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.
CYCLAMATE Ronald G. Wiegand Among artificial sweeteners, the most acceptable product is neither saccharin nor cyclamate, but a combination of the two. The combination has a better taste than either sweetener alone, and it has the further advantage of reducing the intake of each. The value of the combination to an individual consumer derives from one or more of three possible advantages: l. For the obese, the desire for sweets is satisfied while the reduction of calorie intake is made more palatable and therefore more likely, thus contributing to the medical treatment of the problem of severe overweight. 2. For the diabetic, where sugar intake has even more immediate medical consequences, the cyclamate-saccharin combination helps control the diet while allowing a more normal variety of diet. For the diabet- ic and the obese, incorporation in foods and tabletop use are important, in addition to use in beverages. 3. For the simply overweight people who wish to achieve a more nor- mal weight, artificial sweeteners contribute to effective restriction in caloric intake. This third category of use is admittedly the least necessary and the most difficult to defend. But let us not divert our- selves from recognizing that this is the largest single use of cyclamate and saccharin in the world. And I would put it to you that this is properly so, for in a society such as the United States or other major developed nations with a per capita sugar intake of about l00 pounds a year, reduction in sugar intake is a worthwhile goal. Saying that people can simply eat less sugar is failing to deal with the fact of that intake. l77
l78 Back in l969 and l970, cyclamate was banned in the United States. The ban was based primarily on two studies, one by Oser and one by Friedman. The Oser study was the first, a two-year chronic toxicity study of the cyclamate-saccharin combination in rats, sponsored by Abbott at the Food and Drug Research Laboratories in New York City. The Friedman study came a few months later, and this was a metabolism study in rats at the Food and Drug Administration. Both studies showed tumors in the urinary bladders, and while the data can be discussed pro or con, the decision in hindsight is not disputed. The significant point now is that 20 specific carcinogenicity and cocarcinogenicity studies have been completed in the years since then, and that all of them are unequivocally negative. These new carcinogenicity studies, accomplished in all cases without funding or even consultation with Abbott, formed the basis of our food additive petition in November of l973. They were performed in several countries around the world, as well as by the FDA and the National Cancer Institute here in the United States. The World Health Organiza- tion last year reached the conclusion based on these data that cycla- mate is not carcinogenic. In regard to Abbott's food additive petition, the Commissioner of the FDA recently asked the National Cancer Institute to convene a panel of international experts to give the FDA their opinion on the carcino- genicity of cyclamate. This will give an authoritative position on which the FDA can rely. Abbott's position on the carcinogenicity of cyclamate is that it is not. We would not have submitted our food additive petition if we had not reached that conclusion, and reached it firmly. A corporation whose continued livelihood depends on its reputation in the health-care field does not lightly try to reverse a decision on carcinogenicity, and then plan to market the product. There are some further scientific questions on cyclamate that have to be resolved. These deal primarily with the testicular effects, which have been studied recently at BIBRA, as well as some of the re- productive data, some behavioral effects, and some cardiovascular effects. It is our position that these data are not inconsistent with the safe use of cyclamate, and our detailed analysis of all the data is a matter of public record. The ultimate question relating to cyclamate usage in the United States in the short term relates in a practical sense to the levels of use that are found appropriate. Back in l969 the usage of cyclamate in the United States was about l7 million pounds. This sounds like a lot of cyclamate, but it amounts to the equivalent of 2 pounds of sugar per person annually. Thus it is not a significant fraction of the sweetener intake of about l00 pounds of sugar. Looked at another way, the l7 million pounds of cyclamate corresponds to an average per capita ingestion of 0.l gram per person per day. This is one-fiftieth of the generally accepted safe daily intake in l969, which gives generous allowance even for the inordinate user. Our current position is that the allowable daily intake could be 2.5, still giving an adequate safety factor.
l79 Certainly the single finding on cyclamate of greatest current visi- bility is the testicular effects. This occurs, at levels of ingestion which are reasonable, only when feeding the metabolite of cyclamate, cyclohexylamine. In a recent letter to Abbott, the FDA suggests as reasonable a no-effect level of 0.5 percent in the diet, which is equiv- alent to a daily intake of 48 to l08 grams of cyclamate by a 70-kilogram person, who converts 55 percent of the cyclamate to cyclohexylamine. Increasing this safety factor is the fact that the rats ingest all the cyclohexylamine in a few hours, rather than over the day, as man con- verts unabsorbed cyclamate. Further, the effects on testicular weight occur only in the presence of a 30 percent reduction in total body weight, and everything we heard today and yesterday says that this nec- essary precondition is not obtained in man. Additionally, a safety factor considerably less than l00-fold can be accepted when there is so obvious a warning sign as 30 percent weight reduction, as away from an unobservable effect. These factors are part of the judgments that enter the question of determining a safe level of use, and I have tried in this particular instance to share some of the actual data with you. Let us look at sweeteners from the standpoint of the use of resources, both the national resources of production and the resources of the purchaser. I have heard it said that perhaps the NAS report on saccharin asks for additional toxicity studies that use a portion of the available national resource for toxicological facilities in a rela- tively unnecessary pursuit -- that is, there are other priorities of higher merit. But consider the resources used in the production of l6 billion pounds of sugar per year (the farmer, the fertilizer manufac- turers, and even the railroad cars tied up in moving it around the country), in light of some of my fellow speakers' saying that a large portion of this sugar ingestion lies someplace between "empty calories" and possibly harmful. This makes doing a few toxicity studies pale in- to insignificance when you look at the relative use of resources. Looked at from the standpoint of the consumer, one can approach it both economically or scientifically. There are no convincing arguments that say sugar is safer than cyclamate, or vice versa, for that matter. The question is never addressed by the FDA, because it operates under regulations from which sugar is exempt as a food additive. The FDA is overwhelmed with enough difficult questions and decisions that it does not have time to search for more. From the economic side of the con- sumer question, there is only so much discretionary income in the United States. Sweetness has proved, as we understood yesterday, to be a highly inelastic commodity, wherein a fourfold increase in price has resulted in only a slight drop in its consumption. This decreases real income and surely has an effect on the nation's nutrition by substitut- ing sugar for some of the dollars available for other foods. I am not trying to raise any dangerous consequences. I really do not think there are any. But everything I see in this picture tells me that artificial sweeteners, including saccharin, cyclamate, and others, have a place in our society today.
