Toxicants Occurring Naturally in Foods. (1966)

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

ANTIGENS AND ALLERGENS 75 Other Defense Mechanisms Although not essential to this discussion, consideration should always be given to the competency of the digestive, complement, properdin, phagocyte, and interferon systems and to mucosal and vascular integrity.12 Antigen—Antibody Reactions These reactions are primarily due to the characteristic of the anti- bodies to form precipitates with their antigens in liquid and semisolid media. They are also demonstrated by hemagglutination, complement fixation, and the ammonium sulfate precipitation of radiolabeled anti- gen and antibody complexes. Passive cutaneous anaphylaxis can be produced in guinea pigs by the use of the human classical antibody and its antigen, but not by human reagin.! Allergen—Reagin Reactions The demonstration of reagins depends upon a predominant charac- teristic of these antibodies, namely their ability to become fixed to the skin and other tissues. Direct intradermal testing with specific allergens results in a wheal and erythema reaction that occurs within 10 to 20 minutes. Frequently, allergic individuals develop an immediate wheal and erythema reaction upon application of an allergen to the con- junctiva, nasal, bronchial, and gastrointestinal mucosa, scarified skin, and even the intact skin in the exquisitely allergic patient. The Prausnitz-Kiistner (P-K) test is elicited in a nonsensitive person when reaginic serum is injected either intradermally or intravenously, and the injected site is then challenged with allergen. The rectal mucosa can also be passively sensitized with reagin, just as in the P-K skin test. This test has the added advantage that the reagin can be neutralized in vitro with the allergen to demonstrate specificity. Moreover, the P-K test has been successful in some monkeys, thereby eliminating the need to make the test in man and thus eliminating the danger of transmitting viral hepatitis. There is no in vitro quantitative method of analysis for the presence of reagin in human serum. Several investigators have attempted, unsuccessfully, to develop a test involving inhibition of standard pre- cipitating systems to demonstrate and to measure reagins.46 Some qualitative success has been achieved by a radioautographic method

76 HERBERT C. MANSMANN, JR. applied to reaginic serum electrophoresed in agar gel using a radio- labeled allergen. The basophil degranulation test has received extensive attention over the past few years. As yet, it is not clear whether degranulation of iso- lated basophils upon contact with antigen is accomplished by the action of IgG or IgA. Although the role of the basophil has been reviewed in relation to urticaria and anaphylaxis,*! the effect of specifically isolated immunoglobulins has not been evaluated using a pure antigen and allergen. Disheartening is the fact that the leucocyte lysis test has most of the same disadvantages as the basophil degranulation test. In this test, human leucocytes, sensitized in vitro with reaginic serum, release histamine in the presence of allergen. Although very time consuming, this procedure appears to augment the clinical diagnosis in selected cases.§ Delayed Hypersensitivity Reactions The local tuberculin reaction is the classical model of delayed hyper- sensititivy,> and can be elicited by the injection of antigen into most tissues of sensitized animals. A sharply circumscribed area of induration and erythema begins to develop in 6 to 12 hours and reaches its maxi- mum within 24 hours. The predominent microscopic picture is that of perivascular lymphocytic infiltration. There is no quantitative method to measure, in vitro, delayed hyper- sensitivity reactions. The effects of antigens on tissue cultures and on circulating granulocytes and lymphocytes has been receiving a great deal of attention in recent years.3 However, these methods have not yet been applied to the study of immunological food intolerance. Foops AS ANTIGENS A review of the evidence that foods can act as antigens in experimental animals and humans is presented. The emphasis is placed on immuno- logical evidence of antigenicity rather than on manifestations of immunological reactions.

ANTIGENS AND ALLERGENS 77 IN EXPERIMENTAL ANIMALS Cow’s Milk Many investigators have demonstrated that cow’s milk is antigenic in guinea pigs and rabbits. The sensitization of guinea pigs with electro- phoretically pure a-casein, B-lactoglobulin, and a-lactalbumin has been measured by systemic anaphylaxis and by the Schultz-Dale reaction using isolated ileum from these animals.39 These experiments demon- strated that the @-lactoglobulin was antigenic even after only a single immunizing dose. However, both a-casein and a-lactalbumin proved antigenic after multiple injections. The effect of heat on the antigenic properties of milk proteins has been extensively investigated.8.22-4042 Cross-immunization and chal- lenge experiments with a heat-denatured milk and unheated milk fractions have been performed in guinea pigs.4° Heat-denatured milk loses the antigenic and anaphylactic activities of a-lactalbumin but retains the sensitizing capacity of B-lactoglobulin and the antigenic activities of a-casein. Whole milk subjected to different temperatures for 15 minutes was analyzed by immunoelectrophoresis against an antibovine colostrum serum capable of forming 16 precipitin lines.22 The precipitating capacity of a-casein was retained after exposure to 120°C; and the a-lactalbumin and £-lactoglobulin precipitating ca- pacity was retained after exposure to 100°C. The major immune glo- bulin component as well as a few other proteins lost this capacity after exposure to 80°C. Normal pasteurization at 64°C for 30 minutes did not change the number of precipitating systems from that of non- pasteurized milk. Using specific rabbit antisera prepared against a-casein, B-lactoglobulin, and a-lactalbumin, the effect of heat on different milks was studied by the Ouchterlony agar diffusion method 8 It was demonstrated in these experiments that heat denaturation completely destroys the precipitating capacity of a-lactalbumin, partially destroys 6-lactoglobulin precipitation and slightly impairs the potency of a-casein. However, the results of similar experiments using a precipitin ring test and the passive cutaneous anaphylaxis test were at variance with the previously described experiments.42 The heating of milk, these authors concluded, could render the bovine serum albumin and the bovine y-globulin components antigenically inactive, and could also reduce somewhat the antigenicity of the a-lactalbumin. The effect of gamma irradiation was to decrease the

