monkeys (av. weight, 6.0 kg.) were allotted to 4 groups and fed diets containing 0 (control), 0.125, 0.25, and 0.5%, respectively, of toxic fat. The toxic fat wu combined with corn oil to obtain similar fat intake levels in all groups of monReceived for publication June 23, 1966. From the Department of Pathology and the Regional Primate Research Center, Dtf. versity of Wisconsin, Madison. WIs. 53706. This research was supported in part by grants HE-08681 and FR-0167 from tb National Institutes of Health. The authors thank Miss Karen Welke and Mrs. Adrienne Cappas for technical assistance. C0unt . "ro' el trol Prothrom- bin time" serum bilirubinT value » urea nitrogen value" and body w e h t the general appearance and^mt <* ytes,' blood day on Possible, the procurement of fresh tissue for UgKd1 electeon^&ted V1 to ensura from heart, lung, liver, spleen, ^SIJ^S^SS^,7?*08 ^^^ skeletal muscle, testis, gastrointestinal t™or ?? ,L i s*ernal b°ne marrow, cm. of intestinal tmrt)7«kSaten2 iSnd ™™ **»»** «* every 10 bellum, pituitary gland, thyroid eland n^™ th^w , ' ¥toey> «™»«<nn, cere- saw ssast t •saw a? =-£s-sS£r r^ "•"« placed on 400-mesh uncoated cooner examined with an electron imcroseope.* uranyl acetate . RESULTS 5.0 and fed the diet with 1.0% toxic fat uajo* -- -••*-.*- WA JAM? J*£ ^th^^L^T^an^S/UTt,1^ "-**• The had mean SU times of 202, 274, and 445 day^ respecttfely' ^™1 Jss"iSSS?K fflrs^-ssanr^ta *• ™- and occurred during the terminal 30 £™£S±*£ t^8 Werc similar survived for less than 4 months or lonSrtt^r^i^ ^eth? tte monkeT from the experimental monkeys w^? beSn&coU^ely1116^0'6' the data HEMATOMOIC EVALUATIONS aiy^S^J^^^^^^^^ approximately 2.0 albumin in the monkeys fedtoxtefat OB n* 1,^7 f ^ Percentage of serum was 35%, whereas that rf K^taTnSSi? JSSd'S? ^ *° death gradual decrease in the cellular elementrnf t^L wlfl f ^, 61%- There was a Pscked cell volumes were reduced from^2^ r^i«2?^ ?ring tte ^^^ents. Bfe. Thesefindingswere suTsSfteTbf tte to^fr^^l,4116 nast M days of areraged 2.5 miUion/cmm. of Wood immUiflr^^0* ^^ ceU counts wnich tenoglobin values fo™ed a stofl™ c^f7^! '!,,*116 monkey «««• The Gm./100 ml. of blood being obtained in ttTfast^^t "^ Valne of 6'° ^serrations were recordedI for white bloodt £n ^ y ^°f ^^ ComParable 1 .Mtecells/cmm. of blood on «uWj±2^E^ ' i S S : M1CrOSCW' ^ °««- o' ^rica, Camden. „. " i i S �316 tal monkeys. Prothrombin times, serum bilirubin values, serum electrolytes, blood urea nitrogen values, and cholesterol levels of serum were not changed appreciably during the experiment CLINICAL OBSERVATIONS The major changes in the monkeys were the development of generalized alopecia and subcutaneous edema 1 to 2 months before death. Edema, first noticed around the lips and eyelids, progressed to the remainder of the face and eventually involved the subcutaneous tissue of the trunk and extremities. Especially obvious was the marked edema of the scrotum and sheath which developed during the last few weeks of life and, In some monkeys, partially obstructed the flow of urine. During the last month of life there was decreased food consumption and the subsequent loss of body weight was frequently as much as 1 kg. Diarrhea developed in 75% of the experimental monkeys during the last few days of life. Results of bacteriologic cultural examinations of feces were negative for pathogenic enteric organisms. GBOSS AND MICROSCOPIC FINDINGS The findings at necropsy substantiated the presence of extensive subcutaneous edema in over 75% of the monkeys fed toxic fat Heart.—Dilatation of the heart was especially obvious on the right aide. This was further clarified when the circumference of the valves was determined. The mean tricuspid and mitral valve circumference of the experimental monkey hearts was 55 and 45 mm., respectively; in contrast, the tricuspid and mitral valves of the control monkey hearts averaged 40 to 36 mm., respectively. Hypertrophy of the cardiac muscle was also apparent in the experimental monkeys. The hearts of the experimental monkeys were 0.55% of the body weight, whereas hearts of the control monkeys were 0.30%. Microscopically, the muscle fibers were distinctly separated by fluid. Individual muscle ceU» were hypertrophic, and their nuclei were enlarged, distorted, and hyperchromle (Fig. 1). There were no distinct valvular lesions in hearts of the experimental monkeys. Lungs.—Lungs of experimental monkeys were not altered appreciably. Isolated areas of atelectasis, congestion, edema, and fibrosis were observed. The proliferation of fibrous connective tissue was associated with the presence of lung mites (Pneumonysis simicoli). Liver.—Livers of experimental monkeys were small, firm, and moderately yellow. On microscopic examination, moderate distortion of the architecture was found. Many parenchyma! cells were enlarged, multinucleated, and had only moderate affinity for stain, whereas other cells were small and markedly hyperchromic. There was also focal necrosis of the parenchyma! cells in the centrilobular zone (Fig. 2). Many parenchyma! cells contained vacuoleg it their cytoplasm which stained positively for neutral fat when frozen sectiow were prepared. Small areas of fibrous connective tissue occurred in the peri, portal area; however, they did not alter the architecture appreciably. Spleen.—Spleens of experimental monkeys averaged only 0,074% of the body weight, and those of the control monkeys were 0.13%. Microscopically, the &*. minal centers were surrounded by only a narrow zone of lymphocytes, tht blood sinuses were practically devoid of cells, and the trabeculae were era*, cially prominent Mesenteric Lymph Nodes.—Iiymph nodes were light tan and edematow Microscopically, the germinal centers were surrounded by a narrow bani tf lymphocytes. The medullary cords were indistinct, and the sinuses were find with proteinaceous fluid. Sternal Bone Marrow.—Grossly, the bone marrow resembled coagulahri plasma. Microscopically, only a small number of hematopoiettc cells were am in the marrow, and those were approximately equally divided between tfe T 317 myeloid and erythroid series (Fig. 3). Most of the bone marrow was composed of fatty tissue and proteinaceous fluid. The blood vessels and sinuses contained only a limited number of cells. Skeletal Musculature.—The skeletal muscle was pale and edematous. Microscopically, the muscle bundles and fibers were widely separated by fluid, but otherwise seemed normal. Testes.—Grossly, the testis seemed normal; however, when examined microscopically, active spermatogenesis was not found. The seminiferous tubules had abundant spennatogonia and Sertoli cells, but only a limited number of primary spermatocytes. There were no spermatids or mature spermatozoa. Interstitial tissue was moderately edematous, but the Leydig cells did not seem affected. Gastrointestinal Tract.—In 18 of the 27 experimental monkeys, marked hypertrophy ol the gastric mucosa occurred in the fundic and pyloric regions. In the same areas, small gastric ulcers penetrated the mucosal layer. (Fig. 4). In 6 monkeys, inflammatory changes in the intestinal tract were seen. Results of bacteriologic evaluations of these lesions indicated no pathogenic organism that could have been associated with the gastrointestinal disturbance. Microscopically, the hyperplastic gastric mucosa was seen to form many large, interdigitating folds. Adjacent to these proliferative areas, the mucosal lining was eroded, and the underlying tissue was necrotic. Large numbers of polymorphonuclear leukocytes were in the necrotic tissue and underlying musculature. Blood vessels of the edematous submucosal and muscular layers of the stomach adjacent to the ulcerated areas were free of obstruction. Enteritis in 6 monkeys had caused moderate denudation of the intestinal mueosa and considerable hemorrhage into the lumen. The mucosal lining near tie base of the crypts was intact, and the underlying musculature was normal. Skin.—There was marked edema of the dermal layer of the skin, causing disarray of the collagen fibers. The epidermal layer was comparable in the control and experimental monkeys. There was an absence of any detectable change in the hair follicles of the monkeys given toxic fat The adrenal gland, pancreas, kidneys, cerebrum, cerebellum, pituitary gland, thyroid gland, and urinary bladder of the experimental and control monkeys were comparable grossly and microscopically. ELECTRON MICROSCOPIC CHANGES The extent of electron microscopic change in the liver correlated well with the level of toxic fat in the diet and the duration of toxic fat consumption. An early change in the parenchymal cells was the disruption of the orderly arrangement of the granular endoplasmic reticulum. The cisternal spaces were dilated, and the loss of ribosomes from the outer surfaces of the cisternae resulted in an apparent increase in the smooth endoplasmic reticulum. As a result of mitochondria! swelling, the cristae seemed shorter and less abundant than those of the control monkey hepatic cells (Fig. 5 and 6). Cytosomes of nriable content and size were moderately prevalent in the cytoplasm. Small reticles and flattened lamellae comprised the relatively small Golgi complex. Fit racuoles were abundant throughout the cytoplasm (Fig. 6). The nuclei contained distinct nucleoli and abundant chromatin dispersed throughout the jndeoplasm. Results of hepatic biopsies made a few weeks before death, and experimental tissues obtained at the time of death, indicated distinct alterations. Many ivenchymal cells were shrunken and electron dense (Fig. 7). These dark cells jere seen in various stages of degeneration. In some cells, the cytoplasmic eganelles were still visible despite the extremely dense matrix. Vacuoles were 4bpersed between the cellular organelles. Myelin bodies were abundant in the cytoplasm, along the plasmalemma, and in the intercellular spaces. The nuclei tire also electron dense, and the nuclear envelope was only vaguely discerniIfe. other dark cells had lost all resemblance to normal parenchymal cells, �319 318 being devoid of discernible cytoplasmic organdies and nuclei- Microvilli were extremely prominent and abundant along the plasmalemma adjacent to Disse's space. The lighter cells (Fig. 8) seemed to follow an entirely different coarse in their degenerative process. The cytoplasmic organelles were quite distinct in the abundant matrix. There was marked disruption of the endoplasmic reticulum, with only short fragments being scattered throughout the cytoplasm. Abundant free ribosomes were quite evenly dispersed between the organelles. The external contours of the mitochondria were frequently irregular, and the matrix was moderately electron dense. Occasionally, bulblike projections were formed by the external mitochondria! membrane. The plasmalemmae were irregular, and the microvilli were short and sparse. Occasionally, there were myelin bodies along the plasmalemmal surface. In some instances, the light cell plasmalemmae had ruptured, and organelles were dispersed throughout the extracellular space. Bile duct epithelium was affected markedly in monkeys fed diets containing toxic fat. Many of the epithelial cells were so electron dense that the cytoplasmio organelles were difficult to visualize, and their nuclear membranes were irregular and extremely dense. The interlocking plicae of adjacent cells were widely separated. As a result of the shrunken condition of the epithelial cells, microvilli on the luminal surface of the plasmalemmae seemed thin and elongated. Endothelial cells in some areas of the liver had a distinct resemblance to the dark parenchymal and bile duct epithelial cells. The dark endothelial cells were shrunken, and their internal structures were distorted. There were widened fenestrations between the endothelial cella Changes in the cytoplasmic organelles were comparable with those observed in the parenchymal cellk Occasionally, large cytoplasmic sequestra from the endothelial cells were observed in the lumen of the vessels. Heart.—The main differences between hearts of control and experimental monkeys were the dilatation of the intercellular spaces and the wide dispersal of the myofibrils in the latter. Between the widely separated groups of myoflbrils (Fig. 9), there were mitochondria, occasionally a segment of sarcoplasmlc reticulum, and abundant matrix. Usually, only the 2 lines could be readily visualized. Cytosomes were much more abundant in the muscle cells of experimental monkeys and were usually found near the nucleus (Fig. 10). In more than 75% of the hearts of the experimental monkeys there was a distinct swelling of the mitochondria. The crlstae were widely separated, and the mitochondrial matrix was abundant (Fig. 11). Large myelin figures were observed within and surrounding many mitochondria. Myoeardial nuclei, components of the transverse tubular system, as well as elements of the sarcoplasmic reticolum, seemed unaffected. There was a noticeable separation of intercalated disks in many of the experimental monkeys (Fig. 12). The various hands formed by the myoflbrils were often masked due to the abundance of edematons fluid in the tissue. Many of the endothelial cells appeared electron dense and shrunken, with distortion of the nuclei and cytoplasmic organelles (Fig. 13). In most instances, the organelles were in close apposition as the result of the cell shrinkage. There were cytoplasmic myelin bodies in many of the cells. Occasionally, intercellular stromal cells had changes similar to those in the endo. thelial cells. DISCUSSION Toxic fat consumption bad a decided effect upon hematopoiesis and sperm*, togenesis. Myeloid, erythroid, and lymphoid elements of the peripheral blood were markedly reduced. Germinal centers in the lymph nodes and spleen and the islands of hematopoietic cells in the marrow were extremely sparse. Iuhfl>ited spermatogenesis was also observed in the monkeys fed toxic fat. s sas/aa The relationship of toxic fat consumption by monkeys and the development of moderate to extensive alopecia has not been establ& PeriodlS Sr normal conditions there is a loss of hair by most monk^s. Howevtr oZ to Sntoxtefat^ ^ alOPCCla beC°me as ertenatre as <*** ot tne monkey^ An interesting and not well-understood lesion was the development of olcers in more than 66% of the monkeys fed toxic fet T^e appearance of the ulcers was similar to that observed in man an f etiologic A ±rf of tox 0? upon the gastric &ctors was underlying direct effect fT^? fat ° ** mucosa and eZV lc . When the blood protein values were tested, a decreased total and a reversal in the albumin :globnlin ratio were foWd ( T w l Th decrease in serum protein values, particularly albumin, could not be attributed to altered renal function because there was no increase in blood iW nit™ or albumin in the urine. The most logical explanatton for tte Z™ ?? changes recorded in chickens that o v �320 321 REFERENCES 1 Allen, J. R.: The Role of Toxic Fat in the Production of Hydropericardlum aai Asdtes in Chickens. Am. J. Vet. Kes., 25, (July, 1964) : 1210-1219. 