l80 In conclusion, cyclamate has been studied even more extensively than saccharin and the data support cyclamate's safety. With this as a sine qua non, cyclamate should be available in the food supply for good medical reasons as well as to improve the quality of life for those who need it. The question of what items in the food supply (such as foods, beverage, or tabletop use) are allowed must be answered by considering the probable intake resulting from uses in different kinds of foods, the benefits to the various consumers, and the daily total intake found safe. This has to be worked out again for cyclamate in light of pres- ent knowledge. DISCUSSION PFAFFMANN: Dr. Sveda, in view of your intimate relation to the cycla- mate discovery, would you wish to make a remark? SVEDA: In response to Mr. Wessel's most cogent remarks this morning about the need to translate information into understandable language for the public, I should like to present some comments and demon- strations regarding the use of cyclamate. About one person in every ten in the entire world has eaten cyclamate safely, and in my opinion, to the benefit of their health. In this country alone, there were three out of four people eating cyclamate before it was completely wrongly banned. There is a further important point to be made. You may remember some scary headlines a half-dozen years ago about all kinds of things that would befall us if pregnant mothers ate cyclamate. Cyclamate was used widely from about June of l950 until at least October l8, l969 -- "Sour Saturday." If we assume that there are roughly 4 million births a year, plus or minus a couple of hundred thousand, this means that during this 20-year period some 80 million people -- more than one-third of the population of this country -- were conceived and born when cyclamate was in very wide use. Every- body from the age of 5 to 25 -- which means everybody in preschool, in grade school, in high school, and now either gainfully employed or in college or graduate school -- was born under those conditions. Where are the flippers for arms that we were told might result? There is nothing in the medical literature reporting any such reac- tions. I think this is a devastating point, and it is why I was so pleased with Dr. Wynder's suggestion that we ought to look at the facts as they are. Here is a sample of cyclamate. If I sell this to a man, knowing that he is going to use it for food or drug, I am committing a crime. This is not a parking ticket crime, but one that gets me a fine of $5,000 or more and six months or so in jail. This is not true in Canada. This is not true in Australia. This is not true in South
l8l Africa, Israel, and so on. Why is this horrible situation in exis- tence? The reason is, in my opinion, and I can document it, that one series of tests was responsible for wrongly taking cyclamate off the market and upon a recommendation of the National Academy of Sciences that was then accepted by the Food and Drug Administration. I will commit a further crime. I will taste cyclamate, and therefore I will compound the crime. This is illegal. Here are a couple of stuffed mice. I could not find any four- legged stuffed rats, so for the purposes of this demonstration I have a couple of stuffed mice in cages. Here is a sample of cycla- mate. Here is a sample of saccharin, and here is a sample of cyclo- hexylamine. All of the rats were fed a mixture of cyclamate and saccharin. Not one of the rats was fed pure cyclamate. Approximately halfway through the experiment, they were separated into two groups. This group was still fed cyclamate or saccharin, but also the cyclo- hexylamine was added to it. How can anybody in the public -- I am not speaking to anybody scientific now at all -- how can anybody in the public look at this demonstration and say that cyclamate caused all the problems, the few cancer things that were discovered? How can anybody say that? Now, I submit that the reason for putting cyclamate back on the market is not the reason that Mr. Wiegand wants. In spite of the tremendous amount of evidence, I think it should go back because it was wrong to take it off in the first place. The public has the right to know. All the other data are nice, but are really immate- rial. Cyclamate should go back on the market because it should never have been removed. PFAFFMANN: I think that Dr. Wiegand's forthcoming presentation and request for a new review would be suitable in relation to the history you have given.
ASPARTAME Alfred E. Harper I have been asked to talk about aspartame, a new type of sweetener that has been developed and tested by the G. D. Searle Company. It was ap- proved by the Food and Drug Administration in l974 for use in a number of food systems. I have been a consultant for Searle on some aspects of the development of the product. My objective is to tell you what aspartame is, what properties it has that make it a useful sweetening agent, how it is utilized or metab- olized by the body, what limitations it has as a sugar substitute, and how it has been examined to assure that consumption of foods sweetened with it will not pose a hazard to health (l,2,3). Aspartame is the methyl ester of the dipeptide, L-aspartyl-L-phenyl- alanine (Figure l). The constituents of aspartame occur naturally in foods. Aspartic acid and phenylalanine are amino acids that are pres- ent in all food proteins. Aspartic acid is a nutritionally dispensable amino acid. If it is not provided in a diet, the body can synthesize it from glucose and ammonia or from other amino acids. Phenylalanine is a nutritionally essential or indispensable amino acid; that is, the body cannot synthesize it. For normal body function, phenylalanine, like any other essential nutrient, must be provided preformed in the food. Methyl esters are common constituents of plant products, espe- cially of substances that impart flavors in fruits and vegetables, juices, and liquors. Aspartame is a white, odorless, crystalline powder that is soluble in water, more soluble in acid than in neutral solutions, and, as is usually the case, more so in warm than in cold solutions. Aspartame is about 200 times as sweet as sugar in taste tests. Its flavor has the characteristics of sugarlike sweetness, and it has no aftertaste. It also has the property of enhancing the flavors of certain foods, espe- cially those with fruit flavors. l82
l83 Aspartame is stable in dry form and can be stored in closed contain- ers at 40 degrees centigrade, that is l04 degrees Fahrenheit, for over one year with little deterioration, and the test is still continuing. When aspartame decomposes, it loses the methyl of the methyl ester, leaving the dipeptide L-aspartyl-L-phenylalanine (Figure 2). This has a tendency to lose water and cyclize to form the diketopiperazine of ASPARTAME H O 1 H H 1 ? 8 H2Nâ C â C- -Nâ Câ Câ 0- 1 CH2 CH0 C = O r VH f â CH. L-aspartyl-L-phenylalanine methyl ester (molecular weight 294.3) FIGURE l CH3OH H 0 H H O 1 I I T I H,NâCâCâHâC â CâOâCH, +H,0 Z i 1 o / fe 2 <JH2 » 6 Aspartame L-aspartyl-L-phenylalanine H 0 H H 0 H2Nâ C â Câ Nâ Câ Câ OH CH2 c=o 2°1l-H2° C-OH 1% Diketopiperazine FIGURE 2
l84 the dipeptide. This conversion occurs slowly in acidic solutions and much more rapidly in alkaline solutions. When sugar decomposes, the products formed initially are as sweet as sugar itself. Aspartame, however, loses its sweetness when it undergoes this type of decomposi- tion. In a solution at pH 4, that is about the acidity of root beer, stored at room temperature, 68 degrees Fahrenheit or 20 degrees centi- grade, 20 percent of the sweetness of aspartame would be lost after 4-l/2 months. In Table l are summarized some observations on the time for loss of 20 percent of sweetness in relation to the temperature of storage. Twenty percent of sweetness is about the amount of loss that becomes detectable in a comparative taste test. In a neutral solution, of course, even at room temperature, half of the sweetness would be lost in a matter of hours. Therefore, although aspartame can be sub- stituted for sugar and for tabletop use in dry products, such as sweet- ened powders for beverages and in puddings and fillings that do not require extensive cooking, it is not suitable for sweetening most alka- line or neutral products that require high-temperature baking, broiling, or frying; nor is it suitable for sweetening nonacid products in solu- tion that will be stored for long periods of time. Its stability in all foodstuffs has not been completely explored as yet, and there may be some interactions that increase its stability. Nevertheless, this means that aspartame is not a general substitute for sugar. Also, of course, it replaces only the sweetness and not the bulk, the preservative properties, or textural properties of sugar. TABLE l Effect of Temperature on the Time for Loss of 20% of the Sweetness of Aspartame in Acidic Solution (pH 4.0) Storage Temperature (°C)a Time for 20% Decomposition (days) l0 387 20 l34 30 5l 40 22 55 5 68 2 80 l a20°C = 68°F; 40°C = l04°F.