78 HERBERT C. MANSMANN, JR. precipitating capacity of a-lactalbumin, yet there resulted no significant change in this capacity of the 6-lactoglobulin and a-casein.!” In these experiments, several heat-denatured milk products demonstrated significantly fewer precipitin lines than did the irradiated milk. Interesting results were obtained in experiments on lightly anesthe- tized milk-immunized guinea pigs.*5 Inhalation of cow’s milk caused arrest of respiration, and many of these animals died quietly without any struggle or excitement. Moreover, signs of the acute emphysema of anaphylaxis were missing. Unanesthetized guinea pigs died of severe anaphylaxis with acute emphysema. Immunoelectrophoretic analysis of cow’s milk against rabbit and antimilk serum has demonstrated 12 separate precipitin bands.2! Three a-casein, one §-lactoglobulin, one a-lactalbumin, and one unidenti- fied antigen were demonstrated in the cow’s milk. By the use of anti- bovine serum, capable of forming 21 bands with bovine serum, 14 bands were removed by absorbing the antiserum with bovine milk.> Two immunologically identical B-lactoglobulins are present in cow’s milk. Recently, these have been separated electrophoretically as A and B.16 Goat’s Milk Because some infants appear clinically to tolerate goat’s milk better than cow’s milk, comparative immunological studies using rabbit antisera have been performed.23 These studies have shown an im- munological relationship between bovine and goat’s milk, a-casein, a-lactalbumin, and f-lactoglobulin. Human Milk Specific rabbit human milk antisera contains antibodies against at least 18 antigenic constituents of human milk identical or related to human plasma proteins.24 At least 13 “milk specific’? components were demonstrated. When human milk was compared with cow’s milk, immunological cross-reactions were present with the serum albumin, y-globulin, and a-lactalbumin fractions. Egg The antigenicity of egg-white proteins has been extensively investi- gated. Recently, fractions of egg-white proteins, prepared with diethyl- aminoethyl (DEAE) cellulose chromatography, were used to produce

ANTIGENS AND ALLERGENS 79 specific rabbit antisera.!! Specific antisera to ovalbumin, conalbumin, ovomucoid, lysozyme, and flavoproteins I, II, and III, which were pure by the Ouchterlony method, were obtained. Moreover, these antisera were analyzed by hemagglutination and were all found positive without cross-reactions between proteins and antisera. Wheat Biochemical and immunochemical properties of wheat have been reviewed.? Rabbit wheat gluten antiserum contains antibodies against four wheat gluten, three rye, two barley, one maize, and one oat components. IN MAN Cow’s Milk Many in vitro immunological techniques have demonstrated the presence of classical antibodies to milk and its fractions in human serum. As early as 1923, Anderson and Schloss demonstrated pre- cipitating antibodies to milk in the sera of infants recuperating from acute diarrhea.2 However, the significance of such a finding still remains obscure, 40 years later, for many asymptomatic infants and children have milk antibodies. What is significant, however, is that these anti- bodies do develop following natural ingestion of milk as food. Fifty percent of 98 adult and 96 umbilical cord sera contained low titers of antibodies to milk, as measured by the tannic acid hemag- glutination method.!8 As only classical antibodies cross the human placenta these data imply that milk proteins can act as antigens. Other studies using the hemagglutination technique have been re- ported. Milk hemagglutinins were demonstrated in sera from half of 25 milk-intolerant subjects and also from half of the controls who were not clinically sensitive to milk.3! Sera from normal adults and ulcerative colitis patients were also compared.‘’ A statistically significant elevation in titer of antibodies against casein and f-lactoglobulin was found in the colitis patients. Sera from patients with chronic or repeated acute episodes had high titers against casein only. In this same study, sera were analyzed for milk hemagglutinins from patients with pernicious anemia, and the results were comparable with the results obtained with normal sera. Sera from children suffering from celiac disease had higher

80 HERBERT C. MANSMANN, JR. titers and more milk precipitin lines than normal controls as determined by the microslide agar diffusion method.?7 Seventeen sera from patients with celiac syndrome were studied and 17 had hemagglutinins to casein, 17 to a-lactalbumin, and 15 to 6-lactoglobulin.48 These results proved to be statistically significantly different from results obtained with control sera. Cystic fibrosis patients, in spite of their decrease in pancreatic enzymes, did not have an increase in milk hemagglutinin titers in their sera.38 It should be appreciated that 10 percent of normal dietary protein reaches the large bowel without being hydrolyzed ;4 increased absorption easily occurs through an already inflamed colon.2-44 The milk precipitin syndrome of Heiner et al.26 has received a great deal of attention recently. The children in his series had manifestations of iron deficiency anemia, hepatosplenomegaly, poor growth, pulmo- nary hemosiderosis, and gastrointestinal and atopic symptoms. Using a microslide method of agar gel double diffusion, seven or more pre- cipitin lines against raw cow’s milk have been demonstrated in sera from seven of these severely ill children. The number of milk precipitin bands demonstrated were: 14 bands in one sera, 10 in another, 8 in another, and 7 in four others. During this study, sera from 18 asympto- matic children, of the 2,200 tested, had one to three milk precipitin lines. An addendum to their report describes an additional 26 patients with seven or more precipitin lines and with various manifestations of the clinical syndrome found among 1,284 children suspected of having the milk precipitin syndrome. Using a similar micro agar gel diffusion method, other investigators,?8 upon analyzing sera from 1,618 infants and children, were able to demonstrate milk precipitins in 87 sera from sick children with similar symptoms. In contrast to findings of the previously discussed study, none of these sera contained more than five precipitating systems, yet these authors used pasteurized undiluted skim milk as their test antigen. Sera from 5 of 380 asymptomatic infants and preschool children studied also had precipitins. A third evaluation of sera from 288 children with symptoms said to be similar to those of the milk precipitin syndrome has been published.78 The milk hemaggluti- nation test was positive in 67 percent, while the precipitation tests, using both the micro-Ouchterlony and capillary-tube methods, were positive in only 25 percent of the sera studied. These authors used raw skim milk, pasteurized skim milk, and protein fractions as test antigens. It is noteworthy that several sera gave positive results by one precipitin method yet not by the other.

