THESlB LIBRARY Michigan State University > This is to certify that the thesis entitled A Study of the Sugar-Amine Reaction Through Mode] Systems presented by Karim Nafisi-Movaghar has been accepted towards fulfillment of the requirements for Ph. D. degree in Food Science '7) - [4'71 , ' ' {aux/62.4 film/tuba Major professor Date 4/2924»; /950 0-7 639 OVERDUE FINES: 25¢ per day per item RETURNING LIBRARY MATERIALS: _____..____________________. Place in book return to remove charge from circulation records m M - w L‘i; @- éfiiflffln wi- ,' 1‘ ~11 u '1 , \\ 1 r I ”WR3RV7$§9“ ., A STUDY OF THE SUGAR-AMINE REACTION THROUGH MODEL SYSTEMS BY Karim Nafisi-Movaghar A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Food Science and Human Nutrition 1980 ABSTRACT A STUDY OF THE SUGAR-AMINE REACTION THROUGH MODEL SYSTEMS BY Karim Nafisi-Movaghar The extent of the browning resulting from the inter- action of ten amino acids (alanine, arginine, aspartic acid, glutamic acid, glycine, lysine, methionine, phenylalanine, serine and o-amino-n-butyric acid) with two sugars (glucose and fructose) was measured as a function of pH, temperature, duration of the reaction, and concentration of reactants. The main findings are summarized as follows: Increasing the pH of the system from 6 to 8 increased the rate of browning (absorbance at 540 nm), though not linearly; The extent of browning was less for fructose than glucose while reacting with amino acids, except in the model systems containing amounts of glycine less than 0.031 M, at pH 8.0; Fructose was found to caramelize more readily than glucose; Among the amino acids tested, lysine and arginine were the most active in the production of brown color when Karim Nafisi-Movaghar reacting with glucose and fructose. However the activity of arginine at pH 6 with fructose was nearly nil; At pH 7 the reactivity of arginine with fructose was greater than that of lysine with fructose; Glutamic acid and aspartic acid appeared to sup- press the caramelization of glucose and fructose; The addition of glutamic acid or aspartic acid to model systems containing lysine and glucose or lysine and fructose diminishes the color develOped by these systems. At a 1:1 molar ratio of lysine to glutamic acid, the inhibi- tion of browning is greater when the total concentration of lysine and glutamic acid is higher. Impregnating freeze- dried potato slices with aspartic or glutamic acid solutions resulted in less brown discoloration upon frying the slices; The activation energy of the glucose-lysine reaction was 13,331 cal/mole, and for the fructose-lysine reaction 10,210 cal/mole; The 010 for the same reactions were 1.87 for glucose- lysine and 1.61 for fructose-lysine. To Terry for all her patience ii ACKNOWLEDGMENTS The author would like to express his very sincere thanks to Dr. Pericles Markakis, Professor, Department of Food Science and Human Nutrition, for his expert and imagi- native constructive criticism, professional guidance and encouragement, his understanding and sincere friendship throughout the course of this study; Dr. G. S. Karabatsos, Chairman and Professor of the Department of Chemistry, for spending many of his restricted hours on this study; Dr. Richard C. Nicholas, Professor of Food Science and Human Nutrition, for acting both as my guidance com- mittee member and my supervisor, and for all his friendly discussions; Dr. Charles M. Stine, Professor of Food Science and Human Nutrition for his valuable advice on many subjects; Dr. Andrew Timnick, Professor of Chemistry, for accepting the agony of being on my committee and for his valuable encouragement. The author wishes to express his gratitude to the Department of Food Science and Human Nutrition, Michigan iii State University, for the financial aid paid to me as a graduate research and teaching assistant. Thanks also goes to all of the author's fellow graduate students, particularly D. Fardiaz, for their use- ful discussions. iv TABLE OF CONTENTS LIST OF TABLES C O O O O O O O C O O O O O O O O 0 LIST OF FIGURES . . . . . . . . . . . . . . . . . INTRODUCTION LITERATURE REVIEW . . . . . . . . . . . . . . . . Non-Enzymatic Browning . . . . . . . . . . . . A. Production of N-Substituted Glycosylamine . . . . . . . . . . B. Production of Amadori Compound . . . . C. Formation of Pigments . . . . . . . . 1. From the Keto Form . . . . . . . . 2. From the Enol Forms . . . . . . . Strecker Degradation . . . . . . . . . . . . . Caramelization . . . . . . . . . . . . . . Factors Affecting the Maillard Reaction . . . Inhibition of the Browning Reactions . . . . . Chemical Methods of Inhibition . . . . . . Nutritional Aspects of the Maillard Reaction . METHODS A. 11100!!!» '11 Amino Acids . . . . . . . . . . . . . Proteins . . . . . . . . . . . . . . . Sugars . . . . . . . . . . . . . . . . Toxicity . . . . . . . . . . . . . . Flavor Produced by the Amino-Carbonyl Reaction . . . . . . . . . . . . . . Other Aspects of the Maillard Reaction (Antioxidant Property) . . . . . . . . AND MATERIALS . . . . . . . . . . . . . . Materials 0 O O O O O O O O O O O O O O O l. 2. Amino Acids and Amines . . . . . . . . Ampoule-holder . . . . . . . . . . . . V Page viii U1-bw (DC!) 12 17 18 19 24 24 25 26 28 29 32 35 35 35 36 E. F. RESULTS I. II. III. IV. V. 3. Ampoules . . . . . . . . . . . . . . . . 4. Phosphate Salts . . . . . . . . . . . . 5. Reducing Sugars . . . . . . . . . . . . General Procedure for Obtaining Data for Reaction Rates . . . . . . . . . . . . . . . Phosphate Buffers . . . . . . . . . . . Amino Acid Solutions . . . . . . . . . . Reducing Sugars . . . . . . . . . . . . . Final Sample Preparations . . . . . . . QWNH o 0 Preparation of Samples Containing Glutamic and Aspartic Acids . . . . . . . . . . . . . Procedure for Determining the Time Dependence of the Browning Reactions . . . . l. Phosphate Buffer Solutions . . . . . . . 2. Mixture of Glucose and/or Fructose with Lysine Solution . . . . . . . . . . 