Navigation bar--use text links at bottom of page.

(Looking at the Science on Raw vs. Cooked Foods--continued, Part 1C)

How important is genetic
adaptation regarding cooking?

Assessing meaningfulness of the distinction between natural and non-natural toxins

Ames et al. [1990] has discussed the effects of synthetic versus natural chemicals in the body, and his arguments can be extended to the question of Maillard molecules. It is assumed by proponents of Instinctive Nutrition that, because plants have long been part of human evolutionary history whereas cooking is more recent, the mechanisms that animals have developed to cope with the toxicity of natural chemicals will fail to protect us against Maillard molecules. But this assumption is flawed for several reasons:

General nature of most detoxification mechanisms blurs some of the distinction between hazards of natural/non-natural toxins

None of this is to say that adaptation is not important (it is), but that one should not be unduly afraid of the fact that possibly thousands of unknown different chemical variants are produced by cooking, because there are also thousands of toxins in natural foods already. Indeed, while humans could not possibly have developed distinct detoxification mechanism for each one of these molecules, there may be no particular need to.

Comparing degree of toxicity between natural/non-natural chemicals is not necessarily easily predictable or black-and-white. Recall that the body's detoxification mechanisms even for the natural toxins it has encountered during its evolutionary history are mostly of a more general nature, and may therefore be reasonably expected apply to a wide range of "artificial" chemicals that might be produced by cooking as well. And we know that some molecules humans haven't previously encountered can be safe, even safer than toxicants occurring in nature. Indeed, artificial preservatives are allowed only after testing on laboratory animals (while much has been said against preservatives, at least no one dies or becomes violently sick from their ingestion); malathion, the main synthetic organophosphate pesticide residue in our diet, is not a carcinogen in rats or mice. On the other hand, intoxications by cyanogenic glycosides (present in apricot kernels and cassava tubers) have been reported, and natural carcinogens are quite widespread (see below).

(Note: It's not being claimed that malathion, mentioned just above, doesn't have other toxic effects; but as stated, it has not been found to be a carcinogen in rats or mice. And it should be mentioned as well that Ames' research referred to above is sometimes misused by the chemical industry to gloss over very legitimate safety concerns. However, the lesson from Ames' research is that some natural toxins are as toxic as some synthetic ones; and that, as well as some artificial chemicals, some "natural" pesticides which are used in organic agriculture should be tested and possibly prohibited if judged too dangerous.)

Plant toxins the result of evolutionary "arms race" to protect against overexploitation by animals. One must be realistic that the very reason the body has evolved detoxification and excretion methods in the first place is because it is quite normal and necessary for it to be able to handle a given load of toxicants within some reasonable, if modest, range anyway. Nutrition is a cost/benefit transaction where nutrients are gained at the cost of dealing with accompanying toxins and other unusable material. All plants have evolved toxin defenses as part of their survival strategy to discourage overpredation by animals. (This is common knowledge among evolutionary biologists but generally unrecognized, or at least unappreciated, in dietary circles.)

Some toxins, of course, may be less prevalent in certain plant parts than in others (in fruits, for example, which are a "trojan horse strategy" by plants to encourage certain animal species to distribute the seeds, via their feces, by eating the fruits). Nonetheless toxins are present at some level in fruits. (Also, very few mammals can exist primarily on fruits; and in any event, fruits are not continuously available in the wild, necessitating a broader range of feeding habits for mammals who do eat fruit as a component of their diets.) Thus, toxins in one form or another are an unavoidable "price" of the search for food.

Even in "wild nature," nutritional cost/benefit trade-offs drive opportunistic human decisions about eating raw vs. cooked. As we will see with the hunter-gatherer tribes covered in Part 3, it may not always be the case that the net available usable nutrients from a food (once toxins and antinutrients are neutralized) is better in the raw plants that are available in a given environment than in their cooked form. Here, we see that rawist ideas about a toxin-free natural diet are exercises in idealism rather than reality, especially when we see some raw-fooders becoming emaciated from missing out on valuable nutrients in their quest to avoid all toxins.

Antinutritional properties of
Maillard reaction products

First, it is important to assess which antinutritional properties have the most deleterious consequences on, say, the average American or Westerner. (Definition: An "antinutrient" is a substance that is not necessarily toxic per se, but is detrimental because it either inhibits digestion of certain nutrients or else binds with them during digestion to prevent uptake.) It appears, from Pao and Mickle [1981], that the "problem" nutrients, i.e., the nutrients that at least 30% of the population are deficient in (<70% of the RDA), are: vitamin B-6, magnesium (Mg), calcium (Ca), iron (Fe), vitamin A, and vitamin C. Adults usually have an adequate intake of proteins, but protein quality is more important for infants and hospitalized patients.

Lysine availability

The loss of lysine via Maillard reactions generally occurs in storage. The main effect of Maillard reaction products is that they reduce availability of some amino acids, primarily lysine.