9 Allen, J. R., and Carstens, L. A.: Electron Microscopic Observations in the Liver of Chickens Fed Toxic Fat. Lab. Invest, 15, (1966) : 970-979. * Allen, J. R., and Lallch, J. J.: Response of Chickens to Prolonged Feeding of Crnfc Toxic Fat. Proc. Soc. Exptl. BIoI. & Med., 109, (1962) : 48-51. * Bowman, R. E., and Wolf, R. C.: A Rapid and Specific TTltramicromethod for Total Serum Cholesterol. Clin. Chem., 8, (1962) : 302-309. 'Canlfield, J. B.: Effects of Varying the Vehicle for OsOi in Tissue Fixation. J Biophys. Biochem. Cytol., 3, (1957) : 827-830. • Friedman, L., Firestone, D., Horwitz, W., Barnes, D., Anstead, M., and Shne. 0 • Studies of the Chick Edema Factor. J. Asgoc. Off. Agric. Chem., 42 (1959) : 129-140 " I Gornall, A. 6., Bardawill, C. J., and David, M. M.: Determination of Serum Protclot by Means of the Biuret Reaction. J. Biol. Chem., 177, (1949) : 751-766. • Hald, P. M.: The Flame Photometer for the Measurement of Sodium and Potaasln* in Biological Material. J. Blol. Chem., 167, (1947) : 499-511. •Harman, R. E., Davis, G. E., Ott, W. H., Brink, N. G,, and Kuehl, F. A • Tht Isolation and Characterization of the Chicken Edema Factor. J. Am. Chem. Soc ST. (1960) : 2078-2079. " M Mallory, H. T., and Evelyn, K. A.: The Determination of Bilirubin with th* Photoelectric Colorimeter. J. Btol. Chem., 119, (1937) : 481-490. II Millonig, G.: Farther Observations on a Phosphate Buffer for Osmium Solution! l> Fixation. Vol. 2. Proc. 5th Internatl. Cong. Electron Microscopy. Academic Press New York (1962) : 8. n Mollenhauer, H. H.: Plastic Embedding Mixtures for Use in Electron Microscom wStain Tech., 39, (1964) : 111-114. "Quick, A. J.: Hemorrhagic Disease. Lea & Febiger, Philadelphia, Pa., 1957. 14 Bosenthal. H. L.: Determination of Urea In Blood and Urine with Diacetyl Monoi. ime. Analyt. Chem. 27, (1955) : 1980-1982. * "Sanger, V. L., Scott, L., Hamdy, A., Gale, C., and Pounden, W. D.: Allmentur Toxemia In Chickens J.A.V.M.A., 133, (Aug 1, 1958) : 172-176. ^ "Schmittle, S. C., Edwards, H. M., and Morris, D.: A Disorder of Chickens Probahli Due to a Toxic Feed — Preliminary Report. J.A.V.M.A., 132, (March I, 1958) : 216-211 " Simpson, C. F., Pritchard, W. R., and Harms, R. H.: An Endothelosls in Chlekw and Turkeys Caused by an Unidentified Dietary Factor. J.A.V.M.A., 134, (May 1, 195J); 18 "Williams, F. G., Jr., Pickets, E. G., and Dnrrum, E. L.: Improved Hanginc Sw» wf Paper-Electrophoresls Technique. Science, 121. (1955) : 829-832. 19 Wintrobe, M. M.: Clinical Hematology. 5th ed. Lea & Febiger, Philadelphia, p. 19ol. 80 Wootton, J. C., and Alexander, J. C.: Some Chemical Characteristics of the Edema Disease Factor. J. Assoc. Off. Agric. Chem., 42, (1959) : 141-148. CALCULATED DIETARY INTAKES OF CHICK EDEMA FACTOR (CEF) FROM DATA PUBLISHED BY ALLEN AND CARATMB Mean survival CEF intake time Dietary level of toxic fat (Percent) Per day (days) per animal (ug.) MONKEYS (MACACA MULATTA)' ai25. 0,50 L0_ 0.25. 5.0. 10.0. . -- . „ 445 274 202 169 90 0.225 0 45 0 90 1 80 80 m g IB 38 "~* CHICKS (DAY-OLD) » 3.O.. 1 1 15 0.36 Jfote on an Improved Cleanup Method for the Detection of Chick Edema Factor in Fats and Fatty Acids by Electron Capture Gas Chroniatography ir PAUL NEAk (Division of Food Chemistry, Food and Drug Administration, »t*ington, D.C. 20201) Calculations based on average daily food consumption of 225 gm. for a 5.0-kg. monkey (F. Sperling, personal, Calculations based on average daily food consumption of day-old chicks (personal studies, see Flick et a) Science, 45,630-36 (1966)). " e electron capture GLC screening test chick edema factor (1, 2) has been i| by replacing the siponific.alion step » snlfnric aciil cleanup (3. J) which per^ reiliiclion in sample cleanup time, modified procedure involves treatment c with sulfuric acid, frad ion.nlion of the petroleum ether extract from the sulfuric acid treatment on an alumina column, and sulfuric acid cleanup of the tKird alumina fraction, followed by electron capture GI.O. Gas rhromafpgrapliic peaks with rr-tcntion time versus aldrin of 10-25 are indii-ative of the presence of chick edema factor. �322 323 Table 1. Comparison of sulfuric acid cleanup with saponification for detection of chick edema factor by electron capture gas chromatography (ECGLC) ECGLC Analysis* H2SO4 Cleanup Sample" Saponification Low positive reference fat (1-5% toxtc fat in USP Cottonseed Oil) 10.1(45), 11.8(48), 17.6(65), 20.4 (60) 10.2(74). 11.9(82), 18.8(94), 20.6 (98) 10.1(147), 11.8(23), 17.8 (>500), 20.4 (200) 10.1 (18), 11.8 (14), 17.6 (40), 20.4 (34) 10.1 (50), 11.8 (23), 17.8 (200), 20.4 (82) 10.1 (48), 11.8 (14), 17.8 (258), 20.4 (98) Vegetable oil soapstock (nontoxic) 10.4 (trace), 13.5 (trace) 10.4 (trace), 13.5 (trace) Oleic acid (nontoxic) 10.4 (trace), 13.5 (trace) 10.4 (trace), 13.5 (trace) Cottonseed oi( (nontoxic) 10.4(23), 13.5(22) 10.4 (trace) Blank 10.4 (trace), 13.5 (trace) 10.4 (trace), 13.5 (trace) Toxic animal tallow Toxic oleic acid Toxic glyceryl monooleate 10.2(56), 11.9 («), 18.8 (85), 20.6 (90) 10.1 (98), 11.8 (10). 17.8 (>500). 20.4 (154) 0 Toxic and nontoxic refer to results of AOAC chick bioassay (AOAC Official Methods of Analysis. 10th Ed 1965, 26.087-26.091). 6 The first values (without parentheses) refer to retention time of peaks at 200°C vs. aldrin; the values in parentheses refer to peak area which is equal to retention timfe(cm) X peak height (cm). Method Reagents and Apparatus Rinse all glassware with appropriate solvent before use. Do not use polyethylene containers to store solvents (5). (a) Petroleum elhcr.—Reagent grade; redistil! in glass between 30° and 60°C (available from Burdick. and Jackson Laboratories, Muskegon, Midi.). (b) Carbon telmchlnride.—Distilled-in-glass. (c) Celile.—Johns-Mansville #545, acidwashed. Wash well with petroleum ether and dry. (d) Filler paper.—#549 S&S Blue Ribbon, or equivalent. Determination Sulfuric acid cleanup.—Dissolve 2.5 g fat in 10 ml CC14 in 400 ml tanker (heat, if necessary). Add 10 ml concentrated HjSO.i and (hen 20 g Celile; mix with henry glass stirring rod during additions and stir until homogeneous mixture is obtained. Add 125 ml petroleum ether, mix well, let solids settle, and filter the supernatant, liquid through filter paper in 90 mm conical funnel. Repeat with additional 125 ml portion of petroleum ether. Evaporate combined petroleum ether filtrate to 5 ml for alumina column fractionat.km. Complete determination ns outlined in the method of Higginbotham ct al. (1). Results and Discussion GLC retention timrs and prak areas for negative, positive, and blank samples were compared for both the sulfuric acid and tte saponificalion methods; see Tabte 1. Resiilu arc comparable as indicators of toxic materuj. Gas chromatographic peak heights were lower in some cases with the sulfuric a-cid cleanup; however, the presence of toxio factor »u clearly indicated in the low positive reference material. The nontoxic cottonseed oil sample would have been judged toxic by the sapoo*. Fig. 1—Gas chromatograms of (a) blank a low positive reference fat after cleanup saponification. GLC conditions: T X * mm i.d. glass packed with 2.5% SE-52 on 60-80 mesh Chrom Q at 205*0. Amount injected: I/50d» alumina column fraction 3. firalion method because of the relatively large GLC peak of R, 13.5, a peak delected at low levels in gas ehromatograms from blanks and other nontoxic samples; see Table 1. Small peaks of Ka 10-25 were observed in both procedures in the blank and nontoxic samples. However, they did not interfere with identification of the toxic fats and differentiation of toxic from nontoxic samples. Gas chromalograms of blank and low positive reference samples after saponification and preliminary sulfuric acid treatment are shown in Figs 1 «nd 2. Acknowledgments The author expresses sincere thanks to David Firestone for guidance and encouragement throughout the development of this project, and to Richard Staaf who performed many of the analyses. REFERENCES (1) Higginbotham, G. R., Firestone, D., Chavez, Linda, and Campbell, A. D., This Journal SO, 874-879 (1967). (2) Higginbolham, G. R., Ress, J., and Firestone, D., ibid. 50, 8S4-SS5 (1967). Fig. 2—Gas chromatograms of (a) blank and (b) low positive reference fat after cleanup with sulfuric acid treatment. See Fig. 1 for GLC conditions. (3) Davidow, B., ibid. 33, 130-132 (1950). (4) Dingle, J. H. P, Analyst 90, 638 (1965). (5) Burke, J., and Giuffrida, Laura, This Journal 47, 326-342 (19C4). Reprinted from the Journal of the Association oj Official Analytical Chemists Vol 50 Pen-inter 1967. ' ' �26. OILS, FATS, AND WAXES ' (1) The official, first-action GLC-microeoulometric method for chick edema factor, 26.092-26.096, was changed by adding the following to 26.092: (g) Ethyl ether for alumina chromatography.—Ether (not >2% alcohol) or absolute ether (not >0.01% alcohol) (available from Burdick and Jackson Laboratories). (2) The official, first action electron capture method for detection of chick edema factor, This Journal 50, 216-218(1967) was changed by addition of th* following to (b) in the Determination section : After ". . . 26.094" in line 6 add "(using ether specified in 26.092(g)" (item (1) above). (3) The following rapid screening method for detection of chick edema factor was adopted as official, first action: PRINCIPLE Samples are subjected to preliminary HjSOi cleanup and extd with petr. ether. Ext. is purified on A12O» column and examined by electron capture GLC, after addnl H»SO» cleanup. Gas chromatographic peaks with retention time relative to aldrin of 10-25 are indicative of chick edema factor. EEAGENTS AND APPARATUS (a) Petroleum ether.—Redistd in glass, b.p. ,30-60° (available from Burdkfc and Jackson Laboratories, 1953 S. Harvey St, Muskegon, Mich. 49442). (b) Ethyl ether for alumina chromatography.—Ether (not >2% alcohol) or absolute ether (not >0.01% alcohol) (available from Burdick and Jackso* Laboratories). (c) Carbon tetrachloride.—Redistd in glass (available from Burdick ud Jackson Laboratories). (d) Celite.—No. 545, acid-washed. Wash well with petr. ether and dry at room temp. (e) Aldrin standard, soln.—0.1 /tg/ml. See Reagents and Apparatus, sectk» (a), JAOAC 50, 216(1967). (f) Chick edema factor low positive reference sample.—1.5% reference torir fat in USP cottonseed oil. (Available from Division of Food Standards tat Additives, Food and Drug Administration, Washington, D.C. 20204). (g) Activated alumina.—See Reagents and Apparatus, revised 2a092/b> JAOAC 50, 216 (1967). (h) Alumina chromatographic column.—To dry chromatographic tube, JJ mm o.d. (14.5 mm i.d.) x 250 mm long, fitted at bottom with coarse porodtj fritted glass disk and Teflon stopcock (tube without fritted disk but boUta* glass wool plug in bottom may be used), add redisd petr. ether, dried be(Z use with anhyd. Na«SOi, until column is % full. Weigh 15 g Al:Ot and*tna*> fer to column in small portions, tapping tube as A12O9 settles. When last M*. tion of AljOs settles and air bubbles stop rising to surface, add 5 g anML Na2SC>4. Drain excess petr. ether so that it is just above upper surface rf NasSO«. (i) Gas chromatographic column.—Glass, 5-7' X Vi" i-d., packed with SE-52 silicone gum rubber on 60-80 mesh Gas Chrom Q (Applied Science oratories, State College, Pa. 16801). Coat support with substrate as follov»Weigh 2.5 g silicone gum rubber stationary phase and dissolve in 300 a| CHiCLrtoluene (1 + 1), heating to dissolve. Add 97.5 g Gas Chrom Q and te stand 10 mm with occasional gentle stirring. Dry in rotary evaporator held to 50° bath. Pack coated material into chromatographic column by adding ^ui amts while vibrating column at packing level with Vibro-graver tool ( Scientific Co., Pittsburgh, Pa. 15219). Fill to within 1" on exit side and entrance side, and fill remaining space with silanized glass wool. column at operating pressure 2-5 days at 250°. (j) Gas chromatograph with electron capture detector.—See Reagent* Apparatus, (e), JAOAC 50, 216 (1967). DETERMINATION (a) Preliminary sufuric acid cleanup.—Dissolve 2.5 g fat in 10 ml CCI, U ml beaker; mix with heavy glass stirring rod while adding 10 ml HjSO» 5 g anhyd. Na,SO< and stir well while adding 20 g Celite, until *• p mixt. is obtained. Add 125 ml petr. ether, mix well, let solids settle, and supernatant thru paper in 90 mm conical funnel. Repeat with addnl 125 ml petr. ether. tionation. Evap. combined petr. ether filtrate to 5 ml for AhOj column frac(b) Fractionation of petroleum ether filtrate 6y alumina chromatography.— Dry solvents prior to use by shaking with anhyd. NasSO*. Transfer petr. ether filtrate from (a) to AlaOs chromatographic column, using total of 15 ml petr. ether. Let liquid level fall to just above top of Na2SO.. Elute sample with 100 ml petr; ether (fraction 1), 50 ml 5% ether in petr. ether (fraction 2), and 100 ml 25% ether in petr. ether (fraction 3). Relatively fast flow rates of ca 8-9 ml/min give satisfactory results. Keep liquid level above top of Na>SO4 at all times. Discard fractions 1 and 2, and collect fraction 3 in 125 ml erlenmeyer. Add several boiling chips and evap. to dryness on steam bath. Transfer residue with petr. ether to 10 ml g-s. graduate and evap. petr. ether soln to 3 ml. (c) Sulfuric acid cleanup of alumina fraction 3.—See Determination, section (C), JAOAC 50, 217(1967). (d) Electron capture gag chromatography of petroleum ether extract.—See Determination, section (d), JAOAC 50, 217(1967). (4) The official, first action method for methyl esters of fatty acids, 26055-26.059, with changes in This Journal 49, 231-232 (1966), was revised ts follows: (A) 26.058(c), revised third paragraph, This Journal 49, 232(1966), after "... to obtain calibration factor." insert "Reference mists simulating most fats tnd oils may be obtained from Applied Science laboratories, Box 440, State College, Pa. 16801; Supelco, Box 581, Bellefonte, Pa. 16823; and Lipids Prepaittion Laboratory, Hormel Institute, Austin, Minn. 55912)." (B) 26.059, change to read as follows: "Two single detns of major components (>5%) performed in 1 laboratory shall not differ by >1.0 percentage unit Two single detns performed in different laboratories shall not differ by >3.0 percentage units." (5) The official, final action lead-salt ether method for determination of satnated and unsaturated fatty acids, 26.040, was changed as follows: (A) Change first paragraph, first sentence to read: "Accurately weigh 10 (for plant fats used in common household cooking oils) or 20 g sample into 300...." (B) Add the sentence "Reserve ether filtrate (contains ether-sol. Pb soaps)." to the end of the second paragraph. (C) Fifth paragraph, line 8, after ". . . HCl-free", insert "(no ppt with AS.VO*)." (D) Revise the sixth sentence of the fifth paragraph to read: "Distill ether, aroiding any loss of fatty acids, and heat over steam bath to constant wt «der controlled flow of N to prevent oxidation of fatty acids. Cover steam lath with towel to prevent splashing H»O into erlenmeyer." (B) Paragraph 6, line 1, insert "reserved" after "Transfer". Line 6, add •Bepeat HC1 hydrolysis until no more PbClj is pptd." after ". . . into beaker." (P) Change paragraph 7, lines 2 and 3 to read, ". . . until HC1 is removed (w> ppt in wash HSO with AgNOa). Dehydrate ether with ca 2 g anhyd. Sa£0t and transfer ether soln ...." (G) Change paragraph 8 to read: "Det. in duplicate I numbers of 0.2-0.3 g 41 from unsatd fatty acid fraction, and from entire satd fatty acid fraction. (I number of satd acid fraction is due to presence of some unsatd acid.)" (6) The following gas-liquid chromatographic method for butylated hydroxySBboIe (BHA) (121006) and butylated hydroxytoluene (BHT) (128370) in earn and rice breakfast cereals was adopted as official, first action. APPARATUS (t) Gas chromatograph.