ANTIGENS AND ALLERGENS 81 An evaluation of milk precipitins has been made with immunoelectro- phoresis in agar gel.2838 Unfortunately, only the milk was electro- phoresed.28 The other report?8 did, however, demonstrate that milk antibodies in 18 human sera were present in the y-globulin, but no mention was made of the particular fraction that contained them. Sera from 79 of 89 milk-intolerant infants and children, proven so by oral challenge,'4 were analyzed for milk antibodies.*3 The results of micro agar gel diffusion and hemagglutination tests were essentially the same as for the controls and could not be correlated with the oral challenge reaction. The antigens used were raw skim milk, colostrum, casein, a-lactalbumin, 6-lactoglobulin, and serum albumin. Although the passive cutaneous anaphylaxis method gave a higher incidence of positive results in the milk-intolerant group (53 of 79), there was no correlation with the oral challenge using the same protein. The longer the patient avoided milk, the greater the likelihood that the passive cutaneous anaphylaxis tests would become negative. Goat’s Milk Sera from patients with the milk precipitin syndrome commonly had precipitin lines against raw goat’s milk.76 Hunan Milk In one hundred infants’ sera analyzed for cow’s milk antibodies by hemagglutination, none contained antibodies against human milk proteins.!9 Moreover, sera from the patients with milk precipitin syn- drome showed no bands against human proteins.*6 Egg The antigenicity of egg proteins has been demonstrated in man. After acute gastroenteritis, egg albumin may appear in the circulation of infants ingesting eggs.44 Many of the infants subsequently develop egg precipitins in their sera. Sera from 23 patients with egg-white intolerance contained one to three hemagglutinins against ovalbumin, conalbumin, and ovomucoid.!! In another study, egg-white hemagglutinins were present in sera from half of the egg-intolerant patients and half of the controls studied.3!

82 HERBERT C. MANSMANN, JR. Wheat Using a soluble product of proteolyzed gluten (fraction III of Frazer) as the antigen, hemagglutination tests demonstrated antibodies in 32 percent of the mothers’ sera and 85 percent of the cord sera, a highly significant difference (P < 0.001).5° Wheat proteins have also been investigated immunologically in celiac disease.2”7 While 14 of 40 sera from patients with clinical celiac disease had precipitins to crude wheat protein, only 37 of 1,196 sera from subjects with other complaints were positive. The gliadin fraction con- tained four different antigenic substances. Of 25 patients with celiac disease confirmed by biopsy, sera from nine had precipitins to crude wheat protein and ten had precipitins to gliadin-1 fraction. Others had precipitins in varying combinations to gliadin fractions 2, 3, and 4. Foops AS ALLERGENS In this section a review of the evidence that foods act as allergens is presented. The allergenic nature of various foods in atopic human subjects is considered, although there is no in vitro method of measur- ing reagins. Atopic diseases have been reported in experimental animals.37 N MAN Cow’s Milk In addition to many other clinical observations reported with direct and P-K tests, followed by oral challenge, milk intolerance has been excellently studied from many aspects by a large collaborating group of investigators.!4-15.43 In this study of 89 children proven milk-intolerant by oral challenge,'4 direct skin testing with four milk proteins was evaluated.!5 The authors reported 50 of 85 (59 percent) milk-intolerant children had positive wheal and erythema reactions to one or more proteins. The authors stated that 14 of 15 strongly positive intradermal reactions did correlate with oral challenge data, using specific milk proteins. No P-K tests were performed and the direct skin test sites were not observed at 24 hours for delayed tuberculin-type reactions. In another review of milk intolerance, the authors considered the patients as belonging to one of two groups, the quick and the slow

ANTIGENS AND ALLERGENS 83 reactors.3 Although the authors do not emphasize the point, their data do show the superiority of the P-K test to the cutaneous scratch test, especially in the very young infant. It is interesting to find, but certainly not unexpected, that milk hemagglutin titers were not significantly elevated in sera from children with atopic disease.*8 In regard to the milk precipitin syndrome, one article” reported that all of seven patients had varying degrees of positive immediate wheal and erythema reactions to intradermal tests with various milks and milk components. This demonstration of reagins by direct skin test was not confirmed by the P-K test as only one of four sera studied gave a mildly positive reaction. A second report? did not give skin test results. Sixteen patients discussed in a third 1eport?® did not have milk reagins as measured by direct skin tests with commercially available cow’s milk allergenic extract. Goat’s Milk In an excellent review on milk intolerance, there is a discussion of one patient with clinically apparent goat’s milk intolerance who had negative scratch tests and a negative P-K test.!3 Human Milk Two of five infants tested had positive P-K tests to human milk, al- though there was no clinical disease reported and the direct scratch test was negative.!3 Egg Skin testing to egg white has been extensively reviewed.*2 In an excellent study on egg-white intolerance,!! sera were absorbed with human red cells coated with a specific egg-white protein fraction. The P-K titer was reduced simultaneously with the hemagglutinin titers in the four sera tested. All 23 patients with egg intolerance had positive direct immediate skin-test reactions. Foops AS DELAYED HYPERSENSITIVITY INDUCERS The role of foods as antigens in the state of delayed hypersensitivity has not been sufficiently evaluated, especially in man. In demonstrating