3. Reducing Sugar Solution . . . . . . . Test for the Retardation of Browning with Aspartic Acid and Glutamic Acid . . . . . . Physical Measurements . . . . . . . . . . . AND DISCUSSION . . . . . . . . . . . . . . . Amino Acids . . . . . . . . . . . . . . . . A. Amino Acids-Glucose Systems at pH 8 . . . Glycine—Glucose . . . . . . . . . . . Alanine-Glucose . . . . . . . . . . . Phenylalanine-Glucose . . . . . . . . Serine-Glucose . . . . . . . . . . . . Methionine-Glucose . . . . . . . . . . Lysine and Arginine-Glucose . . . . mm-waH B. Amino Acids-Glucose Systems at pH 7 . . C. Amino Acids-Glucose Systems at pH 6 . . Sugars . . . . . . . . . . . . . . . . . . . Lysine and Arginine Interaction with Reducing Sugars . . . . . . . . . . . . Temperature Dependence of the Reaction Rate Inhibition of Caramelization . . . . . . . vi Page 36 36 36 36 36 38 38 38 39 43 43 43 44 45 46 50 50 50 50 52 52 54 54 54 54 54 58 58 73 78 Page VI. Inhibition of Sugar-Amine Browning by Glutamic and Aspartic Acids . . . . . . . . . 80 A. Glutamic Acid . . . . . . . . . . . . . . 80 B. Aspartic ACid O O O O O O O O O O O O O O 82 C. Inhibition of Non-Enzymatic Potato Browning by Aspartic and Glutamic Acid . . 82 SUMMARY AND CONCLUSION . . . . . . . . . . . . . . . . 90 A. sugars O O O O O O O O O O O O O 0 O O O O O O 90 B. Amino Acids . . . . . . . . . . . . . . . . . 91 C. Activation Energy . . . . . . . . . . . . . . 91 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . 92 vii Table 1. LIST OF TABLES Page Amino acid, sugar, and pH combinations in sugar-amino acid reaction for 58 hours at 60°C 0 O O O O O O O O O I O O O O I O O 0 42 Amino acid or mixture of amino acids (1:1 M ratio), sugar and pH combinations in inhibited browning reactions per- formed at 60°C for 58 hours . . . . . . . . . 43 Slopes, y-intercepts, and correlation coefficients (r) of the linear regres— sions between color (A4 ) and amino acid concentration for gge reaction of glucose with several amino acids . . . . . 57 Slopes, y-intercepts, and correlation coefficients (r) of the linear regres- sions between color (A4 ) and amino acid concentration for gge reaction of fructose with several amino acids . . . . . 63 Slopes, y-intercepts, and correlation coefficients (r) of the linear regres- sions between color (A4 ) and amino acid concentration for Ege reactions of lysine with glucose and fructose . . . . . 66 SlOpes, y-intercepts, and correlation coefficients (r) of the linear regres- sions between color (A4 ) and amino acid concentration for Ege reactions of arginine with glucose and fructose . . . . 69 Analysis of variance showing the effect of the nature of amino acid, amino acid concentration, pH, and their interactions on the color produced by the reaction of fructose with lysine and arginine . . . . . . . . . . . . . 71 viii Table Page 8. Analysis of variance showing the effect of the nature of amino acid, amino acid concentration, pH, and their interactions in the reaction of glucose with lysine and arginine . . . . . . . 72 9. Kinetic data for the lysine-glucose, and lysine-fructose reaction . . . . . . . . . 76 10. Effect of various concentrations of glutamic and aspartic acids on the caramelization of fructose (0.4 M) and glucose (0.4 M), at 60°C and 58 hours reaction time in buffer solution of pH 8 . . . . . . . . . . . . . . . 79 11. Hunter color difference values for fried potatoes treated or untreated with aspartic and glutamic acid prior to frying . . . . . . . . . . . . . . . . . . 88 ix Figure l. 12. 13. 14. 15. 16. LIST OF FIGURES Schiff's base production from condensation of carbonyl and amino groups . . . . . . . Production of glycosylamine from Schiff's base cyclization . . . . . . . . . . . . . The Amadori rearrangement . . . . . . . . . Melanoidins formation from Amadori compounds General overview of the Maillard reaction . An example of Strecker degradation reaction Acid degradation reactions of D-glucose . . Isomerization reactions (Lobry de Bruyn and Alberta van Ekenstein isomerization) . Degradation of enediol to lactic acid . . . Reaction of aldo sugars with bisulfite ion . Sulfonation of sugar derivatives by sulfur dioxide to produce 4-sulfohexosulose . . . Possible mechanism of antioxidation by reductones in oil . . . . . . . . . . . . An ampoule-holder made of plexiglass. Capacity = 86 ampoules . . . . . . . . . . Assembly of the sample processing . . . . Flowchart of sample preparation . . . . . . Absorption spectra of glucose-lysine reaction . . . . . . . . . . . . . . . . . Page 10 11 14 15 16 22 23 33 37 40 41 47 Figure 17. 18. 19. 20. 21. 22. 23. 24. The ratio of absorbance at 282 nm to absorbance at 450 nm versus time for the reaction between lysine and glu- cose in phosphate buffer pH 8, ionic strength 0.2 at 60°C Color formation by the reaction of a—amino-n-butyric acid, glycine, serine, at pH 8 Color formation by the reaction of phenylalanine with glucose (0.4 M) and I alanine, and methionine (0.02 to 0.04 M) with glucose (0.4 M) for 58 ionic strength 0.2, hours at 60°C fructose (0.4 M) at pH 8, 0.2 for 58 hours at 60°C Color formation by the reaction of ionic strength a-amino-n-butyric acid, glycine, alanine, serine, at pH 7 Color formation by the reaction of glycine with glucose (0.4 M) at pH 6, ionic strength 0.2 for 58 hours at 60°C Color formation by the reaction of I and methionine (0.02 to 0.04 M) with glucose (0.4 M) for 58 ionic strength 0.2, hours at 60°C glycine, u-amino-n—butyric acid, methionine, Color formation by the reaction of serine, and alanine (0.02 to 0.04 M) with fructose (0.4 M) at pH 8, ionic strength 0.2 for 58 hours at 60°C glycine, phenylalanine, alanine, o-amino-n-butyric acid, and methionine (0.02 to 0.04 M) with ionic fructose (0.4 M) at pH 7, strength 0.2 for 58 hours at 60°C Color formation by the reaction of glycine with fructose (0.4 M) at pH 6, ionic strength 0.2 for 58 hours at 60°C xi Page . . 49 51 . . . 53 . 55 . . 56 . . 59 . 60 . . . 