—Barber-Colman Model 5000, or equiv., with H tune ionization detector and strip chart recorder. Establish following operatK conditions: temps—column 160°, detector 210°, flash heater 200°; N flow Hie, sufficient to elute BHT in 3-4 min from QF-1 column and elute BHA in 1-1 min from Apiezon column; H flow rate, ca 40 ml/min for Apiezon L and a 25 ml/min for QF-1; air flow rate, ca 340 ml/min; electrometer sensitivity; •0 X (5 X 10 10) amp full scale deflection) with 5 mv recorder. Adjust H and «fe- flow rates, if necessary. Adjust electrometer sensitivity so 0.1 jug BHA gives ca 50% deflection. Bftpat injections until constant peak heights are obtained on successive injecgHB of identical vol. of std mixt. ef of appearance on Apiezon column (4') : BHA, BHT, di-BHA. Order of rance on QF-1 column (6') : BHT, BHA, di-BHA. �326 327 (Rtpn»icri from Wamre, Vol. 220, No. 5I6S, pp. 702-703, ft1 toxic fa* were raeh" fractionated by'prejmrnlive OliC (thermal conductivity detect ion), uxing a Ijfdoi x -1 mm column at 230" C packed with 10 per cent •n£.;»iN>' on 00/yii ju.-sli 'Gas Chmin <)'- Elated comwucnts fmi« tin* 2.3.4.C-te(nK-iiIoropIii*niiI pyroly^itc and fctin tlx1 rbloriimtiHl (iilM;nxo-7>-di'>xin wrre exnmined by drctrtin ejijit-un- (.:!/* and infrared *sjx-ctr«B!e<il>y. and were iiinlj'siHl by iiwsR sjH-t-trometrj* lo determine inolceu" ] one! niiinlier fif clilorine atoms per molmiK «• n^eonl-tl in Table 2. Tim conijKuieuts Ls*ilati-*l fiwn tlw referriifN- toxie fat have not yet I«vn rxautinor) C JtVALVKI* Jt'D •«« tWBJtYO ASijAV OF IrfOLJTEU COKPOXEXTS ;-.(2»ri Chk*eu embryo " away* percent /.'. *t t1llrk<-lt (rlnliryo ajwayt n-r M-nt iimrtiility - I-8J 5-51 3-1- 8-9 11-9 17-4 20-0 35-0 ]IM) 11-S 100 KO 100 * It «w nUnMtea t tot alwu t » i-« of material mm tnjcetnl bit o melt r»t t See footnote! 3 »nd 4, TalOc I . ? Com|vMKoU from HitorinatM dH^iuo-f-^loxlii. Chemical and Toxicological Evaluations of Isolated and Synthetic Chloro Derivatives of Dibenzo-p-dioxin ONK of the toxic compounds known as hydropericardium factor (chick oedema factor) is 1,2,3.7,8,9-hcxachlorodibewzo-p-dioxin'. Because it is highly improbable that theso very toxic compounds *re naturally occurring components of the fats from which they have been isolated, the question of their origin has aroused speculation. A cluo to the possible origin of hydrnpei-icardium factor in found hi a report by Tomiia ef al.* of the synthesis of polyhalodibenzo-4?-dioxins: eutorophenofs and: their salts, when heated, undergo condensation reactions and form chlorinated derivatives of dibenzo-^J-dioxin. Because of this observation, we studied the pyrolysis of a number of commercially available chloroplienols which arc widely used in agriculture mid industry, and present results of some preliminary experiments which suggest tlmt hydro* pericardium factor could nri-sc from certain chloropheiiols. The coninu-rcinl rhlorophcnols used in this work were pyrolyscd in «rcorrl«/ic-c with the general procedures described*. Benzene extracts of Uio reaction mixtures wero fractionated by passing them through an alumina column. Residues of the bcuzunc effluents were extracted with petroleum ether at room temperature, and the petroleum ether was then removed. The resulting products wi*re sufficiently pure in most cases for examination by electron capture gas-Iiouid chrninatpgraphy ((J1*C) and biological testing by the chicken embryo assay1. TivbJo 1 shows retention times of the GLC peaks on a 7 foot, 2-5 per eent 'A*£-52* coliunn at SOO* C ami ihc results of the cliiefceu embryo assay. Individually isolated hyclropcrictmtium factors of known chemical struct im*s wero not available for uso nx standards; ccmsotaK-ntly. a c*«i«-«lr»lc from the misiii>or/J/i-ibl»?fr«rlH«( of a certain commercial toxic fatty acid nmtoria! was ust*d «,s n <JIX" refrifiiee. Tlie tmsaponifiablu fraction of the fatty acitl material wan known to contain trace amounts of the hydropcricardiiim futitor. The toxic fatty acid material, a by-product obtained from the manufacture of ofoic* ami stearic acids, has been uwd as a reference in all our *n vious chemical and biologicitl work4. TcUe l. £ii3n.fij7os OF cntOKortixxoi Sample pymljfed* 2,4-Dtdilnroptwnol (K) 2,4.&-TrieWon.]ArnoJ fT) £.4.ft-TrtchIorot>)Knol (11) 2.3.4.6-TctrachW.plicBo) (T) Itcfrrcnce loxfe fat poneoli com- 6004 8-9.10-0, 11-2,17-4, 0-25 6-0 20-0, 36-0 35-0 1-0. 1-8, 3-5. 6-5 1-0. 1-6, 3-S, C-5, 8-9. 10-0. ife. 17-4; 20-0,35-0 3-0 •(B)-rcist"iit grade: (D<=technlcal grade. t JJ«= rrti-nUon time* of peaks n4nUv* to the retention time or 1 1 itf^SO n.p.b. In the c-g. Sample injected Into the air fertile ess* before incubation, SotvcuU toed: ctiianol, acetone f Embryonic mortalltr mi 21 Amy*. The mortality of •otvciil-Inji-ctct! controls «s 10-15 per cent. Tlra chromatogrojn of components from the toxic *j acid material contained a number of peak* of varying retention times; four of theso pculu values greater than 10. According to a current r capture GLC test4, the prescnro of ono or more peaks with /.'« values of 10 or more indicates that rn^*; jK-ric-rtrtHiiin factor is present in a fat. Ilio pyrob^^ from tL-clmicnl grado 2,3,4,0-tetntchlorophenol u-»« n^ only prftfluet lliat sliowed a peak pattem indk'athBr ^ presence of hydropcricnnrium factor; ttio poak ptttoM resembled that disploye*! by the long retention tiract^^. poueiils of the refrn-nco toxic fat. When to*t*xi b« ^ official chk-k hioossay* for hytln>pcricar*Iiiun factor |W mixture produced the disease at a dk-laiy IcvnJ ofO-lp*p^ In addition, dibenxu-p-tliuxiu wan chlorinaliil in ttrcJor to prejwrp 2.a,7<8-lotrac)t|»roditxTtzo-i^ This inaleriat proved to bo fxtrviwly toxic in tliu embryo assay (Table 1 ). nml provided a sourcn for Ho of a trichU.ro as well as tho totmchloroflilynxo-p This technical prado 2,3,4 .0-tctrachlorophcnol ttto. chloriimlwi dil>ouxt>-p-dioxiii, and Infhirnl six»ctra of individual components exhibited ttomctcristic absorption bandit in the region of 1,3301480 cm-' attributed to Asymmetric si relcliitig vibration •f C-*-O— C in tlie dibcnxo-p-dioxin ring. The- two prind-m] compounds isolated from the chlorinated dibenzo-^. facin \i-cm 2,3,7 -trichlonxlibenzo-yi-dioxin and 2,3,7,8IgUwliloitMlibenzo-p-dioxin. Tlie infrared spectrum of AM trichloni compound ha* baiids at 870 cin-» and 805 «r* (J,2,4-s»bstitHtcd benzene). «*K! Sfiw cin-1 (1,2,4,5l^Btitutetl Ixuizcne). Tl» infrared t^inKruin of the Vtrachlnro compound 1ms a strong I wind at 868 cm-' (Lt*,5-sulwtitutcd benzene), llw infrared spectrum of d* tctwchloro compottnd inatelw-d tl«j Rpectnim of tS,T.S-fetrne!tloro<n"benzo-yi.dioxin pubhVIied by Tomita Wrf.*. Chiefcra vmbryo assay showed thai the tetrachloro tMipowid was more toxic than UM; trichloro compound. Tl» first four peaks in the chroimito^nnn from the iropIienoI pyrolysnte wero caused by positkina] , eacli containing six chlorine atoms.; the fifth . «d astli peaks were caused by positional isomcrs, each ctoUiiiing serrn chlorine atoms. A ec\-eiith peak in the nun, observed M-hcn the nunple injection was , had a retention time identical to that of a of octaclitorodibenzo-p-dioxin prepai^d by pyrohM of pcnUehlorophenol. The presence of pentaiicrvphcuol in technical grade 2,3,4,6-tetrachIoroplwnoJ ^counts for the formation of the two hcptachloro isomenj ^1 the octachlarodibcnzo-^-dioxiii. ,A fourth hcx»chloro isamc^r, in addition to tlie threo ^petted from pure 2,3,4,6-tetracliloroplicnoU is prolwbly MMrd by tlie prestmce of a tetrachloi-oplienol other tluin «V t3,4,0-isoincr in tlie starting material. Infrared '•Kin of peaks 3 and 4 from Iho 2,3,4.G-tctraclilon>phcnol ato vftan very simitar to published Kpcctra of two factors isolated frr>m n conUminated e eompononJ* isoltit«x) by preparative CVU: from the icu loxic Kit. tlie chlorinated dil)i-ii/.o./i-rlioxin. mul gV •,3,4»0-U1iriichloronheiml pyrolysut<- wt-ro antilvMed by •kctnui cnpt lire GLC at SiKi" C and liy the ehickcn tiiibryo &tf, 11*"* results an- shown in Table 3. The quant it v ^BMtcriiil in "each fnieU-.n from tlie rcfi-iviiec toxic fn't «K wtiintil«l on the Iwisia of the wei-ht of snmpli: inl" tlie preparative gas ehroniiiK.^raph and the of fnictions collected. Tlie amounts of malerinl l fmm tlie ehI.mnat<*ddibeii/,t>-77.|Jiosii, nn<J 2.3.4.0, k»x>phciinl pyix»ly«ite wero rough estimate I«i*xl n}Htri<on of CLC pmk areas of i.solatttl coiiiiH>iR-tit.s 01 UK- Marling mixttm^. Amounts of individual com* pMKitx in the mixtures were estimated frmn uoniializi-<1 pvk nn-ii calni/atinns. Ilic dafa supgor <luit onn liexa in. Grew *r*un by and one hopta-teonicr an> lew* toxic Uwn the other isomers. Work is now under way to prepare- individual comixmnds of high purity in sufficient quantities for additional chemical and biological testing. Reference toxic fat components with low 7?« values (Jcsu thnn /ft* 8-9} <'oti)d bo <tiK? to ehloro-organic pesticides and/or lower chltvodibcnzo-p-dioxins. Additional work is required to kJcritify tho compounds responsible for the pea lot with low Rm values. The peaks with high Ji, values .(tliose «|tiAl to or groatcr than 8-0) correspond to tiiOKO of the IwxacJiloro and hoptnchloro positional isonn-w isolated fr»jm the pyrolysate of 2,3,4,0tctraehloropliennl. The o\-erall results of this work suggest that chloropbctiok iwuld Ixj precuraora of hydropericardium factor. Commercial chlorophonols are widely iucd for such diverse applications as contact herbicides, defoliants and lennite control agents, as well as for eontrol of microbial atiatfc hi the manuitteture of a number of products. There arc many opportunities for fats and fatty acids* the only product* in which liydropcricardium factor baa been found, to become contaminated. When crude fats and tallows ai-e subjected to heating operations (hyilmlyjiw, distillation) in tho production of commercial Catty acids, residues of commercial chlorophcuols prewmt in tiro fat might be converted in part to chick oedema factors. Additional work is rw|uirrcl 10 detcnnine wlietlicr hydropericardium factors are formed from residual chlorinated phouols during prtxhiction of cojJ)nw-rcmJ fatty acids. This laboratory w developing metluKls for detecting chlorophenols in fats and fatty' newts. We tlwnk *Dr Jo-Yim ClK>n, Dr Donald F. Flick, Mr Robert liorron, Mr James Sphun, Dr E. E. Ileynaldo, 3fr William F. Scott ant! 3frs Mar>- K. Mutchler for technical assistance. G. H. Hi ANITA HTJAKO DAVID FIRESTDKK JOHK R>3S . A. D. Bureau of Sciencv, FootI and l>mg Administration, Department of Hralth. Rductitiou atul Welfare, Washington. IV.'. KwrfVttt AttgiUt 5; irvfrnt Sppl(rnil«i 27. 10 , 46. So. 5. S., nn.l VnU. . . if.. l'**wwi£« / Af. 79. • Vrm-lt, M. J,. M.-irli.f-, J., ami Mf LaiialJin. J^ J. A***. Off. Agr. CktmiXt, 47, 1003 (1034). ' Xt-al, V., J. A**of. Offif. AM!, flvt*;*!*. SO. 1S3S (1Q07). • OfffM A/.ffcrf' t.f A»alt»i*. lOtll ^d.. *»f*. 2rt.OST-2fl.001. (Auorlatbm of Official Aerfcnlliir*! I iKtuHt*. Washing ton, U.C., IMft). *lV<w*(on. J. t'.. -trtnwn. ST. R.. **l At-viw*-r. J. <*., J. A*w. Ojfie. Atr. CA. «iHf*. 45. 73<» ( |90£). l A Co.. Lid.. Si. Attatt �328 INTERNAL PRELIMINARY REPORT ANALYSIS OF COMMERCIAL CHLOBOPHENOLS FOR TRACE AMOUNTS OF THEIR CONDENSATION AND PLYMEBIZATION PRODUCTS 329 REFERENCES iLaclalr, J. B., Anal. Chem. 23, 1760 (1951). »Clem. Abstr. S3, 1315d (1959). By G. R. Higginbotham and John Ress Laclair (1) has reported an ultraviolet absorption study of the components of technical pentachlorophenol, separated by vacuum sublimation, and disclosed only pentachlorophenol (POP), 2.3.4.6-tetrachlorophenol, and an unidentified, dark brown, high melting, "chlorophenol" containing 58.3% chlorine. The unidentified material was presumed to be a polymerization product produced during process of manufacture. Tomita et al. (2) have demonstrated that chlorphenols and their salts, when heated, undergo condensation reactions and form chlorinated derivatives of dibenzo-p-dioxin (chick edema factors, CEF). Recently, we examined the unsaponifiable fraction of a number of commercial chlorophenols and obtained results which suggest that the impurities of chlorophenols consists of chloro derivatives of dibenzo-p-dioxin in addition to other components which have not been identified. Thirteen chlorophenol samples were analyzed for chick edema factor. The samples were carried thru the 2.5 gram saponiflcation procedure and examined by electron capture GLC after AlsOi column chromatography and HtSO extraction [JAOAC, 50, 884 (1967) 1. GLC peaks indicative of chick edema factor were observed in most of the chromatograms from the thirteen samples. Six of the thirteen samples were selected for further study. The unsaponifiables from each sample were extracted and chromatographed on an Al»Qi column, according to A.O.A.C. Method 26.092-26.094. The unsaponifiable fractions were submitted to the egg embryo biassay. Two components (Ra 6.3 and Ra 11.6) were isolated by preparative GLC from the unsaponifiable fraction of Eastman's technical grade 2,3,4,6-tetrachlorophenol. The components were examined by mass spectrometry and submitted to the egg embryo bioassay Results are tabulated below. Table I shows the amount of unsaponifiables isolated from 100 g samples of six representative commercial chlorophenols. In every case the amount of CEF in the total sample is estimated to be less than 0.3%. If a lipid sample is contaminated with pre-formed chick edema factor,* high levels of poly chlorophenols would also be expected to be present in the sample. Results from the chick embryo assay are recorded in Table II. Table m summarizes the results obtained on two components isolated from 2,3,4,6-tetrachlorophenol. The molecular weight and the number of chlorines for each component are not consistent with a chloro derivative of dibenzo-p-dioxiiL The Ra values of these components are not consistent with known chiA edema factors. The compounds may be photodecomposition products of technlcal grade 2,3,4,6-tetrachlorophenol. Further work will be required to characterize the unsaponifiables from commercial chlorophenols. The overall results of this preliminary study emphasi» the need for a rapid method for polychlorophenols in fats and fatty acids. Th» unsaponifiable and CEF (hexa-, hepta- and octachlorodibenzo-p-dioxin) content of a number of reagent and technical grade commercial chlorophenols are shown in Table 4. CEF content was determined by GLC using a synthetic CEF mixture as a reference material. CEF was found in all the chlorophenols examined, varying from a trace (ca 0.001 pg/g) to 205 pg/g. * Pre-formed chick edema factor Is defined here as a group of chloro derivative* rf dibenzo-p-dioxln Initially present In a sample before It Is subject to any type of KM* treatment. ^** TABLE I.