In summary, the conclusion of the above is that, except possibly for infants, the reduction of lysine availability is not an important health hazard for humans. We will also see in Part 2 of our article that cooking may sometimes improve the nutritional value of a food for other reasons, so the net benefit may or may not be positive, depending on the food used and the cooking method. Certainly, all arguments speak against cooked milk, but as some foods may benefit from cooking, generalizing from experiments on milk to cooking in general is of doubtful value.

Effects of Maillard products on vitamins

Maillard molecules also have antinutritional properties with regard to vitamins.

When spray-dried whole milk powder was stored for 9 weeks at 60°C (140°F) and underwent early Maillard reactions, there was a moderate loss (less than 20%) of some vitamins, especially B-1 and B-6. At 70°C (158°F), however, there was a much more rapid and dramatic destruction of the vitamins B-1, B-6, B-12, and pantothenic acid (over 50% loss in 3 weeks) [Ford et al. 1983].

Again, these conditions of storage were extreme, and milk is particularly sensitive; thus, parallel to the situation above with lysine, it is likely one might expect considerably less loss under more commonly encountered conditions. Also, as we will see in a later section of this paper, the total amount of vitamins lost in foods after conventional cooking procedures (i.e., from all causes, including heat destruction) are relatively modest. Thus, the remaining portion of the loss that may be due to Maillard molecules produced during normal cooking must also be relatively small. Unfortunately, however, at the time this paper was written we were not able to locate more relevant data to assess the exact effect. (The role of heat in the destruction heat-labile vitamins will be covered in Part 2.)

Effects of Maillard products on minerals

Mineral metabolism is complex because of certain factors that can affect absorption, retention, and excretion, and which don't necessarily come into play with respect to other nutrients such as most vitamins. For example, where absorption is concerned, chelation or other binding effects (such as occurs with phytic acid) will impact absorbability of divalent minerals. Where retention is at issue, a calcium ion is not necessarily stored in bone, for instance, because the acid/alkaline balance can affect excretion, or loss of calcium from bone. A few studies have been made to determine whether Maillard molecules induce mineral depletion.

Effect on zinc possibly mildly problematic. Citing two studies, Johnson [1991] notes that Maillard products in solutions for parenteral nutrition--that is, when fed intravenously--increased urinary losses of minerals: specifically iron, zinc, and copper. Johnson [1991] discusses relevant studies she has conducted, e.g., feeding human volunteers a diet high in Maillard products (via baked goods, which included browned grain cereals in some studies) and other studies on rats. She also discusses the results of other, related research studies. The results of these studies are mixed; there is some evidence of zinc-binding in substrates of browned albumin due to the presence of Maillard reaction products, but human feeding studies don't necessarily show lower zinc absorption as a result of diets that include "browned" foods (i.e., that contain Maillard reaction products).

Johnson [1991] also reviews the results of published studies on feeding rats mixtures that include Maillard reaction products. The work of O'Brien et al. [1989] (as cited in Johnson [1991]) of feeding rats mixtures of glucose, monosodium glutamate, and MRPs found increases in urinary losses of several minerals: calcium, magnesium, zinc, and copper. The effect on zinc was highest among the minerals mentioned.

Furniss et al. [1989] studied the effect of feeding rats diets that included varying types of Maillard reaction products, and found that feeding casein-glucose MRP mixtures caused a 6-fold increase in the urinary excretion of zinc; casein-lactose MRP mixtures a 2-fold increase. They noted that urine is a minor excretion route for zinc (when compared to fecal excretion) and that Maillard products appear to have no influence on overall zinc retention or on zinc status in rats. However, Lykken et al. [1986] found that zinc absorption in humans was 47% from corn grits and 37% from corn flakes.

Problems with zinc perhaps more of a concern in relation to other factors (phytates in grains). From the above results, keeping in mind once again--as with the data on lysine and vitamins--that the storage conditions were extreme and that the Maillard reaction occurs at a much faster rate in casein-lactose mixtures than in foods other than milk, it seems that the only mineral deficiency likely to be caused by cooking would be a (mild) zinc deficiency. Let's add, as a side note, that zinc deficiencies are not uncommon in people eating raw, especially those who are vegetarians. The reason is that zinc is more available in animal sources than in vegetarian ones, where it is often bound to phytic acid (as in grains). There is evidence [Donovan et al. 1995] that vegetarians often have a lower zinc status than omnivores. (That the same is true with raw vegetarians is anecdotal evidence, since virtually no studies are available on the latter.) For more on the question of zinc in omnivorous compared to vegetarian diets, see Key Nutrients vis-a-vis Omnivorous Adaptation and Vegetarianism: Zinc (go about three-fourths the way down the page) elsewhere on this website.


(Do Maillard Products Carry Potential Risks for Mutagenicity and Carcinogenicity?)

Return to beginning of article



GO TO PART 1 - Is Cooked Food "Toxic"?

GO TO PART 2 - Does Cooked Food Contain Less Nutrition?

GO TO PART 3 - Discussion: 100% Raw vs. Predominantly Raw

Back to Research-Based Appraisals of Alternative Diet Lore

   Beyond Veg home   |   Feedback   |   Links