l85 On the other hand, because of the amount of aspartame required to impart sweetnesses is so small, it will have some unique uses for which sugar is not suitable, such as to sweeten foods that now cannot be sweetened with sugar because the large amount of sugar required alters some critical property. How does the body handle aspartame? As peptides and esters are nor- mal constituents of foods, one would expect the peptide and ester bonds of aspartame to be split by the enzymes, the peptidases and esterases of the digestive tract -- just as are the peptides and esters of food products -- and that these products would be absorbed and utilized just as they are when they are consumed as constituents of foods. The results of metabolic studies with isotopically labeled aspartame in rats, dogs, monkeys, and in man support this assumption. Unchanged aspartame was not detected in blood plasma after the administration of aspartame to subjects, but free aspartic acid and phenylalanine were. By using aspartame labeled with an isotope specifically in each one of its constituents -- the methyl group, the aspartic acid, and the phenyl- alanine -- the ultimate fate of each of these could be examined in ani- mals. The formation of carbon dioxide, the end product of oxidation of each of these substances in the body, followed essentially the same time course, whether the amino acids or the methanol were administered individually or whether they were administered in aspartame. Since aspartame is metabolized in the body in the same way as amino acids, unlike saccharin and cyclamates, it is a nutritive substance. Weight for weight, it should yield the same amount of energy as carbo- hydrates or proteins. Being a peptide, one would expect it to behave as proteins do and provide 4 kilocalories per gram. However, because it is required in only minute amounts to sweeten foods, with a sweeten- ing power of 200 times that of sugar, it will still contribute only negligibly to total energy intake when providing sweetness. Because it is a nutritive substance, aspartame has not been classified as an arti- ficial sweetener, so this means that its use will not be restricted to special dietary foods. As the diketopiperazine of L-aspartyl-L-phenylalanine can form during the storage and preparation of the product or of foods containing it, the metabolism of this product also has been examined. It is not high- ly soluble, and it is biologically rather inert. When it is injected directly into a vein, it is excreted unchanged in the urine. When it is fed to germ-free rats, it is not metabolized. However, from studies on animals with the usual intestinal flora, evidence was obtained that intestinal microorganisms can split the diketopiperazine to give aspar- tic acid and phenylalanine and some metabolites of these. The safety studies that have been conducted on aspartame and its diketopiperazine derivative were reviewed in a symposium held last November (3), so I shall mention only briefly the types of studies that have been done. It is important in assessing safety studies to remember that there is a toxic level for most, if not all, nutrients and other chemical
l86 compounds. Safety studies are not undertaken to determine whether or not a substance can induce an adverse or toxic reaction, but to deter- mine what amount is required to cause adverse effects or toxic reac- tions, and to ensure that the level of use will be well below what is found to produce such effects or reactions. In biological studies in which dosages of aspartame greatly in excess of the projected consumption levels were administered to animals (gram per kilogram of body weight quantities compared to projected intakes are in the order of milligram per kilogram of body weight quan- tities), no adverse effects were observed in the cardiovascular, gastro- intestinal, endocrine, reproductive, or central nervous systems. In rats, with extremely high doses of 4 grams per kilogram of body weight, mild behavioral changes and food intake depression occurred. Similar effects were observed after administration of comparable amounts of L-phenylalanine in the free form. In monkeys, no effects were observed with l gram per kilogram of body weight. Some monkeys receiving 3 grams per kilogram of body weight showed a type of seizure. This is about 300 times the anticipated intake level. Similar observations were made by these same investigators using phenylalanine at comparable levels. Both aspartame and the diketopiperazine have been tested for toxicity in the chronic and acute studies in several species of animals, again at levels greatly in excess of anticipated ingestion levels, levels of from 2 up to as high as 8 grams per kilogram of body weight, versus antici- pated ingestion levels of l0 to 20 milligrams per kilogram of body weight. Both have been tested for their ability to cause tumors, malformation of the fetus, and mutations. Studies have been conducted with human volunteers to assess the tolerance of normal, obese, and diabetic in- dividuals for aspartame. In none of these tests was evidence of adverse effects from either compound obtained except when the amount administered was sufficiently high to provide enough phenylalanine to retard growth. Adverse effects have long been known to occur in ani- mals consuming excessive amounts of several essential amino acids, among them phenylalanine. Some of these tests are continuing. What are the specific concerns with aspartame? It contains phenyl- alanine, which cannot be degraded by individuals with the genetic de- fect of phenylalanine metabolism, phenylketonuria. One child in l0,000 is born with this defect, about 400 infants a year. The disease is de- tected by a screening program that is required by law in all but seven states. If infants with this disease are to develop normally, their intake of phenylalanine must be restricted, just as diabetics, who make up probably l to 2 percent of the total population, must restrict their intake of sugar, and individuals with many types of kidney disease must restrict their intake of protein. Aspartame will therefore be labeled to indicate that it contains phenylalanine, so that if it is used in the diets of individuals with phenylketonuria, the amount consumed can be included as part of their allowed allotment of phenylalanine. One person in 70 carries the reces- sive gene for phenylketonuria. These people show no abnormalities, but
l87 may have greater than normal elevation of blood phenylalanine concen- tration after ingesting foods rich in phenylalanine. In tests that were done on a group of such people over a period of six weeks when they consumed as much as 8 grams of aspartame per day -- between l0 and 20 times the amount they would be likely to ingest, or 3 to 5 times the average anticipated daily intake in a single dose -- no abnormal reactions were noted, nor was there evidence of prolonged or unusual elevations of blood phenylalanine concentration. Another possible concern is with the aspartic acid provided by aspar- tame. Glutamic acid and aspartic acid administered in very large single doses, l gram or more per kilogram of body weight, to newborn rodents will produce lesions in the hypothalamus, an area of the brain, that will result in obesity. In lifetime studies in rats in which up to 400 times the anticipated use level of aspartame, about 8 grams per kilogram of body weight per day, was administered to pregnant females in their diet, and then sub- sequently to their offspring in a lifetime study, although the brains were not examined in this study, no physical abnormalities were observed in the offspring. Doses of aspartame in amounts found to cause brain lesions in newborn mice were not found to cause such lesions in neo- natal monkeys, even when combined with glutamate. In human studies in which subjects were administered 34 milligrams of aspartame per kilo- gram of body weight, no elevation of serum aspartic acid was observed. A pint of milk in a single feeding would provide about 6 grams of aspar- tic acid and glutamic acid. What about the probable consumption of aspartame? Average sugar consumption generally, as we have discussed several times, is between l00 and l50 grams a day, so not more than 0.5 to 0.8 grams of aspartame would be required to provide sweetness equivalent to all of this. Be- cause aspartame is not suitable as a replacement for sugar in some foods, such as products that are neutral in reaction and must be baked at high temperature or liquids that are neutral in reaction and have to be stored, average consumption of aspartame should not exceed half a gram per day. This would provide 280 milligrams of phenylalanine, 226 milligrams of aspartic acid and 54 milligrams of methanol. A six-ounce glass of milk would provide more phenylalanine and aspartic acid than this, and three ounces of beef would provide considerably more, as would wheat (Table 2). An eight-ounce glass of fruit or vegetable juice would provide somewhat more of the methyl esters than the aspartame, and a double martini would provide a great deal more methyl esters than aspartame. This amount of aspartame would represent about 5 percent of the average daily intake of phenylalanine, and less than that of the average intake of aspartic acid. Some individuals will, in all probability, consume more than the estimated intake, and a few undoubtedly will consume much more. But as the dose level that was without effect in the safety studies was about 200 times the estimated consumption of l0 milligrams per kilogram of body weight per day, it will be difficult to conceive of an individual consuming enough aspartame to cause any adverse effects.
l88 TABLE 2 Probable Consumption of Amino Acids from Aspartame Phe Asp (mg) -OCH3 (mg) (mg) Aspartame (0.5 g) 280 226 54 Beef (3 oz) 653 l336 -- Wheat (3 oz) 490 54l -- Milk (6 oz) 3l0 443 â Tomato juice (8 oz) 37 Gin (3 oz) -- -- 90 In summary, aspartame is an odorless, crystalline compound made up of substances that occur naturally in foods. It is about 200 times as sweet as sugar in taste tests, and has been tested extensively for safety in biological, toxicological, and other trials without effects from l00 to 200 times the estimated level of consumption.