—EXAMINATION OF CHLOROPHENOLS FOR PRE-FORMED CHICK EDEMA FACTORS Wt Unsap.. Sample (100 g.) Big. Wt. AljOi ft. 3 Percent CEFi 2.3, 4, 6-Tetrachlorophenol (Eastman) > 4,5-Trichlorophenol (Pract) 13, 4, 6-Tetrachlorophenol (Baker) zliVichlorophenol (Baker) i Estimate based on weight of AW, ft. 3 and assuming that the fraction consists only of CEF «A second extraction of the unsaponifiables (after acidification) of the soaps afforded 5 0 mg of material g i A second extraction of unsaponifiables gave 9.8 mg of material material. TABLE 2.-CH1CK EMBRYO ASSAY Sample (unsaponifiables) /ig/egg Percent mortality jl 5-Trkhlorophenol (T) 0.55 15 0.6 4.8 {346-Tetracnlofophenol (Eastman). 30 0.60 55 70 50.0 40 y-Dichlorophenol CO TABLE 3.—EXAMINATION OF TWO COMPONENTS FROM THE UNSAPONIFIABLE FRACTION OF TECHNICAL 2,3,4,6TETRACHLOROPHENOL R. of GLC component Chicken embryo assay No of Mol.wL Cl atoms ig/egg 123 UJ 408 442 7 8 Percent mortality 2.0 1.0 0 0 TABLE 4.-EXAMINATION OF COMMERCIAL CHLOROPHENOLS FOR CHICK EDEMA FACTORS (HEXA-, HEPTA- AND OCTACHLORODIBENZO-P-DIOXINS) Compound* 24-Dfchlorophenoi(R) ZS-DichlorophenolOO 24,5-Trkhlorophenoirr) 2.4.6-TrichtorophenolCT) £3A6-TetrachloropheiH>I<T) 1 1 CTrirhfnmnhpnftl fO\ 2A4,6-Tetrachlorophenol(T). Penlachlorophenol(R)".III"Iim DowcideB (sodium 2,4,5-trichlorophenate <T).V" 3 Dowcide 6 (sodium pentachlorophenate (T) 4 Reference toxic fat » t-rtigent grade. T=technical grade. • Hen-, hepta- and octachlorodibenzo-p-dioxins. mg. Unsap. ^t/g. of CEF* 2.0 0.018 0.219 2.4 V9 5 12.3 1.9 22.5 3.0 35.0 67.1 1.4 18 25.8 223 trace trace 0.013 205 0.021 96.5 121 0.167 47.0 3.0 �331 330 Reprinted from Acta Cryslallttgraphica. Vol. B2S, Part I, January 1969 PRIftfTTJi IN DENMARK «-3-l7 material, docs indeed produce the hydropcricardium condition in chickens.* Experimental Acta Ciyst. (1969). B2S, 150 The Identification and Crystal Structure of a Hydropericardiura-Producing Factor: 1,2,3,7,8,9-HexachIorodib'enzo-p-dioxin BY J.S.CANTRELL,* N.C.WEBB AND A.J.MABist The Procter St. Gamble Company, Miami Valley Laboratories, Cincinnati, Ohio 45239, U.S.A. (/JnrricK/4 December 1967) A crystalline material, isolated from a contaminated animal feed fat, and capable of producing hydropericardium in chicks, was shown by solution of its crystal structure to be 1,2,3,7,8,9-hexachIorodibenzc-p-dioxin (CizChHzCU). The triclinic unit cell has the dimensions a=7-952±0-005, i=9-379 ±0-01, c=9-433±0-01 A, i=92-35°±0-20°, £=92-39°±0-20°, j.= 109-92'±0-30°. The calculated density is 1 -958 g.cm-3 for Z=1, compared with 2-01 g.cm~3 measured for the bulk material. A statistical treatment of the 1158 measured reflections indicated a center of symmetry; the space group was therefore assumed to be PT. The structure was solved by the symbolic addition method of Karle & Karle. The nearly planar molecules are almost parallel to the (Oil) crystallographic planes. No unusual bond lengths or angles were found. The structure was refined to R= 10-5%. Two types of crystals were isolated from a warm benzene—hexane solution of the a-3-17 material. The bulk of the crystalline material appeared to differ in phase from the material used for this study. No crystals of the bulk phase were found to be satisfactory for single-crystal studies, and only two crystals of the studied phase were isolated. Measured j-spacings of X-ray powder patterns taken of the bulk phase material did not match (/-spacing; calculated from the unit cell of l,2,3,7,8,9-hexachlorodibcnzo-/Mlioxin. However, when the bulk phase was heated to just below the melting point (230"C) a phase change occurred. Measured rf-spacings from X-ray powder patterns of the transformed bulk phase match the calculated dspacings of 1,2,3,7,8,9-hcxachlorodibenzo-/>-dioxin reasonably well. Therefore, it was assumed that the material used for this crystal structure determination was t high temperature phase of the bulk crystalline material known as ot-3-17 HPTF. The single crystals used were diamond shaped and had the approximate dimensions 0-18 x 0-10 x 0-08 nun (flxixc). The uni t-cellparameters were determined from singlecrystal data using a General Electric single-crystal orienter and Ni-filtered Cu KB radiation (>.= 1-5418 A). The parameters of the triclinic cell chosen according to Dirichlct (Balashov & Ursell, 1957) are as follows: «=7-952+0-005A «= 92-35±0-20° Ce = l-958g.cm-' i-9-379 + 0-01 ^=92-39 + 0-20 e.=2-01 g.cm-' }>=109-92±0-30 Z=2 e«9-433 + 0-01 K=662-8 A' where <?o was measured for the bulk phase by flotation. Two-dijnensionally integrated equi-inclination Weissenberg data were collected for the a-axis zones, Okl4lcl, and for the fr-axis zones AO/-A5/ using the multipleJhn technique (one pack each of four films, Eastman • The composition for the structure reported here, namely CiiOzHzCta agrees well with unpublished microchemical uuhscs performed by Professor Wolfgang J. Kirsten, Unitosily of Uppsala, Uppsala, Sweden, at a very early stage of Ifais structure work. Quantity Introduction The isolation, chemical analyses, and spcctroseopic data on the hydropericardium toxic factor (HPTF) material have been described by Wootton, Artman & Alexander (1962), and by Wootton & Courchcnc * Present address: Miami Univcrshy.Departmenl of Chemistry, Oxford, Ohio, U.S.A. t Reprint requests should be addressed to this author al ihc Procter & Gamble address. (1964). One of the active fractions of material isobfci was that called a-3-17, where this nomenclature rcfen to the vapor phase chromatographic behavior as fe scribed by Wootton er al. (1962). Wootton and his a*. leagues proposed that HPTF was a chlorinated kxthydrophenanthrenc with the empirical CUH1(,CI0. Following the molecular idon herein reported, Wootion (1965) showed that thctic hcxachlorinated dibcnzo-/)-dioxin, whose pkjv ical properties arc remarkably similar to the isoterf Non-ccntrosymmelric 0-886 0-736 1-000 Kodak No-Screen). Intensity data were recorded for both crystals, reduced separately, then compared, edited, and averaged. Absorption corrections were made separately for each crystal using Busing & Levy's general absorption correction program as modified by Jeffrey (1964). Owing to the very tiny crystals, and in part to the . integration, very long exposures of approximately 150 hours were required to obtain satisfactory multiplefilm data. The entire Wcis^enbcrg camera was placed inside a plastic bag and a helium atmosphere was provided to reduce background due to air scattering. Of the 3030 possible reflections, 1158 (38%) were recorded; 397 of these reflections had intensities !css than a minimum threshold value and were classified as 'less-thans*. The intensities of most of the reflections were measured by a Joyce Locbl microdensitometer scanning at right angles to the longer integration direction. The weakest reflections were estimated visually. A standard intensity strip wss prepared and used for the visually estimated intensities. To ensure tliat both types of intensity data were on the same scale, a sufficient number of medium intensities were measured both visually and by the densifometer. Radiation damage effects were found to be negligible by retaking data for earlier crystal settings. Statistical treatment of the intensity data by K.imachandran & Srinivasan's (1959) modification of the method of Howclls, Phillip & Rogers (1950) indicated a center of symmetry. The space group was assumed, therefore, to be AT(CJ) and this assumption was confirmed during the direct method calculations. Solution and refinement of the structure Initially we knew the weight of the molecule and the number of chlorine atoms per molecule, and we knew that the molecule possessed some aromatic character. Attempts to solve the structure from the three-dimensional Patterson map were not successful. The symbolic addition method of Karle & Karle (1963, 1966) was then employed. The phases were determined for the 251 most intense reflections in terms offour algebraic quantities, a,b,c,g. A summary of the calculation of the unitary structure factors or £-values used for this determination is compared with theoretical values and is as follows: Centrosymmetric 0-79S 0-968 1-000 0-3% 5-0% 32-0% Karte a al. (1964) Wndolyl- ftccttc acid 0-769 0-772 0-9.14 0-970 1-031 1-000 02% 0-4% 3-3% 4-5% 36-1% 30-8% 1-158 reflections 1-2&) reflections 761 non-zero 86"5 non-7ero 397 unobserved 424 zero (-Icss-than1) �332 The overall temperature factor was 2-24. A summary of the sixteen cases that resulted from the sign permutation of the four algebraic assignments is as follows: Case 333 itfying some of the 'less-thans' to observed reflections man (1959); those for oxygen by Ucrghuis, llaanappel, oa the basis of visual estimations of the weakest re- Potters, Loopstra, MacGillavry & Vecncndaal (1955); fections. and those for chlorine by Dawson (1960). The R value dropped to 13% during the next Icastimurcs refinement. Fixing the chlorine atoms and reDiscussion faing only the carbon skeleton icsulted in an K value of |l-8% at which time the aniso'ropic temperature The molecule is nearly planar and the molecular plane idtnemcnt on the chlorine atoms was undertaken and is tilted 8" to the (Oil) plane. Deviations of atoms from ftn the final R of 10-5%. Of the 397 'Icss-than' rellec- the (034) plane essentially containing the molecule and l»BonIy42 calculated were larger than their threshold talucs and none by more than 36%. A weighting funcbon similar to that given by Hughes (1941) was chosen V, as to have little dependence on !•'„; it was taken as some changes mode in carbon and chlorine a»ij». mcnts resulting in, ideally, a plan;.:- 1,2,3,7,8,9-hco. chloroanlhracene with mm symmetry. This change • the structure dropped the R value lo 35%. The plax 10 1 II 12 13 U 4 H- No. rcfi. generated No. incorrect signs % wrong sign Objections to choice •f 119 - 132 117 183 68 132 114 137 125 153 98 132 142 109 133 46-6 52-6 49-8 52-6 53-0 116 135 118 133 251 0 121 130 121 130 115 136 125 126 121 130 129 130 129 121 128 104 130 35 122 129 61 24-3 m 12» 123 IB 117 120 47-8 tU, *, where / r «,=21-4,andlheqtiantityminmu>cd as t uiFo - Fcf. 'Lcss-than' reflect ions having 7"c > Ft (totshold) contributed like an ordinary reflection, but Aose having FC<1-,, did not influence the refinement jl all. A full-matrix least-squares procedure was used Anwghou! the refinement, and in the las( cycle all parameter shifts were less than JCT. In a difference map ofculatcd at the R = 11 -8% stage, the residual electron fcnsity ranged from -0-8 to +0-6 e.A-', and the dxima and minima did not show any chemically or Uncturally significant features. The final position and thermal parameters are given • Table 1. The observed and calculated structure factors jrc listed in Table 2. In this work, the scattering fetors are taken from International Tables for X-ray Ojaottography (1962); those for carbon arc by Frec- J </ Electron density peaks v.-ere found at centers of symmetry when the special sections .r=0 antj x=\ were computed t It is very unlikely that all 251 most intense reflections would be positive. * This correct case was selected after cxamininj: three-dimensional electron density maps for Cases 9 to 16. Case 13 was chosen over the other seven acceptable choices because of the appearance of a chemically reasonable structure in the electron density plot in an (034) plane. Jt had already been determined lhal the molecule had to be oriented approximately in alternate (034) planes, based on (a) the early analysis_of the Patterson map, (6) the very high intensity of 022, (c) electron densities calculated from models based on the Patterson vectors, and (d) packing considerations. A nearly planar hexachlorinated phenanthrene structure was initially fitted to the £-map peaks-located near an (0?4) plane. An K value of 50% was obtained for the initial trial coordinates ajid a three-dimensional electron density map suggested that the phenanthrene skeleton should be changed to an anthracene one with of the molecule was original! . in alternate (0?4) plaaa but required some tilling and when these changes »T»I made the K value dropped to 24%. One cycle of lew squares reduced the residual :o 19%. At this point tic two bridging atoms were recognized as oxygens, suxe their relative electron densitu-^ were 30% higher tla^ those for the carbon atoms and the individual te». pcrature factors for these atoms had gone neguhe. Subsliluling oxygen atoms tbr these carbon atoms * the proposed model resuhec in an R value There was a carcfttl ediling of the data, especially «f those reflections classified as 'less-thans1, at the Jt* 1 6% stage of refinement. This editing consisted of Ay leting a few doubtful reflections recorded near the cfe of a film, correcting transcription errors, and recfa*. Tiite 3. Distances of atoms from least-squares plane!: nuances for atoms nor defining tlic plane are marked with an •ant):; S-d. is the standard deviation of the atoms dctininc •r fbnc. Under the (04M) heading are listed (he deviations tw the (044) plane containing tlic molecule. The phncs are *faol in direct space by equations Px+Qy+Kz=s. Table 1. Final parameters and their standard deriations V &^B gm The fractional coordinates have hccn multiplied by 104 and the temperature factors by 102; the vtandard deviations uv • «. rcntheses. The anisolropic tempera urr factors of tlic chlorine atoms are in the form | a1 1f 2 cxp[-i(fl ,/j;0*i-t-2A i2//a'A-fr* + #2:& i*2. ..)' X 0(1) CK2) CI(3) CK4) Cl(5) CK6) 0(1) C(2) C(3) C(4) C(6) C(7) C(8) C(9) an) C(12) CO 3) CO 4) CX5) CHIO) -0099 (6) -0145 (6) 3314(7) 9J93 (6) 6188 (6) 2919 (6) 1762 (24) 1718(21) 32S6 (25) 4753 (24) 7712 (23) 7706 (2(1) 6219 (27) 4754 (21) 3292 C.1) 4776 (24) 6226 (21) 4752 (20) 6312(15) 3266(15) V 3158 (6) 0579 (6) -0257 (6) 7375 (6) 8434 (6) 6936 (6) Z 5185 (6) 2931 (5) 2621 (5) 2VK (.-!:> 10547(5) 10752 (5) 8629 (5) 4879 (20) 3979 (IS) 3853 (20) 4673(21) 8394(19) 93S8 (16) 9503 (22) 8539(17) 5778 (19) 5629(19) 4811 (19) 5271 (I?) 7461 (18) 7591 (17) 3742 (M) 6-196(13) 2601 (22) 1502(20) 1118(22) 1920 (23) 5516(21) 6572(191 7051 (24) 623S(I'J) 3471 (21) 4591 (14) 671.1 (13) B or fin 287 (20) 362 (23) 487 (26) 372 (23) 395 (24) 282 (21) 354 (41) 262 (J5> 3S6 (43) 370 (42) 321 (39) 219(34) 427 (45) 249 (35) .325 (39) 36'J (421 240 (34) 213 (32) 355 (27) 334 (27) *H 376 (29) 378 (30) 411 (30) 457 (30) 375 (29) 421 (29) •Bsj 487 (26) 417 (27) 376 (26) 347 (24) 309 (23l 473 (27) Jin 149 (21) 025 (22) 163 (24) 174(23) 076 (22> 148 (21) Bit fa -035(17) -162(19) -056(21) -OJ8(18) -Oftca -dura -OUOK -I05c3 034 (18) 028(18) -owS -|02taJ 1 £ \ | ;c --y ;i , ^ ^i J. > '1 • '^1 "I 7 <% •; ,t* i '-?• : 0") cini vn->( 00) CX4) an at> an era 01) CM) Q*) an Oft cm am 0121 an CII4) «S) wo Oripn li r Q f s All -0-10 -0-11 0-00 -0-16 -0-05 0-13 -0-02 -0-06 -0-02 0-06 0-00 -0-05 -0-01 -001 006 <M>5 005 006 0-08 0-09 1-57* 0-07 -1-1589 -5-9655 6-8500 1-5675 C+O -0-10' -0-12* -0-03* -0-18* --0-06* 0-13* -0-03 -0-08 -0-05 0-03 -0-03 -0-07 -0-03 -0-02 0-05 0-03 0-02 0-05 0-05 0-08 1-55* 0-05 -1-1086 -6-0031 6-840-1 1-5497 (0?4) 033« 0-10* -0-25' -0-45* 0-12« 0-54* 0-15* 0-02* -0-16* -0-17* -0-25* -0-21' 0-03* 0-13" 0-13* -0-09* -0-10* 0-12-0-17* 0-25« . 