61 Figure 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. Effect of pH on the color development in the reaction between (a) glucose (0.4 M) and (b) fructose (0.4 M) with different amino acids (0.04 M) at 60°C, ionic strength 0.2 and 58 hours . . . . . . . . Color formation by the reaction of lysine with glucose (0.4 M) and fructose (0.4 M) at pH 6, 7, and 8, ionic strength 0.2 after 58 hours at 60°C . . . . . . . . . . . . Color formation by the reaction of arginine with fructose (0.4 M) at pH 6, 7, and 8, ionic strength 0.2, for 58 hours at 60°C . . Color formation by the reaction of arginine with glucose (0.4 M) at pH 6, 7, and 8, ionic strength 0.2, for 58 hours at 60°C . . Effect of pH on color development in the systems glucose-arginine, glucose-lysine, fructose- arginine, fructose-lysine, after 58 hours at 60°C. Concentrations: sugars 0.4 M, amino acids 0.04 M . . . . . . . . . . . . . . . . Color formation during the reaction of lysine (0.04 M) with glucose (0.4 M) at pH 8, ionic strength 0.2, at 50°, 55°, and 60°C . . . . . . . . . . . . . . . . . . . Color formation during the reaction of lysine (0.04 M) with fructose (0.4 M) at pH 8, ionic strength 0.2, at 50°, 55°, and 60°C . . . . . . . . . . . . . . . . . . Arrhenius plot of the lysine-glucose and lysine-fructose reaction data . . . . . . Molecular structure of (a) glutamic, and (b) aspartic acids . . . . . . . . . . . . . Effect of glutamic acid concentration (0.02 to 0.04 M) on the retardation of the browning reaction between lysine (0.02 to 0.04 M)-fructose (0.4 M) and lysine (0.02 to 0.04 M)-glucose (0.4M) at pH 8, ionic strength 0.2, after 58 hours reaction time at 60°C . . . . . . . . xii Page 62 65 67 68 70 74 75 77 80 81 Figure Page 35. Effect of lysine concentration on color development in the glutamic acid- fructose system at pH 8, ionic strength 0.2, after 58 hours reaction time, and at 60°C . . . . . . . . . . . . . . . . . . . 83 36. Effect of lysine concentration on color development in the glutamic acid—glucose system at pH 8, ionic strength 0.2, after 58 hours reaction time, and at 60°C . . . . . 84 37. Effect of lysine concentration on color development in the aspartic acid-fructose system at pH 8, ionic strength 0.2, after 58 hours reaction time, and at 60°C . . . . . 85 38. Effect of lysine concentration on color development in the aspartic acid-glucose system at pH 8, ionic strength 0.2, after 58 hours reaction time, and at 60°C . . . . . 86 39. Reflectance spectra of untreated and treated (with glutamic and aspartic acids) fried potatoes at 210°C for 7 minutes . . . . . . . 89 xiii INTRODUCTION The carbonyl-amine interaction which results in the formation of brown pigments in foods has attracted the attention of many investigators since the first pertinent observations of Maillard in 1912. This aptly named brown- ing reaction is desirable (even obligatory) in some foods (e.g., cakes, bread, coffee, and beer) as it results in both an attractive color and a pleasant flavor. In other cases, browning is not desirable because it results in discoloration (e.g., in dried or canned fruits and vege- tables, white wine and dried eggs). Furthermore, the non- enzymatic browning of foods may affect the nutritional value of foods by reducing the availability of certain amino acids; certain products of the Maillard reaction may be even hazardous for health. To prevent browning reactions, physical methods, such as the control of the moisture content and storage temperature of foodstuffs, has been used. Chemical methods have also been investigated. Sulfur dioxide has been used in the prevention of the browning of dried fruits for cen- turies. Sorbitol, a substitute for glucose, has been reported to cause less browning. An aluminum salt of phytic acid has been considered as a retardant of non- enzymatic browning. A careful review of the literature indicated that a systematic investigation of the reaction of different amino acids with the two most common reducing sugars, i.e., glu- cose and fructose, under varying conditions. A comparison of fructose to glucose in terms of their potential for causing non-enzymatic browning seemed especially to be necessary in View of the large quantities of high fructose corn syrups that are being used currently in the food industry (1.5 billion pounds in 1973). In the present study, the reactivity of glucose and fructose with ten individual amino acids was investi- gated under a variety of experimental conditions. During the course of those experiments, the observation was made that aspartic acid and glutamic acid failed to react and even inhibited the reaction of the other amino acids. Consequently, these two dicarboxylic amino acids were then tested as browning inhibitors in a food commodity, fried potatoes. LITERATURE REVIEW Non-Enzymatic Browning The formation of brown pigments (melanoidins) while heating a solution of glucose and glycine was first observed in 1912 by the French chemist L. C. Maillard. Similar dis— colorations have been observed when nitrogen-containing compounds such as ammonia, amines, amino acids, peptides, and proteins react with sugars, aldehydes or ketones. Browning occurs also during heat pyrolysis of sugars, which is a different phenomenon though the appearance is the same. It is now known that the Maillard "carbonylamine" reaction, a major cause of browning during the prolonged storage or heat processing of food, involves a condensation between the a-amino groups of the amino acids or proteins and the carbonyl group of the reducing sugars. After the first interaction of the amino group with the carbonyl group, a series of complex and continuous reactions occur. The final products of these reactions are colored, soluble and insoluble pigments called melanoidins. A. Production of N-Substituted Glycosylamine The condensation product of an amine and an Opened ringed form of sugar is one molecule of water and 3 Schiff's base. Schiff's base has never been isolated from the reaction medium (Figure 1). The stoichiometry of the reaction is one amine to one sugar (Hannan and Lee, 1952). However, reactions of up to 6 moles of amine with one mole of sugar have also been reported (Erickson, 1953). Subse- quent cyclisation of Schiff's base produces N-substituted glycosylamine (Figure 2). The initial reaction is reversible, but irreversible reactions (not chemically irreversible) soon follow (Haugaard et al., 1951). H O H N-R \ / \ c ’ c 4’ | l (CHOH)4 + RNH2 -——- (ff-10H)4 + H20 CHZOH CHZOH D-Glucose Schiff's base Figure l. Schiff's base production from condensation of carbonyl and amino groups. B. Production of Amadori Compound N-substituted glycosylamine goes through a series of rearrangements in the presence of protons with an iso- merization often called the Amadori rearrangement (Figure 3). N-substituted l-amino 1-deoxy-2-ketose is called the Amadori product. The "Amadori" rearrangement involves the transition from an