ANOTHER VIEW OF ASPARTAME John W. Olney I think I should first clarify my relationship to the topic of this Forum. I have a long-standing interest in the developing central ner- vous system and in toxic mechanisms that might adversely affect the immature brain so as to give rise later in life to neurological or be- havioral disturbances. Several years ago, I reported (l) that the widely used flavor additive monosodium glutamate (MSG) destroys nerve cells in the brain of experimental animals, particularly young animals, when given orally in relatively low doses (2). The relationship between glutamate-induced brain damage and aspartame, the sweetener Dr. Harper has just described, is that one of the major moieties of the aspartame molecule, aspartate, has the same type of brain-damaging potential that glutamate has, and when glutamate and aspartate are administered together they act in concert by an addi- tive toxic mechanism to destroy brain cells (2). Furthermore, evidence generated in my laboratory in St. Louis and submitted recently to the Food and Drug Administration clearly demonstrates that aspartame itself, when administered by feeding tube to young mice, causes the same type of brain damage that glutamate or aspartate causes. We think the aspartate moiety of aspartame is responsible for its brain damaging ac- tivity and that it is because aspartate resembles glutamate in molecular structure and excites neurons as does glutamate that it shares gluta- mate 's neurotoxic properties. These two compounds, in fact, belong to a family of neuroexcitatory toxins or excitotoxic amino acids, as we have come to designate them. All of the members of this family of excitotoxins (Figure l) are structural analogues of glutamic acid (top center) or monosodium gluta- mate, which is the popular name for the sodium salt. It is of interest that several of these compounds, in addition to having excitatory and l89
l90 ^CH-COOH CH-COOH 0^ N-CH2-CH-COOH CH2-CH2-COOH CH-COOH Straight Chain Analogues Heterocyclic Analogues CH2-COOH CHj-SOjH CH2 CH-COOH CH-COOH H3C-Ci pCH2-COOH H2N H2N I LcOOH ASPARTIC ACID CYSTEIC ACID KAIHIC ACIl OVOV50^ ox H2N N-NtTHYL-ASPARTIC ACIl) HOMDCYSTEIC ACID QUISQUALIC ACID NH-CO-COOH 1 /ox CH, CH,-S-SO,H y \_ CH-COOH 1 1 II If I CH-COOH CH-COOH I! U NH, / / »0 H2N I12N ODAP CYSTEINE-S-SULFONIC ACID IBOTEHIC ACID FIGURE l Excitotoxic structural analogues of glutamic acid. toxic effects on central neurons, also mimic glutamate in stimulating taste receptors. The two columns on the left depict straight-chain analogues that have excitotoxic properties. Some of these molecules are more potent than either glutamic or aspartic acids in exciting and destroying nerve cells. For example, tf-methyl-DL-aspartic acid, al- though differing only slightly from aspartic acid in molecular struc- ture, is l00 times more powerful in excitotoxic activity (3). 3-tf- oxalyl-L-a,B-diamino-propionic acid (ODAP) is a straight-chain gluta- mate analogue that is found naturally in the chick-pea and is thought to be the neurotoxic agent responsible for neurolathyrism, a serious neurodegenerative condition occurring endemically in regions of the world, such as India, where chick-peas sometimes comprise a high per- centage of the diet. Other straight-chain analogues such as homocysteic acid and cysteine-5-sulfonic acid are of interest for their possible involvement in the pathogenesis of mental retardation syndromes in human metabolic disorders such as homocystinuria and sulfite oxidase deficien- cy (4,5). The molecules depicted on the right are heterocyclic analogues of glutamate. Kainic acid, which was recently shown to produce the gluta- mate type of brain damage (6) is about 200 times more potent than
l9l glutamate in excitotoxic activity. This compound is found naturally in seaweed and has been used as an ascaricide in Japan to rid children of intestinal worms (7). Ibotenic acid is an interesting analogue of glu- tamate found in nature as the poisonous principle in the amanita mush- room. It is a potent neuroexcitant and is about 20 times more potent than glutamate in stimulating taste receptors. This has led one of the larger manufacturers of monosodium glutamate to consider developing this compound as a food-flavoring agent. Quisqualic acid, which occurs in the seeds of quisqualis indicus, is considered about 500 times more powerful than glutamate in neuroexcitatory activity. Nothing is yet known about the neurotoxic properties of this compound as it was only very recently discovered. I have briefly described this group of excitotoxic amino acids, some of which may arise exogenously, others endogenously, to point out that the hazards of aspartame cannot be fathomed fully by merely concentrat- ing on aspartame itself. We should be concerned about how it or its metabolites may interact with other excitotoxins in the body. Two of these excitotoxins are approved food additives (aspartame and monosodi- um glutamate), a third is under consideration as a flavorer (ibotenic acid), a fourth is a potential pharmaceutical (kainic acid), a fifth is found in the chick-pea, a commonly ingested vegetable, others might be generated endogenously in unidentified human hosts with metabolic dis- orders, and still others have yet to be discovered or identified as excitotoxins. It is in context with the above and with the fact that glutamate is an additive in extremely widespread use throughout our food supply, in- cluding the food supply for our young, that I have expressed concern over the introduction of aspartame into that food supply. When the Food and Drug Administration approved aspartame in July l974, I objected in a memorandum to the FDA Commissioner (August l6, l974) that it was a premature action since the combined toxicity of aspartame with gluta- mate or other excitotoxic amino acids had not been studied nor had the neurotoxicity of aspartame itself been tested appropriately on immature animals even though immature humans appeared to be a major consumer target projected for it. FDA has taken these objections under consid- eration and has expressed its intent to convene a public board of in- quiry to review the matter sometime in the near future. At this point I would like to say a word about margins of safety. Depending on one's assumptions regarding use levels, no-effect dose levels, and the age of the consumer concerned, one can come up with quite a wide range of margin of safety calculations. The best way I know to present the other side of the picture from the one that Dr. Harper just presented regarding the safety of aspartame is to begin with glutamate, which must share with aspartame a single margin of safety if they have combined toxicity. Going back to l969, when glutamate was being added rather freely to baby foods, a subcommittee of the National Academy of Sciences met to investigate the practice and established that the highest concentration of MSG being added to baby foods was 0.6 percent. This means that a
l92 4-l/2 oz (l30 g) jar of such baby food provided a 6 kg infant with 0.l3 g of added MSG/kg of body weight. The amount of glutamate that from a single oral load will cause irreversible destruction of nerve cells in the hypothalamus of the immature mouse is 0.5 g/kg. The difference be- tween 0.l3 and 0.5 g/kg, i.e., the margin of safety, is not the l00- fold margin we often hear about, rather it is about 4-fold. From studies on older animals, one finds that the minimal effective dose goes up, but not sharply. For example, in the 2l-day-old mouse at the age of weaning, it requires about l g/kg by oral intubation to pro- duce the brain lesion, and at 45 days of age, which is roughly puberty in the rodent, it requires about l.5 g/kg. The margin of safety, then, just for glutamate alone -- if one can make extrapolations from animals to man, and I fear we have to, because it is neither safe nor ethical to do such experiments in the human -- may increase with age to perhaps an upper limit of between l0- and 20-fold. If we then add aspartame to children's foods, no matter how safe it may seem from experiments not designed to reveal its toxicity for the immature nervous system, I am afraid it will add to the neurotoxicity of glutamate, which means that it will reduce glutamate's margin of safety, a margin already too slim to begin with. Before leaving the issue of aspartame's safety margin, I would like to point out a flaw in FDA's safety evaluation of this sweetener. In approving aspartame for general use, FDA represented that this sweetener would have nearly a l00-fold margin of safety when used as approved. This margin was arrived at by using the body weight of a 60 kg adult human and applying no-effect dose data pertaining to adult animals; in other words, adult referents were used exclusively even though approval was given for the sweetener to be used in children's foods. This is highly inappropriate; as I emphasized in my memorandum to the FDA Commissioner, it is absolutely essential that the child's body weight and no-effect dose data for immature animals be used in evaluating the safety of any additive which will be fed to children. Even if the pic- ture were not complicated by MSG or other excitotoxins, the margin of safety for the consumption of aspartame by children (if calculated from appropriate referents) is nowhere near the l00-fold level. At the time a food additive is being approved it is imperative that the special vulnerabilities of the immature human be figured into margin of safety calculations because experience tells us that after approval is given, the product will be promoted and marketed indiscriminately for consump- tion by the mature and immature alike. In closing I should point out that only the risk aspect of the risk- benefit comparison has been focused upon here. In the absence of evi- dence for real benefits, i.e., that it contributes significantly to the health needs of children or infants to have either monosodium glutamate or aspartame added to their foods, I am at loss to understand why any- one would favor exposing vulnerable young consumers chronically during their developmental years to diets heavily supplemented with these excitotoxic compounds.