1-67* 0-00 0-0000 -6-6618 6-6618 1-6655 viz-y+i/4) »• Fig. 1. Molecular packing in the (04"4) plane containing the molecule. Large "shaded circle, ai c O, solid circles ate C, and open circles arc O. ,01(3') Fig. 2. Projection onto Tf7. plane. �334 from the least-squares planes of the entire molecule and of the carbon-oxygen skeleton arc given in Table 3. The molecule appears to be slightly bowed in the middle and slightly twisted about a line from Cl(3) to Cl(5). The packing arrangement of chlorines 4, 5, and 6 appears to be more crowded than that for chlorines 1,2, and 3. This packing difference could account for the slight twiit of the molecule. Fig. 1 pictures the molecular packing in the (054) plane containing the molecule, and Fig. 2 gives a projected view of two adjacent molecules related by the center at (i,i>'D- Intermolecular distances in this (044) plane of less than 4-0 A are shown in Fig. 3. Between centrosjTnmctrically related molecules there are a number of CI(n)-CI(m') and equivalent Cl(»/)-CI(H') distances of 4-0 A or less. From the parent molecule to the one related by the center at (i,i,-V) the distances are CI(2)-C1(4') = 3-85 A, Cl(3)-CI(5') = 3-66 A, and d(3)-CK6') = 3-S3 A; by the center at (0,0,-l)-CI(l>Cl(3') = 3-84 A; by the center at (0,},4)-Cia)-Cl(l') = 3-39 A and C!(2)-CI(6') = 3-9S A; and by the center at \3-46 3-SS/ 335 (1,1,1)-CI(4V-C1(6')=4-00 A. The least-squares piano of the two adjacent molecules related by the center at (i,i,-l) are 3-13 A apart; between these two molecule* the shortest interatomic distance is 3-30 A froma Cfia toanCXIO). Fig. 3 indicates the bond distances and angles, the mean standard deviations arc as follows: ac~c» 0-025 A, o-c-ci=0-OI9 A, ffoo=0-022A; for angle, <r=2-0'. The bond distances are not significantly dg. fcrent from those found by Davydova & Slrudilcor (1962) and Gafner & Hcrbstein (1962) foi l,4,y. tetrachloronaphthalenc where molecular over-crontf, ing results from the presence of many chlorine atogtt substituted on adjacent aromatic positions. This compound belongs in group (I) according to the classiS. cation due to Harnik, Ilerbs5ein, Schmidt & HirshfcU (1954) for compounds that are affected by molecular over-crowding. An electron density map plotted in the (04"4) plane containing the molecule is shown in Fig. 4. The authors wish to acknowledge fheir appreciation to DrJ. M.Stewart of the University of Maryland, who furnished the X-ray 63 computer program and pn>. vided much information on its use. In addition, «e wish to thank Dr Lyle Jensca for a number of helpful discussions on the use of the X-ray 63 computing tyv tern and on approaches to the solution of the structure in general. We wish to express our appreciation to Drs Jer«K and Isabella Karlc who provkk-d assistance in appl)i« the direct method for determining the phases tt a number of the most intense rcHcctions. Thanks are. due to Mr Robert Gloss who obtained part of the data and provided the computer program used in generating the relations between reflection necessary for applying the direct method. References BALASHOV, V. & URSELL, H. D. (1957). Ada Cryst. 10,50. BERGHL'is, J., HAANAVPEL, U. M-, POTTERS, M., LoonnA, B. O., MAcGitiAVRY, C. H. & VEENENDAAL, A. L. (\9tn JaaCryst.S.m. * DAVYIWVA, M. A. & STRUCHKOV, Yu. T. (1962), n Strukt. Kliimli, 3, 184. DAWSOS. B. (1960). Ada Crysl. 13, 403. FtiumK, A. i. (1959). Ada Crysl. 12, 261. GAFNER, G. & HERIKTEIN, F. H. (1962). Ada Crm i« IOS1. '^ GAFMR, G. & UERKSIEIN, F. H. (1963). Nclure Load JM 130. "^ HARNIK, n., 1 IERBSIEIN, F. H., Scmtnn-, G. M. J. ra.n, F. L. (1954). J. Cliem. S<*c. p. 328S. HOWH.I.S, E. R., I'Hit urs, D. C. & ROGERS, D (19301 (*) Fi'E.3. (0) Interatomic distance*. Primed atoms are on wiphboring moiL'ciiks in the same plane. oc-c~0-025A; ffc-ei "0-019 A; oc-u •--= 0-022 A. View correspond:, to Fig. 1. (4) Bond angles, n-2-0*. Aaa Cr.nl. 3. 210. "^ HIKJHCS H. W. (1941). /. Aimv. Chan. Sac., 6\ 1737. Inlcniaiiiinat Tables far X-rflj- Cri-stallogi-rijfhv, (1962). Vot 111. Hirminfihiim: Kynoch Press. jErTKi-v, G. A. (1964). Private communication. Cl(21 FSg.4. Electron density in the (034) plane containing the molecule. Contours are at 1 e.A~3 starting at 2 e.A~3. The x marks are * projections onto (074) from the electron density maxima, which in roost cases are a sbort distance from (074). KAKU.I. I—Bams, K. & GUM, P. (1964). Ada Cryst. 17,496. WOOTTON, J. C, ARTHAN, N. & ALEXANDER, J. C (1962). /. Assoc. Offe. Agr. Chemists, 45,739. raaJe, I. L.& KAKLE, 1. (1963). Aaa Cryst. IS, 969. WOOTTON, J. C & COURCHESE, W. L. (1964). X Agric. JCA»LE, J. & KARLE, I. L. (1966), Ada Cryst. 21,849. JUMACHANDRAN, G. N. & SRINTVASAN, R. (1959). Ada Food Chan. 12, 94. WOOTTON, J. C (1966). Unpublished results. Crysl. i �336 S T S INCOBPOBATED, SCIENTIFIC TRANSLATION SERVICE Ann Arbor, Mich. CLINICAL PICTUBE AND ETIOLOGY OF CHLORACNE By K. H. Scultz, University Dermatology Clinic, Hamburg-Eppendorf Chloracne is the name for forms of occupational acne which develop as a result of intoxication with certain chlorinated aromatic compounds. The name dates back to Herxheimer who described the first case in 1899 and still assumed that in analogy to bromine and iodine, the acneifonn eruptions are the result of free chlorine as the etiological toxin. This view proved to be incorrect. The proposed designation "perna disease" of Wauer, Teleky and others based on the finding that this clinical picture occurred more frequently under the influence of perchlorinated naphthalenes also does not go to the heart of the matter, since other chlorinated aromatics in addition to chlorinated naphthalenes are also etiologically important. The clinical picture of symptoms primarily affects the skin. Beyond this, internal organs may be affected and nervous system and emotional disorders may appear. The skin symptoms are in the regions of the follicles. Comedones, resulting from a follicular hyperkeratosis, predominate and frequently are so numerous that hardly a single follicle remains untouched and the affected region of the skin obtains a dirty-gray appearance. In addition, at the peak of the disease, fairly large sebaceous cysts, inflammatory nodules, pustules and furuncles appear and in some of the patients, large spots or patches of pigmentation appear in regions exposed to light. Preferential sites are the face as well as the exposed areas of the throat and neck. Frequently, the external ear, especially, the ear lobes, are involved where small cysts can be easily palpated. In more pronounced forms, changes can also be found on the back, chest and extremities and in males, on the genitalia. Hands and feet usually are not involved. It is not rare that the symptoms of acne are preceded by a dermatitis with erythema and edema. In this phase of the condition, photosensitivHy frequently exists which evidently contributes to the development of dermatitis and the mentioned pigmentations (S. Braun, Grimmer). Generally, a differential diagnosis is not particularly difficult. The primary problem is to define the condition compared to other forms of occupational acne and acne vulgaris, which is generally possible with consideration of the clinical aspect, localization and especially the patient's history. Acneiform dermatoses caused by tars, pitch and mineral oils are found primarily on the extremities and trunk, while the face is more rarely involved. The predominance of inflammatory changes, such as folliculitis and furuncles in oil and tar acne and of comedones in chloracne are other characteristic features. Drug, caused aeneiform exanthemas due to iodine, bromine or cortisone also have a picture differing from cnloraene. The course is eminently chronic. In spite of intensive local and general therapy, recidivism may occur even years after the elimination of the causal toxins. Healing frequently takes place with extensive pitted, permanently disfiguring cicatrization (Schmidt and Boslet). The skin is not the only indicative region of intoxication with chloracnecausing substances. Damage of internal organs is not rare, with the liver being in the foreground. Several authors have reported on grave damage of hepatic parenchyma accompanied by icterus and functional disorders, including a number of fatal cases of acute atrophy of the liver (see reviews of W. Braun and A. Risse-Sundermann). The pronounced liver-toxocity of chloracne-causing substances was also confirmed in animal experiments (Bennett, Drinker and Warren ; Hofmann, Oettel; Schulz). In addition to liver damage, changes in the kidneys, pancreas, gastrointe*tinal tract and myocardium can also be observed, although much more rarely. Nervous system and psychological disorders were found primarily among workers occupied in the production and processing of chlorinated phenols (Trubant et al.). General fatigue, weakness of the legs, headache, attacks of vertigo, paresthesias, muscle pain, tendency to orthostatic collapse, local paresis and disturbed sensibility, anomalies in reflexes as well as an autonomir syndrome with lowered drive, depression, reduced power of recall and concea- 337 tration, disturbed sleep, irritability, loss of appetite, reduced libido and impotence have been reported as the most frequent neurological and psychopathological symptoms of intoxication which become manifest often only several months after it occurs (Bauer, Schulz and Spiegelberg). With regard to the question of distinguishing the latter from psychoneurotic obsessions (wish for compensation), reference is made to the discussions of Spiegelberg. Etiology.—When we review the literature, we find that periods of greater incidence of chloracne have existed in the last 60 years, which can be correlated with industrial development W. Braun has described these relationships in his monograph. The first cases were observed near the turn of the century when chlorine and hydrochloric acid began to be produced by the electrolytic route. At that time, the condition occurs primarily among workers having the assignment to clean the so-called hydrochloric acid towers, but only when tar was used as the protective coating of the walls. Although the causal toxin could not be determined at the time, it can be assumed on the basis of our present knowledge that the reaction products of chlorine and aromatic components of the tar must be considered as etiological factors of this condition. The next period of increased chloracne frqucy coincides with the introduction of so-called halogenated waxes. These are mixtures of highly chlorinated naphthalenes and diphenyls with a waxy consistency and a number of valuable properties. They are water-repellent, nonflammable, resistant to acids, are a good dielectric and are not pest-promoting. The halogenated waxes developed during the first world war at that time were used primarily for the manufacture of gas masks. Numerous cases of chloracne occurred in the manufacturing plants. In the middle twenties, these halogenated waxes were used in the mining industry as a water-repellent and nonflammable insulation for detonators. The high incidence of diseases observed in detonator manufacturing plants has been described by Teleky. The next massive occurrence is related with the rise of the electrical and radio industry. Chlorinated naphthalenes and diphenyls were in more widespread use for the insulation of wires and condensors at the start of the thirties. Several hundred cases including one fatality with liver atrophy became known especially in the United States. With the entry of the United States in the second world war, the field of application of these materials expanded also into ship building; this had the following reason: It was found that the halogenated waxes were in the position to insulate ships from the dangerous weapon of German magnetic mines. Consequently, large quantities of these materials were used in American shipyards for the impregnation of ship hulls. Mass incidences of chloracne with several fatalities were the results. In spite of all negative experiences, the use of chlorinated naphthalene waxes did not stop after the last war. Chloracne cases of greater or lesser frequency were repeatedly observed .in the electroteehnical and cable industry (Braun, Grimmer, Risse-Sundermann). From the pathogenic aspect, it is of interest to note an observation of Herzberg of 7 patients who developed intestinal symptoms and acneifonn dermatoses following the use of industrial chlorinated greases for frying. The question of the relationships between chemical structure and acneproducing effects of chlorinated naphthalenes is the subject of several experimental studies. Teleky as well as Drinker and Warren arrived at the conclusion about thirty years ago that the toxicity of the molecules increases with an increasing number of chlorine atoms on the ring. Later experiments conducted by Sehley and Kligman with human subjects and with the use of several chlorinated naphthalenes showed that penta- and hexachloronapbthalenes produced the strongest effects; compounds with 1 to 3 as well as 7 and 8 chlorine atoms were far less toxic of inactive We confirmed this finding in animal experiments using rabbit ears (Schulz 1965). In the last 10-15 years, halogenated waxes have become less important as etiological factors of chloracne. Evidently, this is related with the fact that they are no longer as important industrially and have been replaced by synthetics of the most diverse nature in most fields of application. In the fifties, the incidence of chloracne was observed in entirely different lectors of industry i.e. in the production and processing of chlorinated phenols. Reports of group involvements have been published from at least three �339 338 / industrial plants in "Western Germany. Baader and Bauer as well as Brinkman described 17 workers of a plant in Nordrhein-Westfalen who developed the typical skin symptoms as well as damage of the internal organs and central nervous system disorders in the production of pentachlorophenol. A larger number (about 60 cases) of similar disorders were recorded several years later in the region of southwest Germany among workers occupied in the production of of trichlorophenol (Hergt, Oettel, Hermann). Approximately at the same time, 31 workers in a Hamburg plant became ill after working with industrial 2,4,5-trichlorophenol, an intermediate of the synthensis of trichlorophenoxyacetic acid, a weed killer. Siliar high incidences of the disease occurred a few years ago in chemical plants of the Netherlands and the U.S. during analogous production processes. The high frequency of cases in Hamburg led to studies of the etiology. They were conducted by us together with Dr. Sorge, the former manager of the chlorophenol plant. Rabbit ears were used as the biological substrate on which symptoms corresponding to human chloracne can be produced by local painting as demon, strated by Hofmann and Neumann with chloronaphthalenes. The results, which have been reported earlier (Schulz 1957; Kimmig and Schulz 1957), can be briefly summarized as follows: First, it was found that it was not possible to