84 HERBERT C. MANSMANN, JR. delayed hypersensitivity, technique is important. To eliminate non- specific irritation wheal and erythema reactions—false positive re- actions—intradermal skin tests in suspected atopic patients are usually performed by the injection of 0.02 ml of extract containing 0.01 mcg of protein. Larger volumes lead to false positive reactions, and stronger concentrations frequently result in anaphylaxis. However, in human tuberculosis, frequently 0.1 ml containing 5.0 mcg of purified derivative of tuberculin is required to elicit a positive delayed skin reaction. Thus, in the evaluation of the allergic patient, only 0.05 of a delayed hyper- sensitivity skin test dose is usually given. It is, therefore, understandable why delayed skin reactions are so seldom seen. However, delayed skin reactions are frequently seen when concentrated solutions are used. IN EXPERIMENTAL ANIMALS An extensive volume of published work on the ability of purified food components, such as bovine serum albumin and ovalbumin, to induce delayed hypersensitivity in animals is available.3 12 IN MAN Delayed hypersensitivity can be specifically induced in human subjects by intradermal injections of small amounts of antigen, by application of antigen to the nasal mucosa, and by the intradermal injection of antigen-antibody precipitates. Investigators have deliberately not used food antigens in such studies. The spontaneous occurrence of delayed hypersensitivity to food anti- gens has been reported.‘5 In a series of nine patients with clinically proven milk intolerance, manifested by gastrointestinal symptoms, yet having low milk hemagglutinin titers, skin tests with milk fractions were performed. Although the 15-minute reaction was negative, at 24 hours four were positive to 6-lactoglobulin, one was positive to 8-lacto- globulin, a-lactalbumin, and casein, and four were negative. Four milk-tolerant patients with low milk hemagglutinin titers had positive delayed skin reactions to 6-lactoglobulin at 24 hours. It is noteworthy that these authors skin tested with 10 mcg of a-lactalbumin protein and 20 mcg of casein and £-lactoglobulin protein. This is two to four times the amount necessary to obtain a positive tuberculin test.

ANTIGENS AND ALLERGENS 85 IMMUNOLOGICAL FOOD INTOLERANCE The clinical history is by far the most important tool for the diagnosis of a food intolerance. Clues from the history can lead to trials with the elimination and feeding test method; although at times confusing, this is the only test that is unequivocally significant in the individual case. In many instances there are multiple foods playing the role of toxicants, and their partial removal will only bring partial alleviation of symptoms. None the less, the feeding of the omitted food after a few weeks of abstinence should, and not infrequently does, produce a restoration of the original problem in cases of true food intolerance. When immunological components such as circulating classical antibodies—fixed or circulating reagins—or the state of delayed hypersensitivity have been demonstrated against a specific food, and when clinical food intolerance can be demonstrated, then the diagnosis of immunological food intolerance can in all probability be con- sidered valid. In some persons only special events, such as an attack of enteritis, can unmask a food allergy. Circulating antibodies in such people are harmless until enteritis admits an antigen such as egg or milk protein to the blood stream.?.4 Certain injectable vaccines grown on chick embryos contain sufficient egg proteins to induce anaphylaxis in a sensitive individual. Therefore, the detection of an immunological component can provide valuable information about the involved individual. It is not within the scope of this paper to discuss the evidence for or against the association of immunological responses with the antigenic or allergenic stimulation of food in regard to the pathogenesis of a particular disease. The reviewer’s purpose is to show mechanisms of analysis that are currently available to study these problems, and to illustrate, when possible, how new techniques have been applied in an attempt to understand these diseases more fully. ANTIGEN-ANTIBODY REACTIONS Cow’s Milk The milk precipitin syndrome,” with its milder variants,8 38.43 appears to be associated with a milk antigen-antibody reaction. Although reagins were present in sera of most of the seriously ill patients,© the

86 HERBERT C. MANSMANN, JR. manifestations of the syndrome were not those of allergen-reagin interaction. Most of these patients could tolerate milk long before the disappearance of reagin. The role of delayed hypersensitivity in this group of patients, however, needs to be evaluated more completely. Of the six patients challenged with milk, two had a return of the intense milk precipitin syndrome, while others had evidently become tolerant during the restricted period of 3 to 6 months.6 Following milk restriction, the number of milk precipitin bands and the precipitin titers decreased. Sudden unexplained death in infancy, referred to as “‘cot death,” has been suggested to be the result of a milk antigen-antibody reaction in some cases.*6 These authors included only infants who had been bottle fed at the time of death, and had milk present in their stomachs at post- mortem examinations. The hemagglutination titers in the sera studied, except perhaps in 4 of the 24 sera included, were as high as or higher than the mean titer for sera of children the same age. Moreover, histological similarity was noted between the lungs of the “cot death” babies and the lungs of lightly anesthetized milk-immunized guinea pigs challenged with a drop of milk intratrachially.35 Although this work is not completely accepted, it would seem hasty and ill-advised to ignore the suggestion that a milk antigen-antibody reaction might be the cause of some of the cases of sudden and unexplained death in infancy. None of these sera was subjected to reagin analysis by P-K tests. A case of sudden death that followed the microaspiration of cow’s milk has been particularly well documented.3° The infant, who had regurgitated during sleep, suddenly became cyanotic and died, in spite of immediate medical attention from his father, a physician. Autopsy revealed hyperemia of the bronchial mucosa with narrowing of the bronchial lumens. No aspirated milk was present grossly or micro- scopically. Cow’s milk antigen was present in the bronchial mucosa, as demonstrated by immunochemical techniques, although the serum did not contain any of the milk antigens. Microscopic examination of the lungs revealed extensively contracted bronchi containing shed epi- thelial cells. Specific milk protein antibodies were shown to be present in the infant’s serum by the hemagglutination and complement fixation methods. The P-K test was not performed. One patient with celiac disease proven clinically and by biopsy had a negative hemagglutinin reaction to wheat gluten, yet had high-titer hemagglutinin tests to casein, a-lactalbumin, and £-lactoglobulin.® Improvement occurred while the child was eating a milk-free diet that contained gluten.