l93 DISCUSSION PFAFFMANN: Dr. Harper, do you wish to reply? HARPER: Yes. There are obviously several ways of approaching this question. Certainly the rodent is the species that is most suscep- tible to this type of damage. Newman et al. (4) have reported admin- istering 4 grams of glutamic acid per kilogram of body weight to monkeys without any evidence of hypothalamic lesion occurring. There have been studies in the Searle safety program with aspar- tame together with glutamic acid, 2 grams and l gram respectively of each, being administered directly to monkeys without evidence of lesions developing in the hypothalamus. There is another point that I think Dr. Olney overlooked. In order to produce these lesions, one has to administer the substance by stomach tube within an extremely short period of time. This has to be done to elevate the blood levels of glutamic acid or aspartic acid or cysteic acid to the very high levels required to produce lesions in the hypothalamic area. Frequently this is done by in- jecting a single dose directly into the animal. In the safety tests, no abnormalities were observed in rodents that were administered aspartame up to 8 grams per kilogram per day with their normal diet. In other studies no elevation in blood aspartic acid concentration was observed after something on the order of five times the anticipated daily intake level was adminis- tered in a single dose to human subjects. It is important to keep in perspective how we assess a hazard. We know that if we administer nutrients such as iron, vitamin A, vitamin D to people in huge doses in unique ways, we can develop severe toxic signs. We can administer most of the amino acids in similar ways and produce severe toxicity. We can administer lactose, which is milk sugar that is consumed by children all the time, to rats and produce severe cataracts. We can produce toxic lesions with almost anything if we set about it the right way, and we are interested in being able to do this because we want to know at what levels of intake such effects occur. After that we have to assess the probability of a hazard occurring if the substance is used in quite a different way in a diet, or as a drug or a pharmaceutical. OLNEY: I think there is a misunderstanding about primate susceptibil- ity, and I would like to respond to Dr. Harper's statement. We have performed extensive studies on rhesus monkeys and have demonstrated quite clearly that glutamate given either orally or subcutaneously damages the monkey hypothalamus. All of the monkeys in our series that received glutamate sustained brain damage, whereas our sodium chloride controls were unaffected. The oral experiments involved administering either glutamate or sodium chloride in skim milk through a naso-gastric tube, and the dose of glutamate that produced the lesions by this oral route in the seven-day-old monkey was l
l94 gram per kilo. I do not know what negative study from Japan Dr. Harper refers to. I think Ajinomoto (MSG) Co. in Japan has sponsored some such studies, but I have not noticed their data appearing in any reliable neuropathology or brain research journals. Our findings were reviewed by the most respected editorial board of neuropathologists in the world and are available in any medical library in the form of a 24-page paper with over 30 high-quality photomicrographic illustrations (8). As to the question of gavage being an inappropriate method of ad- ministration, I have heard that complaint made before, especially with reference to my use of gavage to study toxicity in infant ani- mals. The first thing I must point out is that if you want to have excellent control over dosage so that you can make statements about how much the animal really received into the gastrointestinal tract, gastric intubation is absolutely the preferred approach. Secondly, the way we feed human infants in this culture is essen- tially by gavage. We do not leave them free to roam around their cage (home) to nibble ad lib on food throughout the 24 hours of the day. That is the way rats do, but that is not the way humans â human infants, at least -- feed. At a designated feeding time a human infant is fed as rapidly as the food can be spooned into his mouth (additives included): first a jar of processed meat and vege- tables, then perhaps a jar of sweet dessert, and then he is laid to rest with a milk bottle plugged into his mouth. The plan is to fill his stomach to capacity within as short a period of time as possible. Now that is essentially gavage. HARPER: Well, I question whether a baby can drink l2 ounces, l6 ounces, 32 ounces of milk as a gavage. Also, it is interesting to note that 46 percent of the protein in cereals consists of glutamic and aspar- tic acids. About 30 percent of the proteins of meat products is aspartic acid; so if we eat a steak we are probably getting a gavage of glutamic and aspartic acids, too. OLNEY: Yes, but it takes quite some time for the protein to be digested, and the aspartic and glutamic acids ingested as protein are going to be dribbled into the bloodstream over several hours, period of time. Actually, this represents a minor safety factor working in aspartame's favor. Being a dipeptide, it will require a brief digestive process to make the aspartate available for absorp- tion. HARPER: That is right. And I think it is important to note that the gavage technique is effective in producing lesions, because you can overload the stomach tremendously by gavage and cause rapid emptying. If a substance is drunk as a suspension, its entry into the intes- tines is regulated by the rate of stomach-emptying. The stomach is a very important regulatory organ unless it is tremendously over- loaded, and one can overload it by instilling into it large volumes of solution.
l95 OLNEY: I agree with that, although, overloading animals may merely result in defecation rather than absorption of the test material. I do have another concern, though, and that is that the immature human may be dependent for protection from glutamate and aspartate on a transamination mechanism in the gastrointestinal tract which can handle limited loads of these amino acids by transaminating them to alanine. We do not know very much about that in young humans, and above all, we do not know how many humans are deficient in that enzyme system or how easily it might be overloaded when both the ordinary amounts of aspartate and glutamate in the diet are joined by additional free glutamate and aspartame being added in gram quan- tities to that diet. Again by adding more and more of these amino acids on top of what is normally in the diet we may be creating a human gavage situation. HARPER: There have been studies on absorption of amino acids in man, using a double lumen tube. These show, as in the rodent, that glu- tamic and aspartic acids are the most slowly absorbed of all the amino acids and that their concentrations in plasma do not rise in response to a load as readily as do most other amino acids (5). PFAFFMANN: This is an important issue, and there is an obvious differ- ence of opinion on the platform here. As I understand, there will be a hearing involved at which some of this and,other evidence also will be brought forward. As far as we are concerned, we have had the issue presented. The resolution is not going to take place here today from what I have heard of the discussion.
MONELLIN Morley R. Kare About 40 years ago, a report (l) suggested that if there were too much pressure in the womb, one could inject a little saccharin and the fetus would be encouraged to consume the amniotic fluid, thus reducing the pressure. This work inferred that at seven months the fetus will re- spond to sweet stimulation. It follows that at birth a baby might have a well-developed sense of taste. In studies in our laboratory, babies one day to three days of age were tested with a variety of sugars and other taste stimulants (2). Typically, they were offered either water or sugar solution midway between their scheduled feedings. They re- sponded to sucrose at concentrations roughly equivalent to what might be meaningful in the adult. It was concluded that not only will newborns respond to sweet, but they will discriminate among sugars. It is interesting that milk sugar (lactose) is not particularly effective as a taste stimulant. This suggests that it is not chemical imprinting or early experience with milk that develops this sweet taste. In calves, where it is clear that they get lactose and nothing much else in the way of sugars early in life, the story is similar. In a choice situation, the young calf will double its fluid intake, selecting a l percent sucrose solution almost l00 percent of the time (3). Obvi- ously, the drive for sweet stimulation, or its equivalent in animals, can occur independently of early experience. Incidentally, most humans find l-percent sucrose insipid, or even offensive. The suggestion has been made that taste buds must be kept in fighting trim to respond to sweets, that is, a continuous exposure to sweets is necessary to maintain the drive for them. The first time I heard this assertion contradicted it was by Sir Stanton Hicks, who had worked with the aborigines in Australia about 50 years ago. He said that he could get the individuals living in primitive isolation to do just about l96
l97 anything by giving them some sweets. I have heard anecdotally that if you can still find a middle-aged Eskimo who has not been exposed to Western culture, and offer him sweets, he will respond immediately. Apparently, having soda pop and candy bars continuously through life is not necessary to keep a sweet tooth functioning. The taste receptors will work and respond at any age to the initial exposure to a sweet stimulation. 16 l4" a ⢠Z 8 6 4- 2 Wattr .00 .10 .20 .30 Concentration of Sugar Solution (M) FIGURE l Mean volume of sugar solutions and water in- gested by infants offered dif- ferent concentrations of sugar solutions. From Desor et al., reference 2. FIGURE 2 Mean volume of sugar solutions and water (W) ingested by infants offered different sugars. From Desor Gluco»e W FructowW Lactose W SucroteW et al., reference 2.