aldose to a ketose sugar derivative. Amadori rearrangement products are more stable than the original glycosylamines. Reaction products up to this stage are H R \N / I H- C I I Schiff| 5 base V___*_ (fHOH) 3 (I) HO-C I CHZOH N-Substituted glycosylamine Figure 2. Production of glycosylamine from Schiff's base cyclization. colorless. The color starts to be produced after the third stage, which is the production of melanoidins from the Amadori compound. C. Formation of Pigments The "Amadori" rearrangement products undergo a browning decomposition in aqueous solutions. The rate of browning is highly enhanced in the presence of amines. The reaction pathway and its mechanism after the formation of ketose are not clearly understood. Reactions at this stage involve many condensations and polymerization reactions, giving brown pigments or melanoidins at the final stage of reaction. The main reactions occurring are thought to be aldol condensations, the formation of heterocyclic nitrogen com- pounds such as pyrroles, pyridines, imidazoles, and aldehyde-amine polymerization (Hodge, 1953). A simplified pathway of melanoidin formation ori- ginating from Amadori compounds is shown in Figure 4 H R \ ./ b H-f___' CH (I OH)3 o H—c_J I CHZOH N-substituted glycosylamine R H \ ./ N l H - C - H I HO — C I HO - C - H I H - c - on O I H - C l CHZOH fructoseamino acid +H+ *A _ cation of Schiff's (l-amino l-deoxy-Z-ketose) A (CHOH) 4 | CH base Figure 3. The Amadori rearrangement. 2 OH .I R H \ /' N I -H+ CH g _ I JOE I (finou)3 CHZOH N-substituted l-amino l-deoxy 2-ketose (enol-form) H \ ./ H - - H N l c I c = o I c (IHOHI3 CH2OH N-substituted 1-amino l-deoxy- 2-ketose (keto form) I ”C - N \ C - OH enol form I (SHOHI3 CH OH H - C 8 0 I C 8 0 I CH . 2 (CHOHI2 CHZOH dehydration H-C=O I Cco I CH II CH I CHOH I CHZOH \ . o tfio a" i V furfural or Hydroxymethy furfural (m) ¢ N82,, Figure 4. Amadori Compound H o-dicarbonyl intermediate I_. ‘4' @- 94> H C NI’H ’ \ 2 R C - O | keto form (CHOHI3 CHZOH CH3 I C - O I C 8 O I (CHOH)2 CH20H enolization CH . 3 C 8 0 I C - OH H C - OH I CHOH I CHZOH pyravaldehyie. diacetyl, a:e::l hydrcxyd .‘ a:e:;.. -\ t III/I’l’reductcnc Melanoidinsja’llll’ Melanoidins formation from Amadori compounds. 8 (Markakis, 1979). Melanoidins can originate either from the keto form or the enol form of the Amadori compound. 1. From the Keto Form The keto form of the "Amadori" rearrangement product decomposes to a methyl-a-dicarbonyl derivative by the elimination of the amino group. The dicarbonyl compound can undergo various enolizations and ketolizations to yield numerous isomerized intermediates. The intermediates either condense with amino compounds (aminated) to form melanoidins, or break down first, then react with amino compounds and produce melanoidins. 2. From the Enol Forms Deamination of the enol form of the Amadori compound yields 3-deoxy-a-dicarbony1 intermediates. Dicarbonyl inter- mediates are subjected to varying degrees of molecular dehydration. Furfural (2-furaldehyde) and S-hydroxymethyl- 2-furaldehyde (HMF) could be the product of this dehydration. The products of this dehydration condense with amines and produce melanoidins. In summary, in the "Amadori" rearrangement, a 1- amino l-deoxy 2-ketose type molecule is formed from an N- substituted glycosyl amine. The Amadori compound is a turn- table for the reaction which can progress in three different ways: 1. Reactions producing fission products, small carbonyl molecules, such as pyruvaldehyde, diacetyl, acetol and hydroxy-diacetyl. 2. Severe dehydration results in furfurals and 5- hydroxymethylfurfurals. 3. Moderate dehydration produces reductones and dehydro- reductiones. Some of the Maillard reaction products are water soluble and others are water insoluble. The soluble products are called premelanoidins and the insoluble ones melanoidins. Figure 5 shows a general outline of the reactions (Adrian, 1974 and 1973). Strecker Degradation Strecker degradation is not primarily concerned with pigment production; however, it provides reducing compounds essential for brown color formation. The Strecker degrada- tion is more important for flavor development than for browning. It involves the degradation of a-amino acids and the formation of a corresponding aldehyde containing one carbon atom less, which is lost as a carbon dioxide. The degradation is carried out by o-dicarbonyl compounds (Figure 6). According to Schonberg, et al. (1948), only carbonyl compounds containing the structure, 0 O C [ C - C H _ ln _ C _ 10 sugar + amino acid I I N—substituted glycosylamine Amadori rearrangement l-amino 1-deoxy 2-ketose 0) a «4 :3 strong dehydration o c 3 Fission mild dehydration o E o u m ,3 Furfurals, Reductones, '3 dehydrofurfurals, and dehydroreductones 3 banal aromatic substances m pyruvaldehyde, streker diacetyl, acetol degredation and banal aromatic substances aldehyde + CO2 (specific aromatic substances) L____ polymerization and insolubilization LMelanoidinsl Figure 5. General overview of the Maillard reaction. ll .cofluommu coflumpmumop meomuum mo mamfimxm Cd .m muswflm wasnmoam O m _¢ 00 + U Ism + \ \ m m _ Show oumx \m I U I U I m mcouosvmu mar w pfiom ocHEMIa chonumowo mooo I mo I‘m +.m I o I o I m N . = 3 m2 0 O ENOM Hoco m I O u U I m m _ _ 12 where n is zero and/or an integer, can initiate the degrada- tion. If the a-dicarbonyl compound is an osone such as 3- deoxyosone, O 0 II I [H - C - C - CH2 - (HCOH)n - CHZOH] O 0 II II or pyruvaldehyde (H - C - C - CH3) the a-amino nitrogen is not lost in the reaction but only transferred to the a-position of the osone moiety (Hodge, 1959). Caramelization Caramelization is another example of nonenzymatic browning. It is the degradation of sugars either in a solution or dry while in the absence of nitrogen-containing compounds. This degradation is an acid, base or salt catalyzed process. Many substances discovered during sugar degradation are also found during Maillard reactions, so the two pro- cesses should be clearly distinguished. The distinction between sugar degradations and the Maillard reaction is possible upon the black precipitate collected at the end of the reaction. Sugar decomposition produces nitrogen-free precipitation, whereas the Maillard reactions generate nitrogen-containing products of variable nitrogen content. 