Paul-Andre Finot

Historical Perspective of the Maillard Reaction in Food Science

Abstract: Maillards paper of 1912 describing the reaction between amino acids and sugars is both innovative and visionary. It provides original and still‐valuable data on the chemistry of a new reaction and foresees its involvement in many scientific and biological domains, even in human pathology. This paper was ignored by the scientific community until 1941. In 1948 the Maillard reaction was definitely recognized as being responsible for the browning and loss of nutritive value of heated milk powders. There was then a continuous increase in papers on the chemistry of this complex reaction to identify its various pathways: in food science, to evaluate the influence of reaction parameters (pH, T°, time, sugar reactivity, concentration of the reagents, water activity, glass transition temperature) on the evolution of the reaction and on changes in food quality; in nutrition, to quantify the loss of bioavailability of essential amino acids; on the metabolism of the reaction products and on the physiological effects of the ingested Maillard reaction products. The significant scientific advances and the key persons and pioneers who contributed much to the understanding of the Maillard reaction are presented. The food industry is directly concerned with the occurrence of this reaction in processed foods and contributed significantly by its own research to understanding the phenomena and to optimizing the processes and conditions of food preparation in order to preserve the nutritional, safety, and organoleptic qualities of foods.

Nutritional Consequences of the Reactions Between Proteins and Oxidized Polyphenolic Acids

The chemical and enzymatic browning reactions of plant polyphenols and their effects on amino acids and proteins are reviewed. A model system of casein and oxidizing caffeic acid has been studied in more detail. The effects of pH, time, caffeic acid level and the presence or not of tyrosinase on the decrease of FDNB-reactive lysine are described. The chemical loss of lysine, methionine and tryptophan and the change in the bioavailability of these amino acids to rats has been evaluated in two systems: pH 7.0 with tyrosinase and pH 10.0 without tyrosinase. At pH 10.0, reactive lysine was more reduced. At pH 7.0 plus tyrosinase methionine was more extensively oxidized to its sulphoxide. Tryptophan was not chemically reduced under either condition. At pH 10.0 there was a decrease in the protein digestibility which was responsible for a corresponding reduction in tryptophan availability and partly responsible for lower methionine availability. Metabolic transit of casein labelled with tritiated lysine treated under the same conditions indicated that the lower lysine availability in rats was due to a lower digestibility of the lysine-caffeoquinone complexes.

Stearidonic acid, an inhibitor of the 5-lipoxygenase pathway. A comparison with timnodonic and dihomogammalinolenic acid

Leukotrienes have been shown to play an important role as mediators in various disease processes, including asthma and inflammation; thus, their synthesis is tightly regulated. The major precursor of leukotrienes is arachidonic acid (20∶4n−6). Fatty acids which are structurally similar to 20∶4n−6, such as eicosatrienoic acid (20∶3n−6; dihomogammalinolenic acid) and eicosapentaenoic acid (20∶5n−3; timnodonic acid) have been found to inhibit leukotriene biosynthesis. Because of the structural similarity of octadecatetraenoic acid (18∶4n−3; stearidonic acid) with 20∶4n−6, the present study was undertaken to determine whether stearidonic acid also exerts an inhibitory effect on the 5-lipoxygenase pathway. Human leukocytes were incubated with 18∶4n−3 (20 μM or 10 μM), 20∶5n−3 (20 μM) or 20∶3n−6 (20 μM) and subsequently stimulated with 1 μM ionophore A23187 and 20∶4n−6 (20 μM or 10 μM). The 5-lipoxygenase products were then measured by high-performance liquid chromatography. Leukotriene synthesis was reduced by 50% with 20 μM 18∶4n−3 and by 35% with 10 μM 18∶4n−3. Formation of 5S,12S-di-hydroxy-eicosatetraenoic acid and of 5-hydroxy-eicosatetraenoic acid was decreased by 25% with 20 μM 18∶4n−3 and by 3% with 10 μM 18∶4n−3. The inhibition observed with 20 μM 18∶4n−3 appeared to be of the same order as that observed with 20 μM 20∶5n−3; the inhibition observed with 18∶4n−3 was shown to be dosedependent. The inhibition produced by 20 μM 20∶3n−6 was greater than that observed with either 20 μM 18∶4n−3 or with 20 μM 20∶5n−3. The results suggest that stearidonic acid, which is found, for example, in blackcurrant seed oil (which also contains the 20∶3n−6 precursor), may play a role in suppressing inflammation.