ANTIGENS AND ALLERGENS 87 Wheat The repeated clinical observation that wheat gluten will precipitate symptoms of celiac disease and the finding of classical antibodies to wheat gluten in these patients appear to indicate that serious considera- tion should be given to the role an antigen-antibody reaction might have in this disorder.*8 27 ALLERGEN-REAGIN REACTIONS There is a large volume of literature based in varying degrees on clinical history, oral challenge, and direct and passive skin tests indi- cating that foods play a significant role in allergen-reagin reactions. Moreover, an extensive evaluation of the relationship of food pre- servatives to such reactions has been published.> As vegetable gums are also added to many foods, their allergenic capacity should be appre- ciated.4 These references are given in lieu of a complete description of the manifestations of allergen—reagin reactions. Cow’s Milk Children have been known to suffer from bronchial asthma for years until a complete clinical history suggested immunological milk in- tolerance. Time should be taken to listen to the parents for the clues are in the history and there is frequently no need for skin tests or modern immunochemical techniques. One of the most comprehensive studies of the effect of oral challenge with milk and isolated milk proteins has recently been published.'4 These authors accepted the diagnosis of milk intolerance in 89 infants and children among 700 patients suspected of intolerance according to the following criteria: (1) Symptoms disappeared when milk was com- pletely eliminated from the diet. (2) Symptoms recurred in 48 hours following oral challenge with milk. (3) Three such challenges resulted in similar events. (4) Symptoms subsided following each challenge reaction. Forty-five patients were challenged by an equivalent amount of each milk fraction present in 100 ml of skim cow’s milk. Each patient tested reacted to one or more of the purified proteins, but all were not tested to all proteins. Sixty percent reacted to casein, 62 per- cent to 6-lactoglobulin, 53 percent to a-lactalbumin, and 52 percent to bovine serum albumin. Six patients responded to all four proteins, seven responded to three proteins, twelve responded to two proteins,

88 HERBERT C. MANSMANN, JR. and nineteen reacted to only one protein. In a patient studied quanti- tively, sublethal anaplylaxis was produced with less than 1 mcg of casein, G6-lactoglobulin, and a-lactalbumin. This patient was non- reactive to 40 mg of bovine serum albumin. An interesting clinical biochemical observation was the increase in extracellular fluid volume, as determined by chloride space measure- ments, at the expense of intracellular fluid volume, associated with immunological milk intolerance proven by oral challenge and P-K tests.’ This type of internal derangement needs more exploration. Such occurrences may be a more subtle way of evaluating an in vivo allergen- reagin reaction than waiting for gross symptoms to develop. Egg Egg protein intolerance is frequently seen clinically but has not been subjected to extensive immunological study. Chickens fed fish flour produced eggs that were not tolerated by a fish-sensitive patient, yet the patient could tolerate eggs from grain-fed chickens.2° MIXED REACTIONS Although specific evidence is lacking, it can be argued, theoretically, that foods can act as antigens or as allergens. In most persons they apparently do act as antigens; yet in atopic persons the allergenicity of a food might induce reagin formation. The antigenic properties might induce formation of classical antibodies, the delayed hypersensitivity state, or both, even in the atopic individual. Therefore, a patient may have one, two, or three mechanisms operating at a time. It 1s this reviewer’s belief that much of the confusion that exists is a result of mixed reactors. Even though cause-and-effect relationships are easier to study in pure states, human subjects, unfortunately, seldom behave simply. A few examples are given in an attempt to stimulate some con- sideration of these thoughts. A mixed reaction might explain why the very excellent study of 89 milk-intolerant orally challenged patients!*-15-43 failed to disclose the expected correlation between symptoms and serologic tests. These authors might have been dealing with several combinations of reactors. There is a definite need for more exact evaluation of each food-in- tolerant patient. No mention was made in the reports cited of the fact that 58 of the 85 patients who were skin tested were under 1 year of

ANTIGENS AND ALLERGENS 89 age, and that 38 of the 85 were under 7 months of age. It is well known that the younger the infant, the more likely one is to obtain a false negative skin test; therefore, P-K tests should have been performed. No observations of the skin-test sites were made at 24 hours, in an attempt to determine the role of delayed hypersensitivity, even though a strong enough concentration of antigen was used in some cases. However, these investigators excluded patients with clinical mani- festations that occurred later than 48 hours after oral challenge, thereby probably automatically reducing the number of those reported with delayed hypersensitivity. Of the 89 patients challenged by the oral route with milk, 35 had reactions at 6 hours, 26 after 12 hours, 11 after 24 hours, 10 after 1 to 3 days, and 1 after 7 days. Such delays would be expected in systemic delayed hypersensitivity reactions. Mixed reactions may also occur in ulcerative colitis patients. Classical antimilk antibodies, reagins, and delayed hypersensitivity reactions have been described. The onset of symptoms after oral challenge*? seems to bear out these possibilities. Symptoms started anywhere from 2 to 42 days after milk was reintroduced into the diet. In 13 of 200 patients with ulcerative colitis, milk intolerance was a significant contributor to exacerbations. Another example of a mixed reaction is the report that upon oral challenge with milk, a known milk-intolerant adult developed a delayed vesicular eczema of his middle finger.29 Although intradermal skin tests were negative, no mention was made of dose or periods of observa- tion. What was exciting about this report was the marked fall in circu- lating basophils in 30 minutes; the skin rash did not occur for 24 hours. Profound shock with unconsciousness occurred at 1 hour, possibly due to histamine release from the degranulation of the baso- phils. The rash may have resulted from delayed hypersensitivity. Unfortunately, no in vitro studies were performed. NONIMMUNOLOGICAL FOOD INTOLERANCE A consideration of food intolerance, especially if the manifestations are gastrointestinal, requires discussion of other factors in addition to an analysis of immunological components. Although dietary tests may appear to incriminate a specific food, the sequence of events does not in itself establish whether the incitant is operating immunologically or in some other way. Even the presence of an immunological component