13 Alkaline and acidic degradation of sugars such as starch, sucrose, glucose, and fructose, follow a common pathway, and the following steps have been proposed: a. Acid degradation: First, 1,2- enol is formed from the correSponding sugars, then it is followed by a molecular dehydration in a series of stages to 5-(hydroxymethyl)-2-furaldehyde (HMF). Hydroxymethyl furaldehyde is a precursor of the pigment (Wolfrom et al., 1948). The intermediates of these reactions are 3-deoxyaldos-2-ene, 3-deoxyosulose, and osulos—B-ene (Figure 7). b. Base degradation: The effect of aqueous alkali is very complex. It involves isomerizations, fragmentations, and intramolecular oxidation and reduction. Isomerization occurs by way of an enediol intermediary. Therefore, a solution of glucose in an alkali media yields fructose and mannose, and finally 1,2-enediol. These reactions are known as the Lobry de Bruyn-Alberta van Ekenstein transformations, as shown in Figure 8 (Kearslay, 1975). The fragmentation is followed after the 1,2-enol formation. It involves the scission of the parent sugar into smaller fragment molecules, such as dihydroxy acetone, glyceraldehyde, formic, acetic, and lactic acid (Figure 9). Following the fragmentation, condensation between ketones and aldehydes generates polymers and browning products. l4 H l HO - C H - C = O H - C = O I I l (H - C - OH) O ring opening (H - C - OH) -H O C - OH I I w I I H - C CH OH C - H I 2 I CH OH H - C - OH 2 I H - C - OH I CHZOH 3-Deoxyaldose-2-ene rearrangement I H - C = O H - C = O I I _ I l H CH CH H CH CH2 \ I \\l | I HO - C - C C-C -H O CH -H O H - C - OH 2 2 / \/ I4 I <— | H O O cyclization H - C - OH H - C - OH I I CHZOH CHZOH 5-(Hydroxymethyl)-2-furaldehyde 3-Deoxyosulose Figure 7. Acid degradation reactions of D-glucose. CHO H I H - C - OH I HO - C - H HO I V—é H - C - OH H I H - C - OH H I CHZOH D-glucose HO H 15 - C - OH II C - OH I - C - H l - C - OH I - C - OH I CHZOH Enediol fHZOH C = O I - C - H I - C - OH I - f - OH CHZOH D-fructose CHO HO - G - H HO - G - H H - G - OH H - G - OH GHZOH D-mannose Figure 8. Isomerization reactions (Lobry de Bruyn and Alberta van Ekenstein isomerization). 16 glucose, fuctose, mannose I / Enediol H-C-OH II C - OH I HO-C-H I H - C - OH I H-C-OH I CHZOH I I CHO CHOH I " CHOH + C - OH I I CHZOH CHZOH glyceraldehyde Triose enediol OH HO - C ” \H I C = O I CH3 pyruvaldehyde COOH I CHOH I CH3 D-L Lactic acid Figure 9. Degradation of enediol to lactic acid. 17 Factors Affecting the Maillard Reaction The color formation in sugar-amino acid mixtures has been chosen as an index to determine the effect of various factors in contributing to the color development. Various factors such as temperature, pH, and water content can highly affect the extent and browning rate. The Maillard reaction has been shown to have a high temperature coefficient. An increase in thermal energy not only helps to break down sugars, but it also increases the rate of collision between the fragments produced during degradation. Activation energy (Ea) and time for 50% of the reaction to take place (t ) for polylysine-glucose 1/2 systems are reported to be 22 kcal/mole, and 400 min. respectively (Hansen, et al., 1979). The activation energy for ketosamine fructose-glycine formation is 26 kcal/mole (Reynolds, 1959). In model systems, the rate of browning increases 2 to 3 times for each 10°C increase of tempera- ture (Shallenberger, 1975). An increase in pH favors the Maillard reaction. The rate of the reaction is extremely high at the alkaline side of the pH. An example of alkaline food is the egg. In order to obviate browning and discoloration during the prepa- ration of egg powders, the pH is lowered by the deliberate addition of acid prior to dehydration. Willits et a1. (1958), observed that heating glucose with DL-alanine for 20 minutes at 114°C in an aqueous solution up to pH 6 failed to cause measurable browning. The greatest effect of pH is 18 with solutions containing diamino acids and the least effect is with the a-amino acids (Underwood, 1959). The browning reaction in solutions follows the law of mass action, i.e., the rate of the reaction increases by increasing the concentration of the reactant(s). The brown— ing reactions in foodstuffs are more complex. The water activity, (aw), in the system affects the browning rate. At a very low and a high water activity, the browning rate is minimal due to the solubility limitation and the dilution effect, respectively. The water activity that affects the maximum rate of browning varies for different foods. Pea soup mix at a 70% relative humidity has shown a maximum rate of browning (Labuza, 1970). During the dehydration of diced potatoes, at a 15% moisture level (dry basis), the rate of browning is reported to be maximum. Increasing the moisture level to 33% causes the browning rate to decrease 1.5 fold. Similarly, decreasing the moisture level to 4.9% will cause a 6 fold decrease in the browning rate (Hendel et al., 1955). In a model system (described by Hannan and Lea, 1952), consisting of a-N-acetyl-L-lysine and glucose, the rate of loss of the amino groups reached maximum at 40% relative humidity. Inhibition of the Browning Reactions Non-enzymatic browning is usually considered as a deteriative process, particularly by food processors. The loss of the nutritive value of food cannot be compensated 19 for by adding the lost nutrients to the food during process- ing, because the non-enzymatic browning may continue even during storage. Because of the complexity of the food sys- tems, controlling the browning is not without problems either. Primarily the process can be slowed down by simply decreasing the temperature, controlling the moisture con- tent, and by gas packing the food, or by controlling the pH. Non-enzymatic reactions have high activation energies; therefore, lowering the temperature will slow down the reactions. Non-enzymatic browning is accelerated at a high pH value. Therefore, decreasing the pH reduces the browning rate. The processing and storage of food at certain moisture levels in which the non-enzymatic reaction is minimized is another way to control the non-enzymatic browning. All of these ways seem theoretical, and most of the time not practical. Certain dehydrated or highly acidic foods are not marketable. Keeping food at low temperatures or refrigerated all of the time is not economical, and pro- cessing particular foods below the required temperature is not acceptable. Chemical Methods of Inhibition A wide variety of inhibiting chemicals have been used to retard non-enzymatic browning such as, 20 1. Aluminum salt of phytic acid, introduced by Mitsui Totsau Chemicals, Inc., 1976 (Japanese patent). 