l98 CH1LDPCN N = 618 K60] 1 1 ADULTS h! 50- 7 JO 1 1 N = 140 in 1 <J S 40- & 30- 3 | - "> 30- £ ^ ^ I PERCENT â IV) 0 0 â t \ 3 J.N33U3J (VI - 6 ' 6 ' j w. R I i ] n i i .075 .15 .30 .60 SUCROSE CONCENTRAT1ON(M) 05 10 - ZQ 40 NoCl CONCENTRAT1ON ( M) FIGURE 3 The adults responded almost equally to the four sucrose concentrations. However, approxi- mately half of the adolescents (9-l5 years) pre- ferred the highest concentration. From Desor et al., reference 4. All taste buds in all people do not respond to sugar in the same manner. There is an enormous variation among individuals. Over 99 per- cent of humans will respond to sucrose at some concentration. What is more, investigators in our laboratory have found that while there is a great deal of variation between individuals, the response is constant for an individual over a period of time. Does the response to sweet observed in the newborn change in the adolescent? A study on sweet and salt perception was carried out employ- ing 6l8 adolescents and l40 adults as controls (4). Sucrose was offered at four concentrations. The adult population responded, in terms of preference, about the same at all concentrations. However, the adoles- cent responds preferentially to higher concentrations of sweet. This would be in agreement with the data that Dr. Cantor gave you yesterday in terms of consumption in the population. This preference for sweet was independent of socioeconomic background. However, it is interesting that black adolescents responded to the sweet at higher concentrations than did the white adolescents. As we get older, the number of taste buds goes down. A study by Arey (5) in an age group 74 to 85 indicates there is a 60-percent loss of taste buds, and of those that are left, only 50 percent are functional. In this age group, therefore, approximately 20 percent of the taste buds are functional. To assess the changes in taste preferences, if any, with age, we evaluated three age groups: 40 to 45, the controls; 65 to 70, our mature group; and 80 to 85, our aged group.
l99 BLACK .075 .15 .30 .60 SUCROSE CONCENTRAT1ON (M) .05 .10 .20 .40 NoC1 CONCENTRAT1ON ( FIGURE 4 The response to the highest concentration of sweet was more evident in the black adolescent than in the white. From Desor et al., reference 4. 30r MALE N = Z4 FEMALE 25- >- £ 20' 5 Uj 15- in ki S iO- ct 10 5- 37-45 65-70 80-85 AGE FIGURE 5 There is no sig- nificant difference in sensitivity to sucrose be- tween the most elderly and the controls. There are many tissue changes in the oral cavity with age. Limiting myself here to reporting on the response to sugar, sucrose, there was no significant loss in sensitivity, and absolutely no loss in the pref- erence for sucrose. The testing procedure we use requires about 30 to 45 minutes per individual. In addition to the threshold testing, we employed some practical tests with commercial products prepared with
200 different levels of sugar. The subjects 80 to 85 years of age could dis- criminate the sugar content differences equally as well as the controls. One can expect vision problems at 40. Diminished hearing capability can be predicted a little later. It is good to know that your taste preferences will probably be with you all your days. A question has come up repeatedly in the last two days about what function taste serves in the body. I cannot answer all aspects of this question here. I will limit myself to one of the best examples of the function of the sense of taste in the human. Some of the discussion here has focused on nutrition. Nutrition con- sists of more than a precise balance of required foods. The nutritive process begins in the mouth with taste stimulation. Glucose taken by mouth will evoke an insulin response, which occurs before the circulat- ing hyperglycema (6). The same glucose by tube would not evoke a parallel effect. Therefore, oral stimulation with a sweetener can in- fluence circulating hormones. If you pop a candy in your mouth, you know that saliva is secreted. There is no question that the amount of sucrose in food will affect the way you chew the material and also the way you swallow it. But more important, there are many things that you are not aware of. Oral stimu- lation can influence contraction of your stomach. Strong oral stimula- tion can affect the motility of the intestine. Working with dogs (7), we placed chronic fistulas in the stomach and in the intestine. The dogs become familiar with laboratory routine and are relaxed when tested. If clay is applied to their tongues, nothing FIGURE 6 Position of gastric and duodenal cannulae. From Kare, reference 7.
20l ,---COMMON BILE DUCT FIGURE 7 Position of intestinal cannulae in relation to pancreatic duct. From Kare, reference 7. happens to pancreatic flow. If you put some quinine on their tongues (dogs are offended by quinine), nothing will happen. However, if a little sucrose is placed on their tongues, the volume of pancreatic flow goes up and the protein content of the secretions increases. If you put lard, which they like, on their tongues, a similar situation occurs. In summary, an unpleasant taste stimulus had a significantly different effect on the character and volume of pancreatic flow than did a pleasant stimulus. Apparently, pleasant taste stimuli will affect the activity and the secretions along the digestive tract. Currently, Dr. Naim is working in our laboratory identifying the individual enzymes that are affected by oral stimulation, and also with olfactory stimulation. I can only interpret the pancreatic effect in humans in terms of in- formation done by others, not directly related to this research. Before Hollander could operate on children with an obstructed esophagus, he was temporarily feeding them by means of a fistula. When he placed food directly into the digestive system, he found that he had to increase the amount of food considerably to maintain them. If he permitted the chil- dren to chew the food first and then introduced it into the fistula, they could be maintained on a normal amount of food. At our laboratory, Kemper, a pediatrician, was working with infants born with malformed mouth parts, often complicated by a cleft palate. They are commonly tube-fed. Kemper fed these babies spoonful by spoon- ful. It took him considerable time, but apparently oral stimulation considerably improved the prognosis in these infants. This suggests that taste is important, particularly in the young. It has been publicly stated that it is unimportant to flavor baby food, that flavor is there to sell the mother. However, the evidence would indicate that a good case can be made that it is important that the taste of food, particu- larly for very young children, be made appealing to them. There may be room for that sixty-third product Ms. Gussow mentioned. There are some Victorian ethics that have spilled over into the food field: the idea
202 that if it is pleasant, it must be sinful. The case has been presented that the pleasantness of food can have physiological functions. I will move now to the subject that I was specifically invited to speak on, that is, the sweetener known as monellin. Monellin is purified from the fruit of Dioscoreophyllum cwnminsii. This is related to the sweet potato, but of course in the sweet potato you eat the root. The berries are about the size of small grapes. After removing the skin, there is a white jelly and a bitter seed. Morris and Cagan (8) in our laboratory isolated the active material of this sweet jelly, and one pound of the pure material was equivalent in sweetness to approximately a ton of sugar. On a molecular basis, it is 80,000 times as sweet as sucrose; that is, it is intensely sweet compared to sucrose. Monellin is a pure protein, completely free of carbohydrate. As long as sucrose was a few cents a pound, nobody was looking for sweet proteins. This protein has a molecular weight of l0,700. It has 9l amino acids. The only thing unique about the amino acids is that there is no histidine. Cagan (9) and associates have discovered many things about this protein. The tertiary structure is critical for the sweet- ness; that is, the three-dimensional nature of the molecule is important. If they denature the protein, the sweetness is lost. If it is revers- ible, the sweetness will return. 15 cm. Dioscoreophyllum cumminsii FIGURE 8 Monellin, a pure protein sweetener, is isolated from the berry of this fruit.