90 HERBERT C. MANSMANN, JR. does not necessarily mean that the symptoms are caused by an immuno- logical reaction. Other mechanisms might be pharmacologic, bacterial, or toxic. Even the temperature and quantity of the food ingested may be important. Cow’s milk, regardless of whether it is fresh or old, pasteurized or boiled, has been reported to have a local excitatory effect, like that of acetylcholine, on isolated rabbit jejunum; whereas human and goat’s milk lack this activity.33 Disaccharidase deficiencies, either transient or permanent, con- genital or acquired, may account for some cases of nonimmunological food intolerances. These deficiencies should be considered when the liquidized stool pH is acid. CONCLUSION An attempt has been made to present, discuss, and interpret the avail- able evidence that foods and food components can act as antigens and allergens. The antigenicity and allergenicity of bacteria and their products in and on foods, of contaminating substances such as anti- biotics, and of additives such as preservatives and stabilizers need much more investigation than has previously been done. Demonstrations of the induction of classical antibodies, reagins, and the delayed hyper- sensitivity state have been cited in regard to certain foods, yet the deficiencies of the currently available studies are obvious. Aside from the usual overemphasis on one particular aspect of the immune response, there has been insufficient emphasis on the relationship of the time period required after oral challenge to produce symptoms 1n relation to the immune components involved. Immediate reactions are highly suggestive of allergen-reagin reactions. It has been suggested that slower reactions are due to digestive processes changing a food into an antigenic substance, yet this has not been satisfactorily proven. Reactions that occur several hours after ingestion should be evaluated for delayed hypersensitivity. The effect of the presence of one or more of the immunological components on absorption of specific foods has to be evaluated. The reason that the presence of the immune component does not always result in an immunological intolerance must be explained. The use of the dual-ingestion passive transfer test, a P-K test with the skin sites challenged by oral intake of antigen, is an example of the kind of experiments that are needed.

ANTIGENS AND ALLERGENS 9] If such techniques could have been employed in evaluating the tre- mendous number of food-intolerant patients studied to date, a more knowledgeable classification of the patients probably could have been accomplished. It would seem important that those seeing such patients become more aware of recent physical-chemical advances, in addition to the immunochemical, and be ready to apply these techniques when the occasion arises. REFERENCES 1. J. F. Ackroyd, ed., Immunological Methods, F. A. Davis Company, Philadelphia, 10 11. 12. 13. 14. 15. 16. Pa. (1964). A. F. Anderson and O. M. Schloss, “Allergy to Cow’s Milk in Infants with Nutritional Disorders,” Am. J. Diseases Children, 26, 451 (1923). B. G. Arnason and B. H. Waksman, “‘Tuberculin Sensitivity: In:munological Considerations,” Advan. Tuberc. Res., 13, 1 (1964). B. Borgstr6m, A. Dahlqvist, G. Lundh, and J. Sjévall, “Studies of Intestinal Digestion and Absorption in the Human,” J. Clin. Invest., 36, 1521 (1957). E. A. Brown, “Food Processing and Allergy,” Rev. Allergy Appl. Immunol., 11, 307 (1957). W. T. K. Bryan and M. P. Bryan, “The Application of In Vitro Cytotoxic Reactions to Clinical Diagnosis of Food Allergy,” Trans. Am. Laryng. Rhinol. Otol. Soc., 371 (1960). J. D. Crawford, G. A. Kerrigan, and M. B. Arnold,“Observations of a Metabolic Lesion in Cow’s Milk Allergy,” Pediatrics, 22, 122 (1958). L. V. Crawford and F. T. Grogan, “‘Allergenicity of Cow’s Milk Protein. IT. Studies with Serum—Agar Precipitation Technique,” Pediatrics, 28, 362 (1961). G. A. H. Elton and J. A. D. Ewart, “Immunological Comparison of Cereal Proteins,” J. Sci. Food Agr., 14,750 (1963). J. L. Fahey, “Heterogeneity of Gamma Globulins,” Advan. Immunol., 2, 41 (1962). O. L. Frick, “‘Hemagglutination in Serums from Egg-Sensitive Individuals,” Ann. Allergy, 20, 794 (1962). P. G. H. Gel and R. R. A. Coombs, eds. Clinical Aspects of Immunology, F. A. Davis Company, Philadelphia, Pa. (1963). J. W. Gerrard, D. C. Heiner, E. J. Ives, and L.W. Hardy, “Milk Allergy Recogni- tion, Natural History, Management,” Clin. Pediat. (Philadelphia), 2, 634 (1963). A. S. Goldman, D. W. Anderson, Jr., W. A. Sellers, S. Saperstein, W. T. Kniker, S. R. Halpern, and collaborators, Milk Allergy I. Oral Challenge with Milk and Isolated Milk Proteins in Allergic Children,” Pediatrics, 32, 425 (1963). A. S. Goldman, W. A. Sellars, S. R. Halpern, D. W. Anderson, Jr., T. E. Furlow, C. H. Johnson, Jr.,and collaborators, “‘Milk Allergy II. Skin Testing of Allergic and Normal Children with Purified Milk Proteins,” Pediatrics, 32, 572 (1963). P. Gough and R. Jenness, “Immunologic Identity of 8-Lactoglobulins A and B,”’ J. Immunol., 89, 511 (1962).