produce changes in the rabbit ear in the form of chloracne with the use of the chemically pure compound in contrast to the technical grade of 2,4,5-trichlorophenol used in the plant Pure 1,2,4,5-tetrachlorobenzene also was inactive. The toxic factor therefore muat have' formed as a byproduct during the alkaline hydrolysis of tetrachlorobenzene into trichlorophenol. OH + NaOH + CHsOH 1,2,4-tetraehlorobenzene 2,4.5-triehlorophenol Alkaline hydrolysis of 1,2,4,5-tetrachlorobenzene into 2,4,5-trichlorophenol. Since the isolation of well-defined compounds from the distillation residue of trichlorophenol was unsuccessful, a number of especially synthesized substances were investigated which might have formed as a byproduct of the cited saponification process on the basis of theoretical considerations. The majority of investigated compounds proved to be inactive. Only dibenzofnrant with 3 and 4 chlorine atoms (diphenylene oxides) and 2,3,6,7-tetrachloro. dibenzodioxine (tetrachlorodiphenylene dioxide) led to the characteristic changes on the rabbit ear already in low concentrations. Moreover, it was demonstrated that 2,3,6,7-tetrachlorodibenzodioxine had formed by the following reaction route in the industrial process of alkaline hydrolysis of 1,2,4,5-tetrachlorodibenzene. Under the conditions of a salt fusion in a solvent-free state, 2 molecules of sodium trichlorophenolate form 1 molecule 2,3,6,7-tetrachlorodibenzodioxine with the elimination of 2 molecules of NaCL. Dr. Sorge synthesized the compound and in addition, isolated it from the distillation residue of industrial trichlorophenol. Animal experiments conducted with tetrachlorodibenzofuran and 93,6,7-tetrachlorodibenzodioxine showed an extremely high toxicity of these compounds. Even concentrations of 0.001-0.005% of tetrachlorpdibenzodioxine led to severe reactions on the rabbit ear after local application. On human skin in a self-experiment, two applications of 10 7 of the substance produced the symptoms characteristic of chloracne. On the rabbit ear, tetrachlorodibenzofuran showed an activity which was about 10 to 20 times less pronounced- Moreover, the unexpectedly high hepatotoxic action is worthy of note, particularly after tetrachlorodibenzodioxine. Single oral doses of 20-50 y/kg body weight regularly produced lethal liver necrosis, while doses of 10 y/kg were lethal for about 50% of the rabbits. On the basis of these chemical and toxicological findings, it is justified to conclude that 2,3,6,7-tetrachlorodibenzodioxine played an important role in the etiology of the cases of chloracne which occurred during the industrial production of trichlorophenol. It cannot be ruled out, however, that other, as yet unknown chlorinated aromatics of highly toxic properties may form during the industrial process under certain conditions. The results of the study are an example that materials which form only in small amounts as byproducts of Urge-scale syntheses can be of importance in occupational medicine. If such toxic byproducts can be uncovered and their mechanism of formation can be elucidated, this will create an important prerequisite for successful prophylaxis. In our special case, the plant succeeded in avoiding the formation of bignly toxic byproducts by modifying the production process. Our animal experiments were extended to other chlorinated aromatics to which other authors ascribed a chloracne-causing action on the basis of clinical observations. (Reviews of these compounds in the monograph of \f. Brann.) Neither benzenes and phenols with 1 to 6 chlorine atoms nor chlorinated diphenylethers produced an effect in animal experiments; It seems indicated to assume, therefore, that neglected toxic byproducts were of decisive geological importance in these cases rather than the main products. In connection with the acne produced by chloronaphthalenes, the question trose whether toxic byproducts rather than the chloronaphthalenes themselves might not be considered as the true toxins (Oettel). In the production of Industrial naphthalene by fractional distillation of tar, the presence of other aromatic compounds deriving from the tar apparently cannot be ruled out In the following chlorination process, such substances might then also undergo (•^substitution. These question prompted us to carry out animal experiments on rabbit ears using chemically pure chloronaphthalenes of different degrees of chlorination specially synthesized for this purpose.* In agreement with the findings of Shelley and Kligman, we found that naphthalenes containing 5-6 chlorine atoms have a chloraene-producing effect The necessary concentrations, however, were about 100 times higher than those of tetrachlorodibenzofuran (diphenylene oxide) about about 1000 times higher than for 2,3,6,7-tetrachlorodibenzodioxine (tetrachlorodiphenylene dioxide). On the basis of the present state of the art, therefore, the chloracne-productng activity of the following compounds appears to be sufficiently demonstrated or at least highly probable: Sodium salt of 2,4,5-trichlorophenol Saphthalenes containing 5-6 chlorine atoms: 2NaCl 2,3,6,7-tetrachlorodibenzodioxine (2,3,6,7-tetrachlorodiphenylene dioxide) �341 340 Dibenzofurans (diphenylene oxides) with higher degrees of chlorination: Trubaut, R., G. Vitte and E. Broussomart: Research on the toxicology of pentachlorophenoL Arch. MaL Prof. IS, 561 (1952) *>*WHUSJ 01 *° cMorinated hydrocarbons ZbL f. Cl 2,3,6,7,-tetrachlorodibenzodioxine: Boehringer Sohn - "'•-;.- In conclusion, it should be noted that the toxicodennatosls represented by chloracne results from intoxication with certain chlorinated aromatics. -The causally responsible compounds partly involve highly toxic substances, which can cause damage in various internal organs, especially the liver and hervbns system, in addition to the skin. Since the skin does not always represent 'the only manifestation .site, it' is recommended that thorough internal and neurological as well as psychiatric examinations be made in cases of suspected chloracne. UTEBATUBE Baader, E. W. and H. J. Bauer: Industrial intoxication due to pentachloi*. phenoL Ind. Med. Snrg. 20, 286 (1951). Bauer, H., K H. Schulz and U. Spiegelberg: Occupational intoxicatloa. during the production of chlorophenol compounds. Arch, f. Gewerbepath. und Gewerbehyg. 18, 538 (1961). Bennett, G. A., C. K Drinker and M. F. Warren: Morphological changes !• the livers of rats resulting from exposure to certain chlorinated hydrocarbont J. Industr. Hyg. a. ToxlcoL 20, 97 (1938). »u Braun, W.: Chloracne. Monograph Supplements to the journal Berufsdennatosen, Vol. 1, Editio Cantor, Aulendorf/Wurtt « Drinker, C., M. F. Warren and G. A. Bennett: The problem of possible gjitemic effects from certain chlorinated hydrocarbons. J. of industr. Hyg.'••MI ToxicoL 19, 283 (1937). ^? Grimmer, H.: Occupational acne by chlorinated aromatic hydrocarbon (chloracne, perna disease). ZbL f. Arbeitsmed. 5, 76 (1955). t;V Hergt, W.: Comment in discussion Occupational Physicians' Conference Bad' Durkheim, 1955. ^* Herzberg, J. J.: Chloracne following consumption of chlorinated paraflB. Dermat Wschr. 119, 425 (1947). Herxheimer, K.: On chloracne. Munch, med. Wschr. 278 (1899). Hofmann, H. Th.: Paper before the Occupational Physicians' Confe: Bad Durkheim, 1955. Hofmann, H. Th. and W. Neumann: A method for the animal-experiment^ study of the dennatological effect of chlorinated naphthalenes. ZbL Arbeitsmei.' 2, 169 (1952). —"»-; Kimmig, J. and K. H. Schulz: Chlorinated aromatic cyclic ethers as a < of so-called chloracne. Naturwiss. 44, 337 (1957). Kimmig, 3. and K. H. Schulz: Occupational acne (so-called chloracne) W chlorinated aromatic cyclic ethers. Dermatologica (Basel) 115, 540 (1957). ,«r I Oettel, H.: Clinical and animal-experimental experiences with highly torfc chlorinated hydrocarbons; a contribution to the perna problem. Paper befiat the Occupational Physicians Conference, Bad Diirkheim, 1955. Risse-Sundennann, A.: Intoxications by chlorinated aromatics. Monograph. Cologne University. Dissertation, Cologne, 1959. Schulz, K. H.: Clinical and experimental studies on the etiology of acne. Arch. Klin, exper. Dermat 206, 589 (1957). Schulz, K. H.: Unpublished experiments, 1965. Schmidt, W. and W. Boslet: Contribution to a knowledge of permanent i changes in chloracne patients with demonstrable insurance claims, matosen 4, 109 (1956). Shelley, W. B. and A. M. KUgman: The experimental production of acne k* penta-and hexachloronaphthalenes. Arch, of Dermat (Chicago) 75, 689 (19S7L, , Spiegelberg, U.: On the question of delayed and permanent psychopatt»f logical damage following occupational intoxications. Med. Klinik 56, 436 (WOkl Teleky, It: Perna disease (chloracne). Klin. Wschr. 845 (1927)- mi- J Wschr. 897 (1927); Klin. Wschr. 214 (1928). REPOBT ow METHODOLOOT FOB CHLOBDTATED AEOMATICS DT TATS OILS, AND FATTY ACIDS ' By 3 S G B - ' HIGGINBOTHAM, and DAVID FIRESTONE (Divif°%**^ ' don of Food Chemistry and Technology, Bureau of Science, Food and Drug Atainistration, Consumer Protection and Environmental Health Service K^oST*' ^^^^ Ot Health' Education, and ABSTKACT The official, first action electron capture GLC (EC-GLC) method for chick edema factor (polychlorodibenzo-p^loxlns) have been reviewed. This general procedure, which underwent collaborative study in 1967, has undergone several ainor modifications which result in better recoveries of ^SoatbSp^ ^^ 2£d £Sr^sed sP^^city in interpretation of the gas chromatoCTaphic results. The EC-GLC method can be used as a screening tlst, or where? topieflpatternrf GLC peaks is obtained as a preliminary tes^ but confirma^ tats are needed to demonstrate structure and toxicity of polychlorodibeiS florins. Preliminary work with combined GLC-Mass Spectrometry in^ca^ ttat ttis technque might provide a suitable test, if adequate sample cleanup ,*? be accomplished. A chicken embryo assay has been developed to thelpotat rtere toxicity can be observed in three to five days after injectionof eggT ^preliminary procedure has been developed- for isolation and gas chromatography of chlorophenols in fats and fatty acids. Polychlorophenols have ^bera fonnd to be precursors of chlorodibenzo-p-dioxins. The use of a non-sne^c ^*f°^2,°^S? tCSv f0^ ^oroP^enols employing the BacOlus megateriuniw&s tnluated. Chloropheno s were found to produce uniformly graded growth inh^ MUon of the test organism in the range 1-100 jig. : pewidespread use of toxic organochlorine compounds in agriculture and todustry requires development of sensitive methods for their detection in a wide variety of commodities. In addition, it is equally important that methods It developed to detect toxic breakdown or conversion products of organochLorine compounds. One of the most urgent needs in the fat and oU industry is for , rapid and^specific method for polychlorodibenzo-p-dioxins (chick edema factH) in fats, oils, and fatty acids. The official, first action, microcoulometric tod electron capture methods for chick edema factors (CEF) are essentiallv .seening procedures (1,2,3). Both methods, at present, require "rather time «nsuming three-week chick bioassay (4) for confirmation. : ; The purpose of this report is to review the current status of chemical and iWogical methods for chlorophenols and chlorinated dibenzo-p-dioxins in fats tQg, and fatty acids. *«"«» CEF consists of a mixture of chlorinated dibenzo-p-dioxins which occur oeoa itonally as a trace contaminant in fats. Recently, a communication (5) from *" "SUES!?7 S:p?t£AJ?* results of a Preliminary study which demonstrated •••JfffS?^ tha^ CEF conM arise from residues of pentachloro^enS and W4,6-tetrachlorophenol in fats and fatty acids. Chlorophenols and their sa^te •ten heated, undergo condensation reactions and form chlorinated derivatives jfdibenzo-p-dioxin. The following equation Ulustrates ttds cWdWtton w . contain ca 10% of 2,3,4,6-tetrachIorowhich also undergoes thermal condensation reactions and forms hex£ derivatives of dibenzo-p-dioxins. The condensation of 234*6pentecnloro heno1 P *orn« ^o heptachloro derivatives . * capture G L h been developed f r pentachlorophenol fats, §•d 2A4,«-tetrachlorophenolCinmethodoils,a sand fatty adds (6oHowever, �342 343 cries of the two volatile polyehlorophenols were low and varied over a wifc range; nevertheless, the method appears to be satisfactory for qnalit measurements at the 0.5 ppm level. Several samples of oleic add k contain C.&F were analyzed and were found to be contaminated with res of pentachlorophenoL The method requires further study. .; The official electron capture and microcoulometric methods for CEP developed before their chemical structures were known. -The methods screening procedures and are based on the observation that toxic fats < a number of chlorinated components (now known to be polychlorodi^, dioxins) which have greater retention times than chlorinated pesticide* electron capture method has received wide acceptance. It is approi 2000 times as sensitive and requires less sample than the microcoulu<in method. In addition, electron capture gas chromatographic equipment is pier and is in general use in many laboratories. Recently, it has come to our attention that a number of laboratories routinely use the electron capture method for control work are not aware rf improvements (3) that have been made in the original procedure (2). Receftl changes which have not been published include a minor modification of HjSOj cleanup step and a slight modification of the procedure for packing tfct 4 GLC column. The modified method, which includes these changes, replaces d (< existing GLC methods for the chemical assay of CBF. Spirent, add 5 g anhydrous NaiSO*. Drain excess petroleum ether so that It is * t above upper surface of Na,S<X ) Gas chromatographic column.—Glas»i 6-7' long x «" SE 52 siliconegum rubber on 60/80 nwish Gas Chrom Q Coat we£* 2.5^ of? ^ 16801)' rubber stationary *& uand dissolveoin ° ^ ^ •n««rt phase a Weigh 1 80 the silicone gum : m ene i ^ cMoride-toluene, heating to dissot^ AddOT5e of a 1<1Uid mlxture and let stand 1 0 , " ^TwlS diromatograS i ? stirring. n £ to rotary evaporator. Apply vacuum to the ToSn^ Dry column and pack the coated material into the column by addtaV small ) Gas chromatograPh with electron solution) at a sensitivity setting detector.-A tritium source AFS EM METHOD ••-m Reagents and Apparatus Rinse all glassware with appropriate solvents before use. Do not use pd*. thylene containers to store solvents. (1) Concentrated HjSO*.—Reagent grade. (2) Petroleum Ether.—Reagent grade, redistilled to glass between 30 i 60°C (Available from Burdick and Jackson Laboratories, 1953 S. Harvey i Muskegon, Mich. 49442). (3) CCU.—Distilled to glass. (Available from Burdick and Jackson Labc tories, Muskegon, Mich.) (4) Anhydrous Na»SO*.—Analytical reagent grade. (5) Ethyl ether.—Analytical reagent grade (not >2% alcohol) or ether (not >0.01% alcohol). •••HSSSS (6) Iso-octane.—Distilled to glass (Available from Burdick and .JacbHp Laboratories, Muskegon, Mich.) •'-*%%£** (7) Standard- aldrin solution.