Cooking food is the major process adopted to produce flavor chemicals, which give foods a characteristic desirable roasted or toasted flavor. More than 1,000 low molecular weight compounds are known to be produced once a food is heated, and the number of known compounds is increasing steadily. The amounts of these chemicals formed are generally in very trace levels (around the level of µg/kg of food or parts per billion, ppb), but for most of these compounds, detailed information about possible toxicity at this level is not yet established.
One of these compounds is 4-Methylimidazole (4-MEI), which forms from a reaction between amino acids or proteins and sugars, carbohydrates, or lipids when exposed to heat. This past March, an independent study commissioned by the consumer interest group Center for Science in the Public Interest (CSPI) found that four of the best-selling sodas (Coca-Cola, Diet Coke, Pepsi and Diet Pepsi), which...
But I do not believe 4-MEI is harmful at the levels found in these products. Stating that these sodas contain potentially unsafe levels of the chemical is misleading—it is present only trace levels. Furthermore, the supposed cancer link stems solely from a handful of animal experiments; there is currently no evidence of a link between 4-MEI and disease in humans.
I recently conducted a study that demonstrated that adding sulfite, which is one of the most commonly used caramel ingredients, reduces 4-MEI formation. But when we try to reduce 4-MEI, we may have to remove a tremendous number of other desirable chemicals formed in foods by heat treatment. Furthermore, it is important to keep in mind that heat treatment of foods does more than give the food palatable flavors, colors, and textures; heat treatment also kills harmful microorganisms, and it should be stressed that ingesting such microorganisms may cause more toxic effect to humans than ingesting trace levels of 4-MEI.
And this concern over the presence of 4-MEI in popular soda brands is just the latest in the periodic excitement that is stirred up over chemicals that are formed in foods when they undergo processing and cooking. Other potentially carcinogenic chemicals that have raised concerns in the past include polycyclic aromatic hydrocarbons, N-nitrosamines, aromatic amines, and acrylamide. Whenever one of these carcinogens is reported in cooked food, many people from various specialties, including food sciences, regulatory agencies, and medicine, not to mention consumers, react by giving the new chemical concentrated attention and intense discussions about how to reduce it. However, in each case, after some time has passed, attention fades away, in part because there is no practical way to remove these chemicals from cooked or processed foods and also because there is no evidence that the trace levels consumed present any health risk to humans.
It should also be noted that all substances are poisons; only the dose differentiates the poison from a remedy. This concept was developed by Paracelsus, a medical doctor in Europe in the 16th century, and it is useful in evaluating risk in our daily lives by considering examples of well-known substances with low and high toxicity consumed today. For example, water might be considered one of the least toxic substances that we commonly encounter. However, there are many reports of water toxicity in the scientific literature. Too much water can, indeed, be fatal. For example, in 2002 a student was undergoing a fraternity initiation ordeal, in which he was required to drink up to 5 gallons in a short period of time. Consumption of this amount of water resulted in the dilution of the electrolytes in his blood to the point that normal neurological function was lost, and the young man sadly died.
While this example is extreme, the same basic logic must be applied to the toxins that are released in foods during the cooking process. In other words, we must consider not just the chemicals we are ingesting, but how much of those chemicals we consume in our diet. Even well known carcinogens, such as arsenic or nicotine, are only harmful if exposure is high enough. In the case of 4-MEI, the amount present in foods is extremely low and it is hard to predict any kind of adverse health effect at these levels. People have been consuming numerous heat-induced chemicals, including 4-MEI, since the discovery of fire some half a million years ago, and there has been no clear evidence of any specific health problems to date. Thus, while it is extremely important to continue ongoing research to determine the toxic level of 4-MEI, I personally choose to enjoy flavorful, cooked foods, without fear for my health.
Takayuki Shibamoto is a distinguished professor in the Department of Environmental Toxicology at the University of California, Davis, and an associate editor of Journal of Agricultural and Food Chemistry.
Correction (July 16, 2012): This story has been updated from its previous version to correctly state that parts per billion (ppb) is equivalent to micrograms/kilogram, not milligrams/kilogram. The Scientist regrets the error.