203 We are using monellin as a biological model for research. One of the things that is exiciting is that a big molecule like this probably cannot penetrate the taste cell, so you get the sweet sensation with nothing passing into the cell. Our scientists are isolating taste cells in a test tube and studying how a sweet material acts on the surface to produce a sweet sensation. Perhaps the people in the sugar industry here will take issue, but su- crose and the simple sugars are really relatively poor sweeteners. I say this in the sense that saccharin or monellin are so much more effective at the receptor molecules. There will be dozens of new sweeteners com- ing along as we understand the mechanism of sweet. With sweet stimu- lants like sucrose, theoretically, a modifier could facilitate access to the taste cell membrane. The likelihood of increasing the sweetening effect of sucrose is reasonable. Reference by many speakers has been made to the sense of taste in animals. Monellin has not been effective in the animals we have tested it with. There are many animals that do not have a sweet tooth. Chick- ens do not have a sweet tooth -- they have no teeth. Cats, armadillos, some fish, and cowbirds do not respond to sugar. Sea lions, I under- stand, and even whales are in this category. There are many species, however, that do respond to sugar. This response varies. The one sugar that chickens respond to is xylose, which it rejects. Cows love xylose. It is one of the most pre- ferred sugars. But cows do not respond to maltose. Maltose is the most preferred sugar of rats. And rats do not respond particularly well to lactose, but possums love it. I would like to make the point that with synthetic sweeteners, nobody mentioned that most fail to evoke a positive response in animals. At high concentrations, I would guess, they are offensive to many species. Whether or not they produce a pleasant sensation can be meaningful in toxicity evaluations. It can be significant if one were to administer to a level where you are completely distorting the physiological effect that you are expecting from that compound in man. I think it is critical that all of the sweet compounds that are coming along, if they are tested in animals for toxicity, be administered at levels pleasant to the animals. Certainly, in terms of their effect on enzymes, hormones, and digestive tract activity, we should try to have a behavioral effect similar to that encountered in man. One point that summarizes much of what I have been saying to you is this: taste has been criticized here in the last two days for being a vehicle of poor nutrition. It is just as easy to use taste as a vehicle for good nutrition. DISCUSSION PFAFFMANN: I am pleased with the emphasis that Dr. Kare brought to this meeting in which the functionality of the sense of taste has rather
204 taken a back seat. I also cannot refrain from the usual interchange we have when we appear on the same platform. Dr. Kare made the point that there are many organisms that do not have a sweet tooth like the human, and in fact, there is no denying the evidence. The point that I always like to stress, however, is the remarkable fact that there is such a widespread occurrence of the sweet tooth in the animal kingdom. He is interested in the differences; I am interested in the similarities with the human case. The particular point, however, of the different effects of syn- thetic sweeteners and taste modifiers in different species is very well established. Treating sensitivity across all organisms as if it were the same would be a great mistake. But on the other hand, there are instances where a test or model species will have great similar- ities to man. Therefore these can be used in experiments that are not suitable for the human, such as the tracing of the brain pathways and the analysis of the hypothalamic motivational mechanisms. We can rely very heavily on those instances where there is a demonstrated similarity. We have a final session coming up. Perhaps we ought to ask now if there are comments or any points that the Forum wishes to bring up. CHOATE: Dr. Kare, I notice you brought up the reactions to varying increments of sucrose in a rather bland mix as people change by age and by race. Do you know of any studies in which the amount of su- crose has been increased to the point that there has been a decrease in interest in the sweetness? Are there any studies that would es- tablish at what percentage level of sucrose the tongue of the average citizen was unable to detect increased sucrose in a mix? Do you know of any studies which reflect on this by age group? KARE: I invite you, Mr. Choate, to come and visit Monell Center to observe testing people for the level of sensitivity and for preference level. They test up to a concentration where the sweetener becomes nonpreferred. KRAYBILL: I believe, if I remember your charts correctly, you showed the difference in sex on sensitivity. That is the first question: Do you have any explanation for it? The other question is: In terms of this molecule, do you know the structure or do you know the link- up with the various peptides? Do you have any ideas about this? KARE: First, let me answer the one on sex. If I can I will use salt to illustrate the answer because it is some work I did myself. In a group 80 to 85 years of age, the difference between the response is dramatic. The sensitivity threshold concentration where it is first detected in the male goes up. However, it does not go up at all in the female.
205 I could not give you the reason why in the adolescent there is a sex difference. Perhaps it reflects the greater need for calories in the male. On the structure of monellin and the nature of the molecule, rather than take up time here, I will refer you to a series of papers by Cagan and coworkers. (See references.) PIETZ: From studies in physiological psychology, I understand, Dr. Kare, that children have a much more sensitive sense of taste, particularly the infant, than does the adult. Do you agree with that from any studies that you have done? KARE: No, that is not true. I did not go into the modalities other than the sweet one because of the limitation on time. The newborn baby does not respond well to bitter or salty stimuli. The response to sucrose is slightly less than that encountered in the adult. So I would say the newborn1s sense of taste is not as well developed as that of the adult, if we consider all the modalities. PIETZ: Then you would tend to see sucrose being added to baby food as relatively a good thing for getting the baby to eat? KARE: I work on taste, the physiology of taste, not nutrition per se. I would not isolate one compound from the entire baby's diet. It is my feeling, on the basis of everything we have seen, that it is im- portant that the infant's food be highly appealing. As soon as any animal, including the human, is put under stress, food intake drops down; if you make it highly appealing, food intake will go back to the original level. You can see this phenomenon in any weanling animal. When a baby pig is removed from its mother and placed on commercial starter, it is made appealing so that the food intake will be maintained at the normal level. If you place a baby, or a young animal, in a strange environment, food intake drops off. If you make the food highly appealing, in- take rises. So there seems to be reason to make infant food highly appealing. PIETZ: All right. One parting question. Is part of the problem with getting youngsters to eat vegetables -- particularly vegetables that have something of a sour or a bitter taste -- a problem with their taste? Is the young child more sensitive to these tastes than the adult, so that even if he tends not to like these foods now, later as an adult he might learn to like them? KARE: I am lucky I am a professor. I do not know the answer to that, and I can afford to say so.
206 PATRICIA HAUSMAN, Center for Science in the Public Interest: I think that Dr. Kare is implying that carrots and beans and rice do not taste good, and that we have to add sugar to them to make them taste good and to make them appealing. PFAFFMANN: He did not say that. HAUSMAN: I think that baby food without sugar would certainly be ap- pealing enough to infants. I am also familiar with the work of Dr. Desor, and I notice in one of her articles that infants consume more of sweetened foods. I do not necessarily think that this is good with the rate of obesity and overeating in this country. I do not think we need to encourage overconsumption of foods; and I do think that foods minus sucrose are certainly acceptable. KARE: It is a very interesting comment you have, but I do not know how it relates to what I had to say. For all I know, some of the sensory qualities in carrots, by themselves, are very appealing, and may be uniquely appealing. Nobody has ever evaluated that, and to interpo- late it the way you did, it is difficult for me to challenge. What I said is that sensory stimulation, in terms of secretions along the gut, in terms of activity, seems to be as important or more important in the very young animal or the young child than in the adult. The decision as to whether or not you add sugar should con- sider the physiological functions served by taste stimulation. HAUSMAN: It seemed to me that you were saying that foods have to be highly appealing in order to be eaten, and I think that people eat because they are hungry, and that they do not eat highly sweetened foods. Also it was said that there may be room for a sixty-third product, referring to Joan Gussow's remark about cereals, and in the next breath you said that sucrose is a lousy sweetener. Considering that many of the children's cereals on the market are 40 to 50 percent sucrose, I think that there really is not room for the sixty-third product.