Availability of the true Schiff's bases of lysine. Chemical evaluation of the Schiff's base between lysine and lactose in milk.

During the heat-treatment of milk, the Maillard reaction which takes place between the epsilon-amino group of lysine and lactose leads to the formation of two well-defined chemical types : the Schiff’s base in equilibrium with its aldosylamine form and the deoxyketose (Amadori product). Rat growth assays showed that the synthetic e-N-deoxyketosyl-L-lysine was not utilized as a source of lysine and that the true Schiff’s bases resulting from the reaction of lysine with aromatic aldehydes were 100% utilized indicating that the Schiff’s base ⇄ aldosylamine are also 100% utilized.

Black Currant Seed Oil Feeding and Fatty Acids in Liver Lipid Classes of Guinea Pigs

Guinea pigs were fed one of three diets containing 10% black currant seed oil (a source of gamma-linolenic (18∶3 n−6) and stearidonic (18∶4 n−3) acids), walnut oil or lard for 40 days. The fatty acid composition of liver triglycerides, free fatty acids, cholesteryl esters, phosphatidylinositol, phosphatidylserine, cardiolipin, phosphatidylcholine and phosphatidylethanolamine were determined.Dietary n−3 fatty acids found esterified in liver lipids had been desaturated and elongated to longer chain analogues, notably docosapentaenoic acid (22∶5 n−3) and docosahexaenoic acid (22∶6 n−3). When the diet contained low amounts of n−6 fatty acids, proportionately more of the n−3 fatty acids were transformed. Significantly more eicosapentaenoic acid (EPA) (20∶5 n−3) was incorporated into triglycerides, cholesteryl esters, phosphatidylcholine and phosphatidylethanolamine of the black currant seed oil group compared with the walnut oil group.Feeding black currant seed oil resulted in significant increases of dihomogamma-linolenic acid (20∶3 n−6) in all liver lipid classes examined, whereas the levels of arachidonic acid (20∶4 n−6) remained relatively stable. The ratio dihomo-gamma-linolenic acid/arachidonic acid was significantly (2.5-fold in PI to 17-fold in cholesteryl esters) higher in all lipid classes from the black currant seed oil fed group.

Effects of diet-induced hyperthreoninemia. I). Amino acid levels in the central nervous system and peripheral tissues.

Rats were fed four levels of threonine (Thr, 0.4, 0.6, 0.8, and 5.8 g/100 g diet). After two weeks, Thr, serine (Ser), and glycine (Gly) levels were measured in plasma, liver, muscle, and central nervous system. The diet containing 5.8 g/100 g of Thr elevated Thr and Gly concentrations in plasma and nervous tissue in comparison with a standard diet. In muscle and liver, Thr concentrations were also raised but Gly levels did not change. The hepatic Thr dehydratase activity was enhanced. Diets containing moderate Thr quantities (0.6 and 0.8 g/100 g) induced slight elevations of Thr levels in all tissues. Gly concentrations were not modified. The activity of hepatic Thr dehydratase was diminished. Our results show that a high dietary content of Thr (15 times the normal levels) elevates Gly levels in various tissues, including the brain. On the contrary, diets containing 2 to 4 times the normal levels of Thr induce a weak hyperthreoninemia insufficient to modify brain Gly.

Food Processing and Storage as a Determinant of Protein and Amino Acid Availability

Protein is perhaps the most reactive of the major food components. During food processing, the essential amino acids, lysine, tryptophan, methionine and cyst(e)ine, may react with other food components causing a loss in amino acid bioavailability and sometimes a reduction in the digestibility of the whole protein molecule. This review first discusses some recent developments concerning protein-polyphenol reactions, racemization and lysinoalanine formation, and then describes the reactions of proteins and reducing sugars (the Maillard reaction) in greater detail. We report on the chemistry of the Maillard reaction, the nutritional and physiological properties of the newly-formed products and their metabolic transit in the rat. In practice, the Maillard reaction is by far the most important reaction of food proteins. It is especially important in milk products since these are the only naturally-occurring protein foods with a high content of reducing sugar. Lysine is the most sensitive amino acid to damage during processing and storage and its losses may be of nutritional significance to certain population groups, such as babies, who are often dependent on a single manufactured product as their sole source of nourishment.