2. Sorbitol. Sorbitol has been reported to cause less browning (British patent, 1966). This is due to a conversion of the aldehyde groups to alcohol and consequently, a loss of reducing power. On the other hand, other properties such as viscosity, sweetness, and osmotic pressure remain unchanged (Kearsley, 1978). Hydrogenation lowers the sweet- ness and hygroscopicity of glucose and renders it less fermentable. 3. Calcium Chloride. A 0.02%-0.84% calcium chloride solution retards considerably non-enzymatic browning in potatoes (Simon et al., 1955). The inhibitory effect is suggested to be due to the blocking of the amino group. Simon concluded that this blockage might be due to the chelate compound formed with a-amino carboxylic acid. It is also anticipated that the application of an excessive amount of calcium salt may affect the texture of potatoes; consequently, a longer cooking time is required. A positive sinergestic effect was also reported by Simon when the calcium chloride was used along with sulfite. 4. Hydrogen Peroxide. According to Pokorney et a1. (1973), hydrogen perox- ide hinders, but not entirely stops, the formation of brown pigment resulting from the interaction of glycine and 21 diacetyl. Actually, hydrogen peroxide destroys the amino acid rather than blocking it. NH2 - CHz-COOH + H202 + NH4OH + HCHO + CO2 Pokorny et al., reported that hydrogen peroxide also bleaches the brown pigments. 5. Butylhydroxyanisole (BHA) BHA has been shown to retard the oxidation of lipids by interfering with the free radical formation. It has also been shown to reduce amino acid and carbohydrate destruction caused by thermal stress (Magna and Monte, 1977). 6. Mercaptans Mercaptans are reducing agents, and they can prevent non-enzymatic browning (Joslyn and Braverman, 1954). Hodge (1953) suggested that the inhibition might be due to the ability of mercaptans to reduce reductones and eliminate the active dehydro reductones. Ingles (1963) reported that the mercapto derivatives of reducing sugars when reacting with amino acid form an imperceptible color compared with non-derivatized sugars. The use of mercaptans is limited because of its unpleasant odor. 7. Sulfur Dioxide Sulfur dioxide is the most widely and oldest com- pound used to retard non-enzymatic browning. It is also used as a preservative, a reducing agent, and for the pre- vention of enzymatic browning. It is used in the form of its alkali metal sulfites, bisulfites, or as sulfurous acid. 22 Sulfur dioxide has two series of salts, sulfite (50:) and bisulfite (H803). Two tautomers of bisulfite are known. H 0" O" \ / ’ O-S H-S-O \ \ O O (a) (b) At high pH, form (b) dominates to form (a) (Cotton and Wilkinson, 1966). Form (b) has more reducing power than form (a). Sulfur dioxide is a very strong reducing agent, and has a bleaching character. It can be added to aldehydes and to ketones, and it similarly reacts with reducing sugars and unsaturated carbonyls which are the products of sugar dehydration and reductones. Ingles (1959) prepared carbonyl bisulfite by reacting bisulfite ion with aldose sugars, such as glucose, galactose, mannose, xylose, and arabinose, at pH 4 (Figure 10). The sulfonated product would release sulfur dioxide upon acid hydrolysis. H 0 SO \//- I C + HSO -—————- HO — c - H Figure 10. Reaction of aldo sugars with bisulfite ion. 23 Conjugated unsaturated carbonyl compounds, which are important in the development of chromOphors, react with an excess of bisulfite both at the olefinic and carbonyl group at 60°C (McWeeny and Burton, 1962, and 1963). Bronis- law and McWeeny (1974) identified 4-sulph0pentosulose in dehydrated, sulphited cabbage. Ingles (1966) reported bisulfite reactions with glucose at pH 6.5. The sulfonic acid derivative (4-sulfo- hexosulose) is obtained either by adding it to the un- saturated osone or by substitution in 3-deoxyglucose (Figure 11). CH0 CH0 CH0 I I C = O C = O C = O I I I CH2 H2803 CH2 e H2803 EH CHOH CHSOBH CH I I I CHOH CHOH CHOH I I I CHZOH CHZOH CHZOH Figure 11. Sulfonation of sugar derivatives by sulfur dioxide to produce 4-sulfohexosulose. The sulfonic acid derivative obtained at this pH is very stable, and upon reaction with acid or base, even in the Monier-Williams distillation, does not release its sulfur dioxide. According to Hodge (1953), the amine derivative of carbonyl disulfite prepared by Ingles (1959) may have a role 24 in inhibiting those browning reactions dependent upon the Amadori rearrangement. The use of sulfur dioxide for the prevention of non- enzymatic browning works alright, but its safety for human consumption is now being questioned. To prevent the brown- ing of white wine, Safar (1976) fermented must pressed from grapes with sulfur dioxide, producing yeast such as Saccharomyces ellipsoideus and S. carlsbergensis. Safar states that wine produced in this way does not pose the harm for humans that the sulfur dioxide apparently does. Nutritional Aspects of the Maillard Reaction The most important aspect of the Maillard reaction is its role in the nutritional value of foodstuffs. It reduces the nutritional value of the free amino acids, proteins, and sugars. Its products also are suspected to be toxic. A. Amino Acids An amino acid linked to a sugar is resistant to enzymatic hydrolysis. But the regeneration of the initial sugar and the amine by means of chemical hydrolysis is possible. Therefore, from a nutritionist point of View, the amino acid is destroyed right from the beginning of the Maillard reaction (Cook et al., 1951; Evans and Butts, 1948) and from a chemist point of view it is "blocked." The behavior of amino acids in browning reactions is different. The closer the amino group to the carboxylic group, the less active is the amino acid in brown color 25 production. Lysine generates six times more intense color