92 17. 18. 19. 21. 22. 23. 24. 25. 26. 27. 28. 29. 31. 32, 33, 34, 35, 36. 37. HERBERT C. MANSMANN, JR. F, T. Grogan, Jr., and L. V. Crawford, “Allergenicity of Cow’s Milk Protein. Ill. Effects of Irradiation on the Antigenicity of Cow’s Milk as Studied by the Agar Diffusion Technique,” J. Immunol., 87, 240 (1961). M. Gunther, R. Aschaffenburg, R. H. Matthews, W. E. Parish, and R. R. Coombs, “The Level of Antibodies to Proteins of Cow’s Milk in the Serum of Normal Human Infants,” Immunology, 3, 296 (1960). M. Gunther, E. Cheek, R. H. Matthews, and R. R. A. Coombs, “Immune Responses in Infants to Cow’s Milk Proteins Taken by Mouth,” Intern. Arch. Allergy Appl. Immunol., 21, 257 (1962). M. J. Gutmann, “Visible and Hidden Food Allergens,” Acta Med. Orient., 16, 1 (1957). L. A. Hanson and B. Johansson, “Immune Electrophoretic Analysis of Bovine Milk and Purified Bovine Milk Protein Fractions,” Experientia, 15, 377 (1959). L. A. Hanson and I. Mansson, “Immune Electrophoretic Studies of Bovine Milk and Milk Products,” Acta Paediat., 50, 484 (1961). L. A. Hanson and H. J. Andersen, “A Comparison of the Antigenic Relation of Human Milk and Goat Milk to Bovine Milk,” Acta Paediat., 51, 509 (1962). L. A. Hanson, “Immunological Studies of Human Milk with Specific Reference to the Immune Globulins,” Thesis, University of Gothenburg, Sweden, 1961. L. A. Hanson, “Immunological Analysis of Bovine Blood Serum and Milk,” Experientia, 15, 471 (1959). D. C. Heiner, J. W. Sears, and W. T. Kniker, “Multiple Precipitins to Cow’s Milk in Chronic Respiratory Disease. A Syndrome Including Poor Growth, Gastrointestinal Symptoms, Evidence of Allergy, Iron Deficiency Anemia and Pulmonary Hemosiderosis,”” Am. J. Diseases Children, 103, 634 (1962). D. C. Heiner, M. E. Lahey, J. F. Wilson, J. W. Gerrard, H. Shwachman, and K.-T. Khaw, “Precipitins to Antigens of Wheat and Cow’s Milk in Celiac Disease,” J. Pediat., 61, 813 (1962). N. H. Holland, R. Hong, N. C. Davis, and C. D. West, “Significance of Pre- cipitating Antibodies to Milk Proteins in the Serum of Infants and Children,” J. Pediat., 61, 181 (1962). L. Juhlin and O. Westphal, ‘“Degranulation of Basophil Leukocytes in a Case of Milk Allergy,” Acta. Dermato-Venereol., 42, 273 (1962). . H. J. Kaufmann, J. Lantz, and A. Biirgin-Wolff, “Anaphylactic Shock of the Lungs Triggered by Microaspiration of Cow’s Milk: A Form of Sudden Un- expected Death in Early Infancy,” Ann. Paediat., 200, 20 (1963). C. Larose, P. H. Delorme, M. Richter, and B. Rose, “Immunologic Studies on Milk and Egg Allergy,” J. Allergy, 33, 306 (1962). R. L. London and J. Blaser, “The Cutaneous Reaction to Egg White,” Pedi- atrics, 24, 1009 (1959). A. H. Mohamed, O. Zaki, K. Zaki, and L. Fahim, “Effect of Different Types of Milk on Intestinal Motility,” Nature, 185, 773 (1960). D. C. Nilsson, ‘‘Sources of Allergenic Gums,” Ann. Allergy, 18, 518 (1960). W.E. Parish, A. M. Barrett, and R. R. A. Coombs, “Inhalation of Cow’s Milk by Sensitized Guinea Pigs in the Conscious and Unconscious State,” Immunology, 3, 307 (1960). W. E. Parish, A. M. Barrett, R. R. Coombs, M. Gunther, and F. E. Camps, “Hypersensitivity to Milk and Sudden Death in Infancy,” Lancet, ii, 1106 (1960). R. Patterson, “Investigations of Spontaneous Hypersensitivity of the Dog,” J. Allergy, 31, 351 (1960).