Metabolic Transit of Lysinoalanine (LAL) Bound to Protein and of Free Radioactive |14C|-Lysinoalanine

Three alkali-treated proteins (lactalbumin, fish protein isolate and soya isolate) containing respectively 1.79, 0.38 and 0.12 g of lysinoalanine (LAL)/16 g N, were submitted to an “in vitro” enzymatic hydrolysis (pepsin then pancreatin); the higher the level of LAL present in the proteins the smaller was the proportion of LAL liberated in the dialysable fractions of the enzymatic hydrolysates. These same alkali-treated proteins were also given to rats in feeding studies. The faecal LAL varied between 30 and 50% of the ingested quantity, and the urinary LAL between 10 and 25%. The total recovery was always inferior to 100% showing that a certain proportion of LAL was modified in the organism. In the urine, LAL was partially present as free LAL and also as combined LAL, its recovery being higher after acid hydrolysis. The experiments performed with radioactive LAL confirmed the above-mentioned results and gave complementary data: LAL was partially transformed into 14CO2; the urine contained free LAL and several metabolites, some combined LAL, probably acetyl derivatives, and some products of catabolism. The proportions of the urinary products varied widely from one animal to another. The kidneys play an important role in the chemical modification (acetylation) of LAL and in the filtration of the excretion products which is most efficient for the most acetylated catabolites of LAL. Although there is practically no difference in the pattern of the urinary catabolites in the rodents (Sprague-Dawley rat, Swiss mouse, Syrian hamster), the rat kidney retains LAL and its catabolites at levels much higher than the kidneys of other species; this retention is localized in the inner part of the cortex.

Effects of diet-induced hyperthreoninemia. II) Tissue and extracellular amino acid levels in the brain

Growing rats were fed graded levels of threonine (Thr, 0.4, 0.8, and 3.3 g/100 g diet). Free amino acid content was measured in plasma and brain. Extracellular amino acid levels were measured by microdialysis in brain slices. Large quantities of dietary Thr (3.3 g/100 g) raised plasma and brain Thr and glycine (Gly) levels. Brain and spinal cord extracellular levels of Thr were also raised, whereas the other amino acid levels remained unchanged. A moderate level of dietary Thr (0.8 g/100 g) raised plasma Thr and Gly levels and brain Thr but not Gly level. The diet raised cortical Thr extracellular levels but did not modify the levels of the other amino acids, including glutamate (Glu) and aspartate (Asp). These data suggest that brain neurochemical processes involving Gly, Glu, and Asp are safeguarded in rats fed high Thr levels.

Improvement in the Nutritional Quality of Bread

To assess whether the dipeptide N-epsilon-(gamma-L-glutamyl)-L-lysine (glutamyl-lysine) can serve as a nutritional source of lysine, we compared the growth of mice fed (a) an amino acid diet in which lysine was replaced by six dietary levels of glutamyl-lysine; (b) wheat gluten diets fortified with lysine; (c) a wheat bread-based diet (10% protein) supplemented before feeding with lysine or glutamyl-lysine (0, 0.75, 1.50, 2.25, and 3% lysine HCl-equivalent in the final diet), not co-baked and (d) bread diets co-baked with these levels of lysine or glutamyl-lysine. With the amino acid diet, the relative growth response to glutamyl-lysine was about half that of lysine. The effect of added lysine on the nutritional improvement of wheat gluten depended on both lysine and gluten concentrations in the diet. With 10 and 15% gluten, 0.37% lysine HCl produced a marked increase in weight gain. Further increase in lysine HCl to 0.75% proved detrimental to weight gain. Lysine HCl addition improved growth at 20 and 25% gluten in the diet and did not prove detrimental at 0.75%. For whole bread, glutamyl-lysine served nearly as well as lysine to improve weight gain. The nutritive value of bread crust fortified or not was markedly less than that of crumb or whole bread. Other data showed that lysine or glutamyl-lysine at the highest level of fortification, 0.3%, improved the protein quality (PER) of crumb over that of either crust or whole bread, indicating a possible greater availability of the second-limiting amino acid, threonine, in crumb. These data and additional metabolic studies with U-14-C glutamyl-lysine suggest that glutamyl-lysine, co-baked or not, is digested in the kidneys and utilized in vivo as a source of lysine; it and related peptides merit further study as a sources of lysine in low-lysine foods.