compared to norleucine, having only one a-amino group. Similarly, y-amino butyric acid contributes 10 times more color compared to its aisomer (Underwood et al., 1959). The chain length of the amino acids has an effect on the coloring of the sugar solution. The longer the chain length in an homogeneous series, the more intense the color development (Lento et al., 1958). Friedman and Kline (1950) reported that the biologi- cal value of food will be reduced as a result of the reaction of amino acids with glucose. However, during food experi— ments Friedman and Kline observed that the amino acid-glucose complex is more available to microorganisms than to rats. B. Proteins The interaction of sugars and proteins is less under- stood because of the complexity of their systems. The destruction of proteins by the Maillard reaction varies with respect to the amino acids representing the proteins, e.g., first the N terminal of the protein is attacked, then the basic amino acid side chain, particularly the lysine, is damaged. The sulfur containing amino acid, methionine, is destroyed last by the Maillard reaction. When soybean pro- tein was autoclaved alone, there was no evidence of destruc- tion; however, up to 97% less methionine was liberated by enzymic digestion in vitro from protein autoclaved with sucrose or glucose (Evans et al., 1949b). Evans et a1. 26 (1948), also reported that adding sucrose to soybean protein, and autoclaving it, causes about a 50% destruction of lysine level. Among the protein-bounded amino acids, lysine is reported to be the most sensitive to the Maillard reaction, and it can become a protein limiting-factor after the Maillard reaction occurs (Adrian, 1972). This is particularly important in processed milk (Greaves et al., 1933; Mauren et al., 1955; McDonald, 1966; and Payne-Betha et al., 1959). Lysine is the first limiting amino acid in cereals. A slight heat treatment of cereal products is enough to block and may destroy a large amount of lysine. Bread crust and toasted bread has much less nutritional value than untoasted bread (Palamidis and Markakis, 1979; Rosenberg and Rehdenburg, 1951). In vegetables, the protein destruction is fairly low. This is because of the lack of active forms of carbo- hydrates. Among vegetable products, cotton meal-cake seems to be more labile to heat. A 20 min. autoclaving of cotton meal-cake reduces its biological value (B.V.) by 25% (Balica et al., 1959). C. Sugars Reducing sugars are severely damaged by the Maillard reaction (Allen et al., 1973; Haugaard et al., 1951). Hannen et al., (1952) reported that in a model system consisting of a-N-acetyl-L-lysine and D—glucose, at relative humidity of 60% the loss of glucose reached a maximum. 27 The initial hydroxyl configuration and the speed of dehydration are important in the browning reaction. Burton (1963) suggests that the rate of production of brown pigment is closely related to the initial configuration of the aldose. According to Cantor (1942), the order of reactivity of the aldohexose series are in the ascending order of glu- cose, mannose and galactose. In ketose/amines, the initial chromOphoric develOpment is greater than for aldoses (Burton et al., 1963). In general: a. Short chain sugars have greater reactivity. If a hexose destroys 42% of lysine in solution, an equimolar pentose causes 70% destruction of lysine in the same solu- tion. On the basis of their reactivity, pentoses are more reactive than hexoses, followed by disaccharides. b. The reactivity of sugars also depends on the type of protein involved. A hexose reacting with lactalbumin destroys 64% of lysine moiety, whereas the same hexose destroys only 59% lysine of soybean globuline. c. Different isomers of the same sugar behave differ- ently when reacting with the same amino acids (Rubenthaler et al., 1963). Foods low in carbohydrates are subject to less nutritional damage from the Maillard reaction. In foods high in carbohydrates, most of the available amino acids and proteins are destroyed by the carbonyl amino interaction. 28 This somehow would explain the reason for the low sensitiv- ities of legumes to heat damage. According to Evans and Butts (1949), heating soybeans alone will cause a 3% reduc- tion of its lysine; however, the addition of sugar may decrease the lysine content up to 47%. Schroeder et a1. (1961), report that adding 5% glucose to meat, and auto- claving it at neutral pH for 30 minutes reduces its diges- tible lysine by 11%. Sugars with an alcohol group may not cause Maillard reaction (Pomerant et al., 1962). Polyhydric alcohols in general and glycerol in particular may contribute to the browning reaction when they are used as humactants in inter- mediate moisture meats (Obanu et al., 1977). Obanu's results do not rule out the possible mild oxidation of polyhydric alcohols. D. Toxicity The production of melanoid substances during Maillard reaction not only occurs at the expense of nutrients, such as sugar, proteins, and amino acids, but also the melanoidins affect the utilization of the nutrient. Melanoidins inac- tivate amylolytic malt enzymes (Zabrodskii et al., 1960). To obviate this, Zabrodskii suggested fermentation at a temperature which considerably lowers the rate of sugar- amino reactions. Melanoidins modify an enzymatic proteolysis in vitro performed with pepsin and trypsin. 