There were indications as early as 1908 that animal proteins were not always perfectly tolerated by herbivores (Ignatowski, 1908), and by the early 1960's it was obvious that feeding proteins to mammals was not as simple as it seemed. For example, rabbits fed a high fat diet based on casein and beef fat develop arterial disease (Jones et al., 1963), but simple replacement of the casein by soya bean meal prevents the deposition of fat in the arteries (Howard et al., 1965). While some of the apparent protective effect of the soya bean meal could be traced to the essential fatty acids present in the lecithin fraction, this in itself could not account for the entire effect, and the findings, which could readily be reproduced, left researchers at that time believing in "unknown protective factor X", which we now believe are the soy isoflavones. Later studies have confirmed that isolated soy proteins ("soy protein isolates") which are now in common industrial use behave in much the same way (Kritchevsky et al., 1983). In fact, most animal proteins produce some elevation of blood cholesterol in herbivores, while plant proteins uniformly reduce the blood cholesterol levels in such animals (Van der Meer and Beynen, 1987). It would be interesting to see if the effects in carnivores were the reverse of this!

In humans, diets in which some of the animal protein has been replaced by soy protein have substantial serum cholesterol lowering effects in hypercholesterolemic subjects (Widhalm, 1986; Potter et al., 1993; Bakhit et al., 1994), with major decreases in low density lipoproteins (LDL), but studies in normal healthy subjects have generally given ambiguous results. Epidemiological studies have also failed to demonstrate any effect of dietary protein on coronary heart disease.

While there are clear species-related distinctions between animal and plant proteins with regard to effects on lipid metabolism and vascular disease, effects on blood pressure are more difficult to categorize. It has been shown that very high protein diets limit the development of severe hypertension and reduce the incidence of stroke in various strains of spontaneously hypertensive rats (Lovenberg and Yamori, 1984), while low-protein diets have opposite effects (Wexler, 1983a). In this system, fish and milk proteins were somewhat protective; the hypertension that resulted when protein intake dropped to 10% of energy was less severe if the protein came from milk or fish (Wexler, 1983b), and both casein and whey were particularly protective (Ikeda et al., 1987).

Effects on blood pressure in humans are somewhat more conclusive. Epidemiological studies show a beneficial effect of high protein intakes on blood pressure (Pellum and Medeiros, 1983), while clinically, subjects placed on lacto-ovovegetarian diets (milk, egg and plant proteins) showed reductions in both systolic and diastolic blood pressure in relation to omnivorous control subjects.

Relationships between dietary protein and cancer, osteoporosis and kidney disease have also been investigated in clinical or epidemiological studies. Sporadic associations have been found between animal protein intake (mainly meat) and large bowel cancer, breast cancer and pancreatic cancer, but most studies have failed to demonstrate such associations, so the overall results are inconclusive. It has also been shown that bone mineral mass is lower in omnivorous women than in lacto-ovovegetarian women (Marsh et al., 1980), though the exact reason for this remains obscure, but there is no epidemiological or clinical data to support a hypothesis that high protein intakes can lead to kidney disease.

From the foregoing, it can be concluded that the type or types of protein consumed may have some positive or negative health aspects that are probably unrelated to the classical nutritional role in the diet. Plant protein appears to be much better for herbivorous animals than does animal protein, but in humans, which are basically herbivores, the epidemiological and clinical evidence suggests that combinations of plant proteins with milk proteins, optionally with additional egg protein, confer the maximum benefits. Just about all the evidence implies that genuine meat protein may not be such a good thing, and this is analogous to what we already know about fat; the hidden fat of red meat is probably the worst thing you are ever going to eat!

This does not imply that every meal must consist of plant protein with or without milk protein (optionally with eggs), since it is the overall content of the food intake over a daily or longer period that really counts. What it does imply is that, if you are a lover of (red) meat, there are a few things you should be doing! The first is obviously to cut back, and replace the protein that you are missing with plant and milk proteins. The second is to adjust your fat intake to compensate for the hidden fat in the red meat (more omega-3 essential fatty acids in particular).

There has to be a reason why plant proteins, with or without milk proteins, are beneficial, and some industrial groups have invoked the presence of mystical trace substances to account for the benefits. While small amounts of "phytochemicals" such as isoflavones may certainly exert some effect, research indicates that the amino acid profile is probably the most important factor governing the "benefits" of the protein in a particular species. However, the specific amino acids, or combinations of amino acids, responsible for the observed effects have not been identified.

In practical terms, formulators of food products can base their protein composition on the essential amino acid score (AAS), corrected for digestibility, using the essential amino acid requirement pattern for humans 2 - 5 years of age (FAO/WHO, 1990). This pattern was chosen because it is the most demanding pattern of any human age group other than infants, therefore if a protein or protein blend scores 1.0 (or more) against this pattern, it is of high nutritional value:


Methionine + cystine
Phenylalanine + tyrosine

To score a protein, the content (in mg) of each of the above amino acids present in 1 gram of the protein (which may be a mixture of different proteins) is determined, and divided by the appropriate value from the reference pattern. This gives the uncorrected AAS, which is then multiplied by the digestibility factor (this ranges from 0.4 to almost 1 for most proteins). The lowest value is the official digestibility-corrected AAS (PDCAAS), but all the values are of interest from a nutritional point of view. A comparison of the values obtained by this method with those obtained by Protein Efficiency Ratio determinations in rats is illuminating:


Soy protein isolate


Whey protein

Milk protein

Beef protein






> 1.00

> 1.00


> 1.00


Thus for rats, whey and beef are definitely superior to soy, but for humans, soy is better than whey and beef. However, since all these proteins have different amino acid profiles, they can be blended or mixed in the daily protein intake to achieve higher biological values or specific effects. Thus it is quite possible to blend soy protein with milk proteins in order to obtain an amino acid profile that increases the biological value or matches more closely the profile that may be needed to achieve certain benefits. Food technologists may also want to blend proteins to achieve better taste, or better physical properties.

Knowledge of health benefits associated with certain amino acid profiles and the essential amino acid requirements of humans allows nutritionists and food technologists to work together and create the best of both worlds; products that not only taste good, but are also good for you!

- back to top -