THE SEARCH FOR NEW SWEETENERS George E. Inglett Among the sweetener issues and uncertainties that are to be discussed, I think we should consider some of the new sweeteners as potential alternatives. In the formulation of food products, a variety of nutritive as well as nonnutritive sweeteners afford the best opportunities for providing consumers with excellent foods. A food product must be sold; otherwise you will not find it in the marketplace. In order to have successful food products, there are certain sensory and functional properties that sweeteners must fulfill. A variety of sweeteners, in my estimation, is the best approach to this problem. We need saccharin, cyclamate, aspar- tame, and others, particularly to meet the food needs of the diabetic and the diet-conscious. I will cover a few of the other sweeteners that you have not heard much about. One of them is stevioside. This occurs in a plant that grows in Paraguay and surrounding areas of Brazil and Argentina. It occurs to the extent of about 6 percent in the leaves of the plant known as Stevia rebau.di.ana Bertoni. The Indians use this plant material to sweeten their tea and other bitter foods. Another substance -- and it happens to be the sweetest material on the GRAS list -- is known as glycyrrhizin, better known by most people as licorice extract. It is widely used to flavor candy, tobacco, and Pharmaceuticals. There are other sweeteners on the horizon. Some new ones are ob- tained from certain citrus flavonoids. These sweeteners were discovered in the early sixties by Horowitz and Gentili, who hydrogenated naringin and neohesperidin to give unexpected sweet dihydrochalcones. Another compound, osladin, has been discovered that probably does not have any potential as a sweetener, but yet it is a naturally occurring 207
208 sweet substance. Just because it is naturally occurring does not mean that it could not be toxic. We all know that there are some lethal compounds that do occur in nature. Osladin was discovered by some Yugoslavian workers, and I think it has some theoretical interest based on its structural relationship to other sweeteners I have mentioned in having a (l-*2) linked disaccharide moiety as a glycoside attached to another portion of the molecule. Some related compounds are being syn- thesized at the Northern Regional Research Laboratory in an attempt to associate certain types of structures to the sweet-taste phenomena. In preparing a few words on the intensely sweet wild fruits and berries, I was struck with a bit of nostalgia. It was twelve years ago that I was employed by the International Minerals and Chemical Corpora- tion to work on the "miracle fruit," Synsepalum dulcificim. At the time we discussed the project with Dr. Beidler, Dr. Pfaffmann, Dr. Kare, and others. Dr. Beidler discussed miracle fruit earlier in this Forum. It is the remarkable fruit that causes sour things to taste sweet. That is, after you have eaten one of these berries, you can eat an ordinarily sour lemon, and it will taste like sweet lemonade. As you heard earlier from Dr. Beidler, the FDA refused GRAS status for miracle fruit concen- trate. After a few years of working on miracle fruit, we had problems at International Minerals and Chemical Corporation. In an attempt to solve some of these problems, I found some sweet berries and went back into the laboratory to put them in water to filter the material, and found the extract to be exceedingly sweet, similar to a saccharin solution. I called the sweet berries serendipity. The serendipity berries are what Dr. Kare discussed as the Dioscoreophyllum cwnminsii, which gave them monellin. At the time we did not know the botanical name for the berries, but within several months I found a botanist in Sierra Leone who knew the berry to be Dioscoreophyllum cunrninsii. We did a lot of work that has never been published, and I am sure never will be, on botanical searching in Nigeria and certain sections of East Africa around Lake Victoria. We conducted a survey in Nigeria, starting from the rain forest in the south and working from village to village up through to the desert in the north. The objective was to find any material that had sweetness greater than sugarcane. There were some surprises, including some of the successes that we have heard about here and others will be heard about in the future. A new discovery was Katemfe. Botanically it is known as Thaumatococcus daniellii. Its sweetener principle was found by Unilever researchers to be two proteins that account for its intense sweetness. Dr. Kare referred to the high intensity sweetness of monellin; the Katemfe protein sweeteners are also very sweet. Monellin as obtained from the serendipity berries is 2,500 times sweeter than sugar, while the Katemfe sweeteners are l,600 times sweeter than sugar. Many outstanding groups and scientists have continued this area of research, including Dr. Beidler at Florida, Dr. Kare at Monell, a group at Dynapol, and Unilever NV in Holland. I am sure many other people are working on these very interesting taste-provoking molecules.
209 We should continue to expand our knowledge of these materials and add other potential sweeteners. We should not restrict our choices of foods; we should expand our horizon and think of new developments. The Queen of England, some one hundred years ago, offered a fantastic amount of money if she could just taste one mangosteen. This fruit has a wonderful flavor and grows in Indonesia. Obviously, at that time, it was not possible for a real one to be sent to her over such a distance; science and technology had not been developed to make it possible at any price. Today we have magnificent science and technological developments. We can do many wonderful things. It would be a shame if we did not take the pathway of searching and finding new sweeteners to make and provide better foods for consumers. DISCUSSION ROBERT HARVEY: I am the ex-president of the Miralin Company, which spent some seven years developing the miracle fruit products. I would like to clarify and expand a bit on some of the things that George Inglett mentioned. We spent seven years and $7 million developing the miracle fruit enterprise, including substantial research on the metabolic safety of the product. We kept the FDA informed over the years as we con- tinued this work. In the fall of l973, we presented a brief to the FDA. Having complied with all of the then-existing FDA regulations, we informed the FDA that this material was about to be market-tested as a GRAS food item. We proceeded with the market-testing, which gave us some unexpected and interesting results. We found that the average consumer had a lower acceptance of the product than we expected. We found that people interested in dieting had about the expected response. There was a tremendous acceptance from diabetics. About 83 percent of the diabetics who tried our product, involving extensive testing in their homes, preferred it to their present diet and wanted to continue. Mothers of juvenile diabetics were particularly pleased with the prod- uct and were interested in continuing the program. This also was true for late-onset diabetics, that is, diabetics who found it very difficult to practice a new diet late in life. They found our alter- native to the other sweeteners a very interesting one, a very pleas- ant one, and one that they wanted to continue We also had done quite a bit of development work on confection products for children. The one thing this Forum seems to agree on is that there are many products containing sugar, particularly the between-meal snacks and the sticky type of sucrose-containing prod- ucts, that contribute to dental caries. We found that we could develop these very confection products by using the miracle fruit
2l0 concentrate as the sweetener. Tests showed that our confection products were preferred over the sugar-sweetened variety. The whole project came to a halt in September of l974, when we received a letter from the FDA, telling us that this material was now regarded as a food additive, and that since there was no legal food additive order outstanding, the product was to be removed from the market immediately. We were to notify FDA within a specified number of days that we had complied with this regulatory letter. This development was quite unexpected. From the contact and feed- back that I had had from the FDA up to this point, this was contrary to the expected response that we had been led to believe would be forthcoming. Mr. Ronk asked for some feedback; I would like to give him some. In the follow-up since the regulatory letter, the FDA has stated that as far as they know, there is no reason to suspect a lack of safety with this product; yet it was essentially banned from the market. We know that several other sweeteners are sold that are known to have adverse effects. We have heard the discussions of saccharin, cyclamate, and even sugar. There is a controversy over the safety of all these sweeteners. Yet the FDA says that a pure fruit concentrate, one of the sweeteners that was on the market and that they had no reason to suspect the safety of, must be banned. The banning of this product caused the failure of the Miralin Company. The product probably will not be made available even though we think that there are some tremendous health benefits to be derived from its use. Frankly, it is my personal opinion that this was an unreason- able position on the part of the FDA and that there could have been some middle ground -- not only could there have been, but I think that there should have been. The Commissioner and other officials of the FDA have stated on many occasions that they use a risk-benefit analysis to arrive at a final decision on products such as ours. Yet the FDA has never asked us for, and I am quite sure it does not have, the information on the benefits that could have been derived from the use of our product. I feel quite confident that they did not, in fact, make a risk-benefit analysis when arriving at their decision. In fact, the basis of their decision is not clear to me at this date. The FDA still has not chosen to tell me the basis for their action, which I find very embarrassing since I am not able to describe to the stockholders who put up the $7 million why the FDA has in fact taken the action that they have taken. I appreciate the opportunity to bring these facts before this body. I also would like to point out some of the things that I think are dangerous in this action by the FDA. I have talked to most of the major representatives of capital souces of money in this country. Even five or six years ago, these sources of money -- which is one of the ways that new products get developed -- were very much afraid of projects that involve or require the approval of the FDA. I be- lieve that actions such as the FDA has taken in our situation and in
2ll other similar ones -- where you just cannot seem to arrive at a reasonable position with the FDA to negotiate or arrange for any kind of a middle-ground position -- will have an adverse effect in the future in terms of being able to raise money from private sources in order to develop products that are to be regulated by the FDA. There is another consequence of this action: as the FDA takes a much more rigid and tougher position, it is raising the development time and costs of these various products. Even those large companies that can afford these long-range and very expensive development pro- grams can undertake them only if the return on investment indicates that it is justified. This means that such programs will be limited, therefore, to a very few specialized products, which eventually could lead to high volume in the marketplace. There are certain sweeteners that would never find a market other than use by diabetics and some other individuals with special dietary problems. Will these special- ized products be developed in the future if l0 or l5 years' devel- opment time and $l0 or $l5 million investment are required before you can realize them? ANITA JOHNSON, Public Citizen Health Research Group, Washington: I find your plea for a special exemption from the Food and Drug Act quite extraordinary. The Food Additives Amendment to that Act says that all new food additives must be proven safe by scientific evidence submitted to the Food and Drug Administration, unless they are gener- ally recognized as safe. Over the last decade the courts have said that the term generally recognized as safe means that there is controlled scientific evidence in the literature equal to the evidence required to be submitted to FDA in a food additive petition. It should come as no surprise to you that the FDA requires you to prove by controlled scientific evi- dence that your product is safe before you market it on any wide- spread basis, and I find it quite extraordinary for you to complain when FDA was just enforcing the law.
FUTURE OPTlONS: Natural and Artificial Sweeteners