—Dissolve aldrin to iso-octane to make QM fig/ml solution. (8) Chick edema factor low positive reference sampla—1.5% reference ta fat ta USP cottonseen oil or other suitable vegetable oiL (Prepare from ence toxic fat available from the Division of Pesticides, Bureau of Food and Drug Administration, Washington, D.C. 20204). (9) Activated AUO, (Fisher No. A-540, do not substitute).—Activate m" portions by heating 4 hours at 260°C. Transfer without cooling to dry «• tainer and close tightly. Check activity of AUO> by analysis of the low pa reference sample, examining AUO* fractions 2 and 3. With sufficiently vated AljOj, chick edema factor elutes predominantly or entirely to AUO* tion 3 as indicated by the gas chromatograms. (Chromatogram should series of GLC peaks with Ra values between ca 8 and 45). ..,,, (10) AUO* chromatographic column.—To a dry chromatographic column, Jf nun o.d. x 250 mm long, fitted at bottom with coarse porosity fritted glass <H* and Teflon stopcock (a column without the fritted disk but holding a " wool plug to the bottom may be used), add redistilled petroleum ether, prior to use with anhydrous Na*SO<, until column is % full. Weigh 15 g and transfer to column to small portions, tapping the column as AUO* settfel When last portion of AUO* settles and air bubbles stop rising to surface --* , ,„ ..-.feu J.« £ AlJJfc- DETEBMINATIOW 1>5% 6* •;'«.—*>-.;- V ^fcrence toxic fat to USP cottonseed o |(tf the 1.5% reference toxic fat to 10 ml of CCU to a 500 ml glass t IffiTf ?Jt^ %??** With detem**tion as described Tel!w to 25 „&?! %^3^£^>f^.***»••>* **** calibrated gas chromatograph. The resulting ga I^tt>StpaXi?t°R?I£i?ak8 1tt ^ ? 8^5' ^Peldtog on operating condi' f^« 17 o^ dne3 are due to hexachlorodibenzo-p-dioxto isomers 2 ^ to tte 2 ?-f Ra 35-45 is *,™ to octachlorodibenzo-p-dioxto. lrul"JUU lsomeis, anaaa ^P^^orodibeiizo-p-diSnlsomeTMd tat »^^^ due H Preliminarv fsnlfniH/. O«M „!«».,— -rS—;±^X... CCU 30 125 ot wM,m .tier, stopS �344 345 (d) Sulfuric acid cleanup of alumina fraction S.—Add 2 ml of concentrate I HiSO< to graduated cylinder containing 3 ml petroleum ether solution ft * (c), stopper and shake vigorously for 30 sec. Allow layers to separate i decant petroleum ether layer into 10 ml beaker avoiding transfer of HJ_ layer. Add 2 ml petroleum ether to cylinder, swirl vigorously, allow layers 1 separate, and decant petroleum ether layer into beaker. Add ca 0.5 g of i naHCOi to beaker and stir ca % min. Let stand five minutes and dt petroleum ether layer into clean 2 or 4 dram vial. Wash NaHCOi with 2'j petroleum ether and decant washing into vial. Evaporate solvent under N. (c) Electron capture gas chromatography of petroleum ether extract;— _ up residue in 250 pi iso-octane (redistilled in glass), stopper vial tightly i rotate so that solvent wets sides of vial. Inject 1 microliter of sample sol (equivalent to 10 mg fat) into calibrated gas chromatograph. Gas chroi graphic peaks with Ra 8-45 are indicative of the presence of chick a factor. Compare Ra values of sample peaks with Ra values of peaks from hfc'f erence toxic fat See (a) for identification of peaks. If peaks indicattve'Wi chick edema factor are not observed, inject 5 ^1 of sample solution (eqnlvale* to 50 mg fat) into gas chromatograph. Check reagents for possible Interftr. ences by running a blank with each set of samples. The chromatogram ttA the blank should show a smooth low baseline from Ra 8—Ra 45. (Types of sa*.' pies found by experience to be generally free of components characteristic rf toxic fats may be examined by initial injection of 5 pi of solution.) ; • r "* standards and an extract from a 2.5 g sample (positive for chick |(dema factor by EC-GLC). The GLC oven temperature was 220°C with a leUnm flow at 75 ml/minute. Injection temperature was 235°C and silicone : .embrane temperature was about 150'C. MS sensitivity setting was 32 x at 40 ^ •V'l: 1 DISCUSSION Analysis of the low positive reference sample serves as an overall check • instrument performance and sample cleanup. The chromatogram from the'l« positive reference fat should show a distinct peak pattern as Illustrated I Figure 1. The lower chromatogram (B) represents an injection equivalent 5 50 mg of the original low positive reference fat Chromatogram (A) rtn! sents a mixture of synthetic polychlorodibenzo-p-dioxins prepared by pyrotnL-of 2,3,4,6-tetrachlorophenol and pentachlorophenol. As stated previously pate 1 through 4 are due to four positional isomers of hexachlorodibenzo-D-dlarX 1 The isomer associated with the small shoulder (peak 3) is probably causetfp ' the presence of a tetrachlorophenol other than the 2,3,4,6-isomer in the start* material. Peaks 5 and 6 are due to two positional isomers of heptacbkSI dibenzo-p-dioxin. Peak 7 is due to octachlorodibenzo-p-dioxin. ™»>3 Aldrin is used to calibrate the instrument sensitivity for chick edema analyses. The similarity of detector response vs. applied voltage for aldrln «S an extract from the 1.5% reference toxic fat (low positive reference is illustrated in Figure 2. CHEMICAL AND BIOLOGICAL CONFIRMATORY METHODS The need for rapid chemical and biological confirmatory tests has led investigation of mass spectrometry as well as two biological toxicity aan Preliminary work has suggested that rapid EC-GLC screening for chick edou factors can be carried out initially; if the presence of chick edema factoafJL. indicated, then larger portions of sample would be fractionated and cleaned i*! for chemical and biological confirmation. ^s MASS SPECTBOMETBY Results of a preliminary investigation of combined GLC-mass spectroaBiP (GLC-MS) indicated that the use of this technique might be suitable •*^!*1 quate sample cleanup can be accomplished. A 7-foot coiled glass column •' with 2.5% SE-52 on 60-80 mesh Gas Chrom Q was used with an Atla«~L mass spectrometer and single-stage Llewellyn (silicone membrane) separator! re*. ^ |, GLC retention times as well as molecular weight and number of chlorine Ittoms in the molecule were determined for a standard mixture and an extract tfcom the test sample. Two pi of lOpl solution from the test sample was tUjeeted; it was estimated that 2 p\ test sample contained 0.4 jig of hexa-, lepta- and octachlorodibenzo-p-dioxins in addition to other unidentified constit%ents. A. summary of results are in Table 1. Comparison of fragmentation S pattern and relative abundance of the ions from standards and sample might iftfford additional specificity; impurities in the test sample prevented such evalIUtion at this time. CHICKEN EMBETO ASSAY ?•^Extracts from a reference toxic fat and several test samples were subjected i|D the chicken embryo assay (7); 1 1 g samples of fat were fractionated 1 ^cording to AOAC (1965) 26.093-26.094, and alumina fraction 3, and cleaned w with sulfuric acid (JAOAC (Changes in Methods) 50, page 217, section (c) (1867)). Small portions of each extract was retained for EC-GLC analysis and tbe remainder, in chloroform solution, was subjected to the chicken embryo • assay (10-15 eggs per sample were tested by injection of portions of the .jample extract in the air cell) at three levels equivalent to ca 40, 30 and 10 g darting sample. Assay results are shown in Table 2. These results indicate ? that the chicken embryo test can provide a sensitive indication of toxicity as as a measure of specificity due to observations of localized and genralized a. In many instances evidence of toxicity can be observed (by periodic in 3-5 days. B. megaterium TOXICITY TEST The use of a non-specific biological test employing the Bacillus megaterium rts evaluated (8). This test involves observation of inhibition of a seeded etri dish holding a B. megaterium spore suspension in agar medium. Filter iptper discs of sample extracts are placed on the surface of the agar plates "and inhibition zones are observed after 18 hour incubation at 37° C. Rre samples (111 g each) were fractionated according to the general proeeare of AOAC (1965) 26.093-26.094. These test samples consisted of two toxic fats, one low toxic reference material (1.5% TEF In USP cottonoil), one nontoxic oil, and a reagent blank. In addition to the three alu. fractions' obtained with petroleum ether, 5% ethyl ether in petroleum . ether and 25% ethyl ether in petroleum ether, a fourth fraction eluted with ; 400 ml of 100% ethyl ether was obtained. The four alumina fractions were deaned up twice with sulfuric acid according to JAOAC (Changes in Methods) :», <X217 Sect C (1967). The residues of fractions 2,3, and 4 were spotted on flier discs (7.5 mm diameter) at two levels equivalent to 2.5 and 54 g starting ample. EC-GLC analysis of alumina extracts indicated that chick edema factors (polychlorodibenzo-p-dioxins are predominantly concentrated in alumina fraction 3. In addition, solvent blanks, a synthetic reference standard consistjja- of hexa-, hepta- and octachlorodibenzo-p-dioxin (CDPD), and a sample of Ironical grade 2,3,4,6-tetrachlorophenol (2,3,4,6-TCP) were spotted at the folt bring concentrations: 0.01 pg, 0.1 pg, 0.5 /.g, 10 pg, 10.0 pg, and 100 pg. ••: After incubation, inhibition was observed in six cases as shown in Table 3. £ibe sensitivity of the B. megaterium test for CEF appears to be limited to 100 . and even then only a very small zone of inhibition was noticed. It ftfpears, however, that a rapid confirmation test for chlorophenols can be ;*reloped at levels of ca 2-3 ppm or less. These are the low levels at which ' aiatoxin B» shows inhibition of B. megaterium. �346 ACKNOWLEDGMENT The authors wish to express their appreciation to Mr. Joseph Barandj, Drew Chemical Company, Boonton, N.J. and Dr. E. N. Gerhardt, Emery Indn». tries, Inc., Cincinnati, Ohio for suggesting modifications of the H2SO< cleanup procedure which have resulted in improved recoveries of chick edema factor. The contribution of the following members of the Food and Drug Admiafe. tration are gratefully acknowledged: to Dr. M. J. Verrett for the chickca embryo assays; to Joseph N. Damico and Bobert P. Barron for the mass spectrometric analyses: to Bobert M. Eppley for the S. megaterium tests; and Thomas J. Dols for his helpful discussions concerning GLC operating param*. ters. REFERENCES 1 Official Methods of Analysis, 10th ed., Association of Official Agricultural Chemiita. Washington, B.C., 1965, sees. 26.092-26.096. —«* '"Changes In Methods. 26. Oils, Fats, and Waxes," ibid. SO, 217-218 (1967). ' Neal, P., This Journal 50, 1338 (1967) ; "Changes In Methods. 26. Oils, Fate, tat Waxes," ibid. SI, 489-i»0 (1968). • Official Methods of Analysts, 10th ed., Association of Official Agricultural ChemiM* Washington, D.C., 1965, sees. 26.087-26.091. * * Hlggtobotham, G. H., Huang, A., Firestone, D., Verrett, J., Ress, J., and CauDbdL w > A. D., Nature 22tf, 702 (1968). * « Higginbotham, G. K., Ress, J., and Eocke, A., JAOAC, In press. 1 Verrett, M. J., Marliac, 3. P., and McLaughlln, J. ibii. 47,1003 (1964). 'Clements, N. L., ibid. SI, 611 (1968). TABLE 1.—MASS SPECTROMETRIC ANALYSIS Of ISOLATED COMPONENTS GLC peak no. Test sample °8S f 422 388 422 1 Hexachlorodibenzo-p-dioxin 3 Heptachlorodibenzo-p-dioxin 1 Hexachlorodibenzo-p-dioxin 3 Heptachlorodibeiizo-p-dioxin Molecular weight found 388 388 Identity • * I J TABLE 2.—CHICKEN EMBRYO ASSAY Of EXTRACTS EC-GLC analysis for chick edema factor Sample Reference toxic fat Test fat No. ! „ „ Test fat No. 2. Test fat No. 3 Reagent blank... Chloroform solvent Control eggs — _ positive do. do do negative.. '.do do Estimated level of hexa-, heptaPercent and octachloromortality dibenzo-p-dioxchicken ins in fat, in ppm embryo assay Assay obsemtini 2.4 100 edema obs*mt 0.7 06 100 Do' 93 Do' 30 ... 20 7 —— TABLE 3.-RESULTS FROM B. MEGATERIUM TEST Sample Nontoxic USP cottonseed oil Fr. 4(2.5 g extract) TEF-F797,- Fr. 2, (2^ g extract) „ 100 <ig synthetic (CDPD) reference standard Ug2,3,4,6-TCP 10Mg2,3.4,fr-TCP 100;>g2,3,4,6-TCP.- Observation Barley visible around disc. Do. Do. Do. 16 mm inhibition zone. 36 mm inhibition zone. �349 348 SECTION 121.1070 The food additive fatty acids may be safely used in food and in the manufacture of food components in accordance with the following prescribed conditions: (a) The food additive consists of one or any mixture of the following ttraigbt-cbain monobasic carboxylic acids and their associated fatty acids manufactured from fats and oils derived from edible sources: Capric acid, caprylic acid, lanric acid, myristic acid, oleic acid, palmitic acid, and stearic acid. (8) The food additive meets the following specifications: (1) TJnsaponifiable matter does not exceed 2 percent (2) It is free of chick-edema factor or other factors toxic to chicks, as evidenced during the bioassay method for determining the chick-edema factor as prescribed in paragraph (c) (2) of this section. (c) For the purposes of this section: (1) Unsaponifiable matter shall be determined by the method described in section 26.049 of the Official Methods of Analysis of the Association of Official Agricultural Chemists, Ninth Edition (1960). (2) Chick-edema factor shall be determined by the bioassay method described in the Journal of the Association of Official Agricultural Chemists, Volume 44, page 146 (1961). The presence of chick-edema factor shall be determined by a comparison between the mean log of the pericardial fluid volumes of a test group and of a concurrent negative control group. The significance of the difference in pericardial fluid volumes between the test group and the negative control group is determined by calculating a * value according to the formula: The test sample is judged to contain chick-edema factor if the calcu- 100 E e •*~s U 00 z O 0. CO l.J FATTY ACIDS . xt-£. J°l+*J. \n,n. GO where: 2 1 and £e are the means of the log of the pericardial fluid volumes of the test and control groups, respectively; n< and n, are the number of chicks in the respective groups; t? and ««* are the variances of the test and control groups, respectively. The variances are calculated as follows: cr CHICK EDEMA FACTOK n(n-l) where: is the sum of the logs of the pericardial fluid volumes; 2z> is the sum of the squares of the log of the pericardial fluid volumes for either the test t or control c group date. 20 20 GO VOLTS 100 toted t exceeds 1.3 and the mean log of the pericardial fluid volume obtained from the negative control group multiplied by 100 is less than 1.146L (3) "Other factors toxic to chicks" referred to in paragraph (b) (2) of this lection shall be determined during the course of the bioassay test described in fobparagraph (2) of this paragraph on the basis of chick deaths or other abnormalities not attributable to chick-edema factor or to the experimental conditions of the test (d) It is used or intended for use as follows : (1) In foods as a lubricant binder and as a defoaming agent in accordance with good manufacturing practice. (2) As a component in the manufacture of other food-grade additives. (c) To assure safe use of the additive the label and labeling of the additive and any premix thereof shall bear in addition to the other information required by the act the following: (1) The common or usual name of the acid or acids contained therein. (2) The words "food grade," in juxtaposition with and equally as prominent as the name of the acid. �350 SECTION 121.1071 SALTS OF FATTY ACIDS The food additive salts of fatty acids/may be safely used in food and in the manufacture of food components in accordance with the following prescribed conditions: (a) The additive consists of one or any mixture of two or more of the aluminum, calcium, magnesium, potassium, and sodium salts of the fatty addt conforming to f 121.1070. (6) The food additive is used or intended for use as a binder, emulstfler, and anticaking agent in food in accordance with good manufacturing practice. (c) To assure safe use of the additive, the label and labeling of the addiUre and any premix thereof shall bear in addition to the other information required by the act the following: (1) The common or usual name of the fatty acid salt or salts contained therein. (2) The words "food grade," in juxtaposition with and equally as prominent as the name of the salt. Title 21—FOOD AND DRUGS Chapter 1—Food and Drug Administration, Department of Health, Education, and Welfare SUBCHAPTEB B FOOD AND FOOD PRODUCTS PART 121—FOOD ADDITIVES Subpart D—Food Additives Permitted in Food for Human Consumption FATTY Acros The Commissioner of Food and Drugs has received a petition (FAP 6A2003) from Fatty Acid Producers' Council. Division of the Soap and Detergent.'