ANTIGENS AND ALLERGENS 93 38. 39. 40. 41. 42. 43. 45. 47. 49. R. D. A. Peterson and R. A. Good, “Antibodies to Cow’s Milk Proteins. Their Presence and Significance,”’ Pediatrics, 31, 209 (1963). B. Ratner, M. Dworetzky, S. Oguri, and L. Aschheim, “‘Studies on the Aller- genicity of Cow’s Milk. I. The Allergenic Properties of a-Casein, B-Lacto- globulin and a-Lactalbumin,” Pediatrics, 22, 449 (1958). B. Ratner, M. Dworetzky, S. Oguri, and L. Aschheim, “‘Studies on the Aller- genicity of Cow’s Milk. II. Effect of Heat Treatment on the Allergenicity of Milk and Protein Fractions from Milk as Tested in Guinea Pigs by Parenteral Sensitization and Challenge,” Pediatrics, 22, 648 (1958). H. Rorsman, “Studies on Basophil Leucocytes with Special Reference to Urticaria and Anaphylaxis,”” Acta Dermato-Venereol., 42, (Suppl. 48), 1 (1962). S. Saperstein and D. W. Anderson, Jr., “‘Antigenicity of Milk Proteins of Pre- pared Formulas Measured by Precipitin Ring Test and Passive Cutaneous Anaphylaxis in the Guinea Pig,” J. Pediat., 61, 196 (1962). S. Saperstein, D. W. Anderson, Jr., A. S. Goldman, and W. T. Kniker, “Milk Allergy III. Immunological Studies with Sera from Allergic and Normal Children,” Pediatrics, 32, 580 (1963). O. M. Schloss, “The Intestinal Absorption of Antigenic Protein,” Harvey Lectures, 20, 156 (1924-1925). P. Sewell, W. T. Cooke, E. V. Cox, and M. J. Meynell, “‘Milk Intolerance in Gastrointestinal Disorders,” Lancet, ii, 1132 (1963). D. R. Stanworth, ““Regginic Antibodies,” Advan. Immunol., 3, 181 (1963). K. B. Taylor and S. C. Truelove, “Circulating Antibodies to Milk Proteins in Ulcerative Colitis,” Brit. Med. J., ii, 924 (1961). . K.B. Taylor, D. L. Thomson, S. C. Truelove, and R. Wright, ‘An Immunologi- cal Study of Coeliac Disease and Idiopathic Steatorrhoea,” Brit. Med. J., ii, 1727 (1961). S. C. Truelove, “Ulcerative Colitis Provoked by Milk,” Brit. Med. J., i, 154 (1961). R. Wright, K. B. Taylor, S. C. Truelove, and R. Aschaffenburg, “‘Circulating Antibodies to Cow’s Milk Proteins and Gluten in the Newborn,” Brit. Med. J., li, 513 (1962).

F. M. STRONG Pressor Amines in Foods A great diversity of amino compounds, many of them possessing high physiological activity, exists naturally in food and feeds. This dis- cussion considers only those related to the aromatic amino acids, phenylalanine, tyrosine, tryptophan, and histidine. These compounds include such well-known and highly potent biochemicals as histamine, tyramine, tryptamine, and their metabolites such as serotonin (5 hy- droxytryptamine, 5-HT) and norepinephrine. It is perhaps surprising that such typical constituents of animal tissue occur in plants, but that many do is indicated by the data in Table 1. The quantitative data in this table were derived by various analytical procedures, some based on chemical and physical methods, including chromatographic separations, while others were bioassays measuring a particular type of activity rather than an individual sub- stance. Nevertheless, the figures quoted probably are a reasonably close approximation to the amounts of the various compounds (or their physiological equivalents) actually present. It is quite probable that many other fruits and fermented foods also contain pressor amines. A systematic survey might be worthwhile. Intravenous administration to human beings of any such amounts of these amines as would be obtained by eating normal quantities of the foods listed in Table 1 would have disastrous consequences indeed. What then may be the hazards of consuming such foods? A number of factors must be considered in this connection. The amine may not be well absorbed from the intestinal tract or may exist in the original food in the form of an inactive conjugate that requires enzymatic hydrolysis to release the active principle. Conversely, the free amine, if absorbed, 94

PRESSOR AMINES 95 TABLE 1 Pressor Amines in Foods APPROXIMATE CONCENTRATION FOOD AMINE (mg/100 g*) REFERENCE Bananas, ripe Serotonin (5-HT) 3 16 Passion fruit Serotonin (5-HT) 0.1-0.4 4 Pawpaw Serotonin (5-HT) 0.1-0.2 4 Plantains, boiled or fried Serotonin (5-HT) 4.7 4 Plantains, green Serotonin (5-HT) 2-6 4,12 Plantains, ripe Serotonin (5-HT) 4-10 4 Pineapples, green Serotonin (5-HT) 5-6 5 Pineapples, ripe Serotonin (5-HT) 2 5 Pineapple juice Serotonin (5-HT) 2.5-3.5 2 Tomatoes, ripe Serotonin (5-HT) 0.34 17 Bananas Norepinephrine 0.2 16 Plantains, green Norepinephrine 0.2 12 Plantains, ripe Norepinephrine 0.25 4 Bananas Dopamine 0.8 16 Cheese, Camembert Tyramine 200 1 Cheese, Stilton Tryptamine — 1 Cheese, Stilton Phenylethylamine — 1 Lemons Octopamine _ 15 Lemons Synephrin — 15 Sauerkraut juice Histamine 40 mg/liter 9 Wine Histamine 5-20 mg/liter 11 © Fresh weight basis. may be detoxified by conjugation or by conversion to an inactive derivative. One particularly prominent route of detoxification of primary amines is oxidation to the corresponding carboxylic acid through the agency of monoamine oxidase (MAO). Detoxification of the naturally occurring osteolathyrogen, 8-amino- propionitrile (BAPN), for example, is accomplished in this manner with the formation of cyanoacetic acid.!° In view of the probable participa- tion of MAO in this conversion, the effect of an MAO inhibitor was investigated and a distinct potentiation of BAPN toxicity was ob- served.!3 Thus, feeding 0.02 percent of BAPN fumarate and 0.04 per- cent of 1-isonicotinyl-2-isopropylhydrazide phosphate (an MAO inhibitor) separately to turkey poults had very little effect on the birds, but, when the two were fed together at these levels, 70-90 percent died from the aortic rupture and massive internal hemorrhage characteristic of BAPN intoxication in this species. Similar results were obtained