- • ciation, 295 Madison Avenue, New York, N.Y. 10017, proposing that § the food additive regulation providing for safe use of fatty acids in food and in the manufacture of food components, be amended: A. To provide for the use of a screening method for determining the pretence of chick-edema factor in the fatty acids that, within certain condition*, may be used in lieu of the bioassay method prescribed by paragraph (c) (2), and B. To delete references to "other factors toxic to chicks" from the section. From the available information it can be concluded that the anomalies presently identified as due to other toxic factors, which may be evidenced during the bioassay method for determining chick-edema factor, are directly associated with the same conditions or substances producing chick-edema factor, and the proposed physicochemical method is adequate as a screening test tar detecting the chick-edema factor complex of toxicants. Based on the information sbnmitted in the petition, and other relevant material, the Commissioner has concluded that the regulation should be amended at petitioned. In addition, the references identifying the chick-edema bioasar procedure was updated to refer to the Official Methods of Analysis of toe Association of Official Agricultural Chemists. Therefore, pursuant to the provisions of the Federal Food, Drug, and (V*. metic Act (sec. 409(c) (1), 72 Stat 1786; 21 U.S.C. 348(c) (1)), and under the authority delegated to the Commissioner by the Secretary of Health, Education, and Welfare (21 CFB 2.120; 31 F.B. 3008), § 121.1070 (b) (2) and (e) (2) and (3) are amended to read as follows: 351 SECTION 121.1070 * * (b) It is free of chick-edema factor: * *** (2) FATTY ACIDS * * (1) As evidenced during the bioassay method for determining the chickedema factor as prescribed in paragraph (c) (2) of this section; or (ii) As evidenced by the absence of chromatographic peaks with a retention time relative to aldrin (RA) of five or more using the gas chromatographicmicrocoulometric method prescribed in paragraph (c) (3) of this section. If ehromatographic peaks are found with RA values of five or more, it shall meet the requirements of the bioassay method prescribed in paragraph (c) (2) of this section for determining chick-edema factor. (0 * * * (2) Chick-edema factor shall be determined by the bioassay method described in Official Methods of Analysis of the Association of Official Agricultural Chemists, 10th Edition (1965), sections 26.08T through 26.091. (3) The gas chromatographic-microcoulometric method for testing fatty adds for chick-edema shall be the method described in Official Methods of Analysis of the Association of Official Agricultural Chemists, 10th Edition (1965), sections 26.002 through 26.096, except that the following procedure is substituted for that described in section 26.092(b): Activated alumina.—(Fisher No. A540 or equivalent.) Activate 250-gram portions by heating 4 hours at 260°C. Transfer without cooling to dry container snd close tightly. Use within 1 week after preparation. Check activated Al>Oi by analysis of a reference standard by examining fractions 2 and 3. Chickedema factor should elute in AUO> fraction 3 as indicated by the gas cbromatopam. (A sample of the reference standard may be obtained on request from toe Bureau of Science, Food and Drug Administration, Washington, D.C. J0204.) * * * * » Any person who will be adversely affected by the foregoing order may at tor time within 30 days from the date of its publication in the FEDERAL REGISUS file with the Hearing Clerk, Department of Health, Education, and Welfrre, Boom 5440, 330 Independence Avenue SW., Washington, D.C. 20201, writtea objections thereto, preferably in quintuplicate. Objections shall show wherein the person filing will be adversely affected by the order and specify with particularity the provisions of the order deemed objectionable and the grounds for the objections. If a hearing is requested, the objections must state the issues for the hearing. A hearing will be granted if the objections are supported by grounds legally sufficient to justify the relief sought Objections may le accompanied by a memorandum or brief in support thereof. Effective date. This order shall become effective on the date of its publication in the FEDERAL REOISTEB. (Sec. 409(c) (1), 72 Stat 1786; 21 U.S.C. 348(c) (1)) Dated: August 18,1966. J. K. KIEK, Acting Commissioner of Food and Drugs. [F.R. Doc. 66-9263; Filed, Aug. 24,1966; 8:47 a.m.] �Table 12 EiT.bryotoxicity of Chlorophenola, DibanKO-p-dioxins (Chick cdeir.a) Estimated Dose Levels to Produce Indicated Mortality (Percent) VL Compound Phenols (Dose in mg) Mortality *^ 100 50 0 0.4 .-x 5x10-3 exK)"3 , ;,lxlo.2 0.05 IxlO"3 0.7 1.5 2,4-Dichlorophenoxyacetic 2,4,5-Trichlorophnnoxyacetic acid Eionetlcs Bow #120449 0.2 0.2 0.01 2.5xlO" 6 2.5x10-3 l.OxlO" 4 2.5x10-2 1.0 .1.0 0.25 SxlO" 6 0.005 2xlO"4 >0.5 2.5xUT 5 1.0 7wt.ehloroph.tol 0.4 0.01 l.SxlO'2 2xlO~ 3 5xlO"4 .05 1.2xlO"3 6xJ-0~4 • ^0.1 1.5 0.05 2,4-Dichlorophenol ~-2 2,4,5'Trichlorophanol 2,4,6-Trichlorophenol 2,3,4,6-Tetrachlorophenol 2.5 Uns^ponifiable* Fraction (Dose in mg) 100 50 0 Pyrolysis* Produces (Dose in mg) 50' 0 100 6xlo.3 , xlo - 3 ' 0.25 < .05 1.0 svO.5 1.0 Polyehlorodibenao-p-Dlo:cins * F'iberiso-p-dioxin Chlorinated diber.KO-n-dioxin Trichloro diber.zo-p-dioxin Tetrachloro dlbenzc.-p-dioxin ;Is::achloro isciucrs (4) a. Xost toxic . b. Least to:;ic HcyC£Chloro isontcrs(2) £. Most toxic b. Least to:ci« L'npi:3listed data of J. Vsrrecc * Xany of these ars cixtura* CoJ«: 1 raj - 20 ppn 1 pg - 20 ppb 0.5 1x10- " 2x10"* 5X10-2 2. 5x10'5 2xiO"3 2xlO'4 2.5x10-6 5x10-5 1.25xlO"4 2x10-3 SxlOJ4 3 1.25xW 2.5xlO' lxio'3 4 Air Call Injections Pre-incubation ttRY K'.SC&t C» YSHAXOtOClf SYUDIKS I'I'IH PICXIS KSI1JG CO / Avg. v:t. of latua i£S?.2«?;! Atioui':t Ivi£u!>ate<l r-rya S - 10 Fc.tt . litter Total i< of Fo.ti CoaCrol 0 8.6 43 41 Koac Chloi-odibanf-o-n- 9.1 ..g/kg/day 10.3 62 11 eyo anoKolioc 5 1 - 4 :JD, 47 ID 1.33 (21'.i trichloro, 2.0 S/bs/do.y 10.8 43 41 Snr,tro- 2 - 2 :•!> 1.47 0«5 '"g/1^;A.^y 10.5 63 62 hctEorrhage sa'ieroInteattMil i . 1 a) 1.66 0.13 ; s/3i3/dcy 11.6 58 5S Hoiic 2 - 2 CD 1.S3 r Bern t.livti ayntUcstcsa in I-DA) Tcrr.ta t Cord 2 - 1 ED, 1 LD 1 1.55 •D - Uarly rvat", ^ . Lctc F5LOH: r-r. T.F.X. Coliila3, t!r. W.H. Hcasea, I>r. C.H. Willlaasa Risittuo Tox'eoloKy Brcr.ch T'i.virlca o£ Pacelcido ^cr.-tstry t Tor.tcolosy t'fflcc of ?'.":ticldr:s 6 Prtciticfc �304 355 DEPAETMENT OF HEALTH, EDUCATION, AND WELFAEE, vJ Jl^ .; £i £1 jji c; i-. «w & • to > < O t-. •* ii i in «-< \Q ^* #-< tH • »-l « r-» • VO CM • • VO \a • »-i \Cj ' ^~ ?> -a 0 • »H h « a g 1-4 - 8 t-i S " i «! 9 9 0 i-4 « tv: a" a CM •-i Cl •* •» CF* \e st ** 8 . «'t O» c>i « <f I 1 I M ** C -* c,. • sg ?s f-4 t» f-l C» R eo o ca t: « c ts i 1-1 a • ^4 ,; ' t? A*1 £ CJ <!l «5 o 5-"i S C to « o p o « f: o ta i-i A ta -ri ^i rtt rt o Es o &s to e S ta ^' ^ C f! S « *-• § ^^ •«i O CO co ^ r~ iri O V£> f-r »-< 'i - Z "s " *!S«.! & C! t-i « ^.5 -3 ^ 2 * ^ >w W o i2 g ««-? n ^. O3 ^** O • •jj a^-& f -r^ s OiJr4 r: 14 *> « & » »J £i n 4J y r<t •-! •H i-* J< •-* 5 *5 y IA* 2 4J '^? « t=< •-" •-; «> v> t«co <t 03 «n m . <-i sf O «H H 0 f-i •r--| 0 .Jjl-w C'lu {<*!•••> ti -H sh . in • 5 f* 3 cJ S <a • •* » •w • t-l «-4 «-l c-i « o* M <r »-l '« * «•» • n o «-< t-l •CJ o • CJ o ri t-l •v* fJ It t^ ? JJ rt 3 ^—i > > re 4-1 ''- | H u *£. ; " » 1-4 t"l > y <« ^~ o fe O 5, > , « > > o >. c? (^ ts 4J • --l-O X i? 5. 5. ^ 5 . *^ m •~-.v3 g- o o k'~ ifo ••-.•<!• r o go ~-m >H <-< ts ta &. —. .'^^-^ ~~<7< o : c; t>3 ,«.-* •* ta ji /^. tj AJ ^, •v^<c c v-i i-l 5i .OH W ooi! t> 5; s-* ^ *J tM a go J5iJ 5) H x-v t> ^ "s?t-< a e K H -^ 3 »-< FJ »** v- rj ';•; Hi cl t-i o u 11 —I ; ! " '-< o m {> "i >» r' ort sr*<° ««a - '«-! -» o ;j ••) h -,J O =^10 r^ t^i o i-- rt 5 *J ^ <0 ^ ••: ~" >» so lJr-1 0 •••• cj "vj S "-.~; r. cv -•- « -o c-i ••• ^-"i »H tl fc t> to £ •3 K -Ij W :-< B. i r. m -•«•- JJ •' ,''^> c >cJ r-i "• a "i>,-: m u r'.i SO u «. ^ ""iJ 0 x-v -a x^ O x-% C •t4 1 O o «,4 x-x U t t! « 1 4 E f" 3 i-l .0 C *.. , ^ , 3S'-"> "^ W e to c a i tt ^ £Jl4fJ44V) L; V f-* K t-l G) O .U O C> 4J '^ _t c Si2 o _ l-< K. DIANE COTTETNET, Ph.D, Pharmacology and Toxicology Branch, B t/ c v es •- • c o r A r*> <n m ^-1 Uw ** « 5| •:i|i I j^ P *-< . K 0 VO rl £^ Jj I en s "*- x» I r4 B 1 I e»-) at «J C gl K? •O f. I ) Research Triangl DE. J. MCLATTGHLIN, Food and Drug Administration Washington, D.G. -* ' K ^ 1 ! r: : r.i H i <> « £I£S 1^1 ,S ; — -^ ---^ 1 <j| ja| Washington, D.G. THE DOW CHEMICAL COMPANY, Midland, Mich., February 9,11 , Division of Toxicology, Departi, Food and Drug Administration, assay on the sample of 2,4,5-trichloro^ „ -^— you is as follows: V-I^UH yrouuction batch 120449) whidh we recently sent to PEECENT 2,3,7,8-tetrachlorodibenzo- p-dioxin = (0.5 ppm) 2,6-dichlorophenoxyacetic acid = <0.02 2,5-dichlorophenoxyacetic acid = 0.42 2,4-dichlorophenoxyacetic acid = 0.05 2,3,6-trichlorophenoxyacetic acid = 0.55 2,4,6-tiehlorophenoxyacetic acid = <0.1 Bis- (2,4,5-trichlorophenoxy) acetic acid = 0.4 3 isomers of dichloromethoxy phenoxyacetic acid = 2.9 2,4,5-trichlorophenol = 0.23 (Max.) sodium chloride = 0.035 2,4,5-trichlorophenoxyacetie acid = Balance Freezing Point = 152.9° C Total assay by acid titration = 100% Sincerely, GEOEOE E. LYNN Government Regulatory Relations, Dow Life Sciences. �356 357 MABCH 2,1970. CHLOBINATED DIBENZO-P-DIOXIN STATTDABD (F768) 1. We have supplied Dr. C. Rilliams with a mixture of tri- and titrachlorodibenzo-p-dioxins (F768) prepared by direct chlorination of dibenzo-p-dioxin at room temperature for two and one-half hours (see memo of Bess to Campbell 1/30/70 and memo of Firestone to Campbell 2/5/70). 2. This mixture, previously analyzed by EC-GLC and found to contain about 50% tetrachlorodibenzo-p-dioxin (GLC peak area), was reexamined by GLC with flame ionization detection (6 foot glass column; 200°C; 3% OV-101; Packard GC model 871), which gives a more accurate determination of eompr> sition. By this latter analysis, it was determined that the chlorination mixture F768 consists of two components as follows: (a) 38% 2,3,7-trichlorodibenzo-p-dioxin (6) 62% 2,3,7,8-tetrachlorodibenzo-p-dioxin MARCH 26, 1970. SAMPLES FOE CHICKEN EMBRYO TESTING 1. The following samples were delivered to Dr. Verrett on 3/17/70 for chicken embryo testing: Remarks Our. No. Identification F871 0.0878 mg in 10 ml of acetone. 0.3806 mg in 10 ml of acetone. 2,3,7,8-tetrachlorodibenzo-p-dioxin. FDA, prepared by Dr. Pohland (99.5% 0.1735 mg in 10 ml pure by GLC). of acetone. 2,4,5-trichlorophenoxyacetic acid; Dr. Williams' purified sample: further ex- 1.712 grams. tracted 3 x with petr. ether and 4 x with 1+1 petr. ether—diethyl ether. Sylvex; Dr. Williams' purified sample; further extracted 3x with petr. ether and 2.519 grams. 4xwith 1+1 petr. ether—diethyl ether. 2,3,7,8-tetrachlorodibenzo-p-dioxin, Dow, pure(ca.95% byGLC).. F877 2,7-dichlorodibenzo-p-d'oxin, Dow, pure(ca.99% by GLC) F883 F881-A-1 F881-B D. FIRESTONE, Head, Fats and Oils Section. CHLOROPHENOL SAMPLES MARCH 26, 1970. 1. About 3 grams each of the following chlorophenols were delivered on 3/23/70 to Dr. Verrett (at her request) for chicken embryo testing. Our No. F888-sub A... F888-sub B. F888-sub C F888-sub D F888-sub E. F888-sub F Identification Pentachlorophenol (purified) Aldrich Chemical Co 2,3,4,6-Tetrachlorophenol (technical) Eastman Chemicals. 2,4-Dichlorochlorophenol (purified) Eastman Chemicals... Ortho-chlorophenol (purified) Fisher Scientific 2,4,5-Trichlorophenol (technical) Eastman Chemical; 2,4,5-Trichlorophenol (reagent) Dow Chemical Co EC-SLCanal PPB *CEF . 167. . 96,500. - No analysis. . Trace. . Trace. "Hexa-, hepta-, and octachlorodibenzo-p-dioxine. J. RESS, Fats and Oils Section, I. Cleft Palate. 21-day Embryo. 0.02 micrograms 2,3,6,7-Tetrachloro-dibenzo-pdioxin. �358 II. Abnormal (incomplete) Development of Eyelid. 21-day Embryo. 0.0025 micrograms 2,3,6,7-Tetrachloro-dibenzo-p-dioxin. 359 III. Edematous ' ng (micrograms), � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Alvin L. Young Collection on Agent Orange Description An account of the resource <p style="margin-top: -1em; line-height: 1.2em;">The Alvin L. Young Collection on Agent Orange comprises 120 linear feet and spans the late 1800s to 2005; however, the bulk of the coverage is from the 1960s to the 1980s and there are many undated items. The collection was donated to Special Collections of the National Agricultural Library in 1985 by Dr. Alvin L. Young (1942- ). Dr. Young developed the collection as he conducted extensive research on the military defoliant Agent Orange. The collection is in good condition and includes letters, memoranda, books, reports, press releases, journal and newspaper clippings, field logs and notebooks, newsletters, maps, booklets and pamphlets, photographs, memorabilia, and audiotapes of an interview with Dr. Young.</p> <p>For more about this collection, <a href="/exhibits/speccoll/exhibits/show/alvin-l--young-collection-on-a">view the Agent Orange Exhibit.</a></p> Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Box The box containing the original item. 106 Folder The folder containing the original item. 2926 Series The series number of the original item. Series V Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Creator An entity primarily responsible for making the resource Verret, Jacqueline Source A related resource from which the described resource is derived Hearings Before the Subcommittee on Energy, Natural Resources, and the Environment of the Committee on Commerce; United States Senate, Ninety-First Congress, Second Session on Effects of 2,4,5-T on Man and the Environment Date A point or period of time associated with an event in the lifecycle of the resource 1970 Title A name given to the resource Statement of Dr. Jacqueline Verrett, Food and Drug Administration, Department of Hew