29 Some commercial solutions designed for parenteral nutrition contain sugar-amino acid complex due to sterili- zation of the solutions. The presence of these complexes resulted in a 2 to 5-fold increase in the urinary excretion of Zu, Cu and Fe in both adult and infant subjects. Stegnik et al. (1977) reported that premelanoidins can transfer across the placenta during infusion into the Rhesus monkey. Premelanoidins are not only antinutritional, but toxic as well. The toxicity of the premelanoidins varies with the amino acids involved to produce them. Among the amino acids, lysine premelanoidins are the most toxic. Reductones are common products of sugar pyrolysis and the Maillard reaction. Reductones become more toxic when they are bound to an amino group. For mice the LD is 850 50 mg/living weight for nitrogen-free hexose reductones, whereas the LD50 value is reduced to 300 mg/kg, if hexose reductones are methylaminated (Ambrose et al., 1961). Premaloidins provoke histological disturbances in the liver. E. Flavor Produced by the Amino-Carbonyl Reaction The products of the Maillard reaction cannot totally be condemned, because they are also responsible for desirable color such as in bread crust, cookies, and for pleasant flavors, such as in roasted nuts, coffee, potato chips, and many other food products. So flavor production could be included as a positive effect of the Maillard reaction, at least from the consumer's point of view. 30 The sources of aroma could be either from furfurals of sugar pyrolysis or aldehydes derived from amino acids during Strecker degradation. The aldehydes formed from degradation of alanine, glycine, leusine, methionine, and phenylaline are thought to be acetaldehyde formaldehyde, isoraleraldehyde, methional, and phenylactaldehyde respec- tively (Johnson et al., 1966; and Self et al., 1963). The development of flavors depends on the nature of amino acids which react with sugars; however, the exact stage of Maillard reaction, water content, pH, and temperature are also responsible for them (Adrian, 1973; and Herz et al., 1960). Bondarovicy et a1. (1967), report that the aroma complex of coffee consists of more than two hundred con- stituents. The volatile components responsible for the aroma of ground coffee were identified by Merritt et a1. (1963), as aldehydes, ketones, esters, heterocyclic com- pounds, sulfur compounds, alcohols, and nitriles. Thus they concluded that heterocyclic compounds, particularly 2-methy1 furan and furan, are among the significant com— pounds in the complex which make up typical coffee aroma. According to Wiseblatt et a1. (1960), the compounds responsible for bread flavor are acetaldehyde, acetone, crotanaldehyde, diacetyl, formaldehyde, furfural, hexanone-Z, heptanone-3, isobutyraldehyde, 2-methy1butana1, methyl- glyoxal, methylketone, n-valeraldehyde, and pyruvic acid. Volatile aldehydes formed by non-enzymatic browning 31 reactions during baking are a major factor in bread flavor. Linko, Y.-Y. et a1. (1963), examined free amino acid con- tent in the crust and crumb of baked bread. They observed a decrease in amino acid content in the crust compared with I the crumb regardless of the type of sugar used. This fact would explain the different flavors in the two parts. Linko et al. concluded that the decrease in free amino acids in crusts, together with the formation of several aldehydes, suggest the importance of Maillard-type browning in flavor production. A gradual loss of carbonyl compounds from the crust parallels the aging of bread. In fresh bread a few amino acids are especially responsible for the flavor, e.g., degradation product from leucine (isovaleraldehyde) (Branes et al., 1948). The products of the reaction between glycerol and proline also generate a bread-like aroma (Hunter et al., 1966). Methylbutanal and methional, products of the Strecker degradation of methionine, are responsible for cheddar cheese flavor (Day et al., 1960). It is proper to mention that flavor production by the Maillard reaction is not desirable in all cases. This again depends on the reaction condition, for instance, the heating casein mixed with 3% lactose for 40 hrs. at 80°C produces an objection— able gluey flavor (Ranshaw et al., 1969). 32 F. Other Aspects of the Maillard Reaction (Antioxidant Preperty The products of amino acid sugar browning not only take part in the color and flavor development of foodstuffs, but also show antioxidative activity. The antioxidant activity is believed to be due to the reductones formed in the browning reaction between the reducing sugars and amino acids (Griffith et al., 1957). Cooney et a1. (1958), suggested a possible mechanism of antioxidation by reductones in oil. Cooney states there are at least four possible reactions for the mechanism of antioxidation in oils (Figure 12): l. The air oxidation of the endiols to produce a- dicarbonyl compounds; 2. The spontaneous reduction of the o-bicarbonyl com- pounds enediols; 3. The spontaneous oxidation of the o-dicarbonyl com- pounds which is independent of oxygen to produce deep-colored products; 4. The oxidation-reduction reactions between the colored compounds and fat peroxides. Reductones are believed to react with precursors of peroxide- forming compounds rather than directly with peroxides. The features of the reductone-treated oils are long induction periods, slow absorption of oxygen, and low rates of peroxide development (Evans et al., 1958). 33 0 (air) 2 1:0 ‘c-mI (1) ea. ‘f :3 - OH ipontaneous (2) /C§ +H20 (3) + H2202 I reduction 0 f' ' " ' ‘| reductone or I ' polyphenol : : H202 ‘- ————————————— - —- - - .1 ' # I I Dimer and Polymers I I + I Bleached Polymer + H20 t J (4) H202¢ —————————— Figure 12. Possible mechanism of antioxidation by reductones in oil (from Cooney et al., 1958). Alcoholic solutions of some dehydro reductones, such as O - H. maltol O ‘= O isomaltol glycine > alanine, the slope representing color change with amino acid concentration places them in a dif— ferent order: glycine > o-amino-n-butyric acid > alanine. 3. Phenylalanine-Glucose Phenylalanine behaves differently than glycine, alanine, and most other amino acids. The relationship between color development and amino acid concentration is not linear. Color development is affected by amino acid concentration much more strongly at higher levels of amino acid concentration (Figure 19). 53 A450 0.5 I 0'4 '- -I 0.3 .. GLUCOSE . 0-2 - + . FRUCTOSE 0-1 1 0-02 0-03 0-04 PHENYLALANINE I M Figure l9. Color formation 0. the reaction of phenylalamne WI h Iucose (0.4 M) and fructose (0.4 M at pH 8 Lame strength 0. 2 for 58 hours at 60 C. 54 4. Serine-Glucose Serine produces less color than alanine, glycine and a—amino—n-butyric acid under the same conditions of reaction with glucose (Figure 18). 5. Methionine-Glucose Methionine shows very little tendency to interact with glucose (Figure 18). Paton et a1. (1948), in deter— mining the nutritive availability of the amino acids, observed a minimal loss of methionine due to the Maillard reaction, compared to other amino acids. 6. Lysine and Arginine-Glucose Lysine and arginine will be discussed separately as they react with glycine in somewhat different fashion. B. Amino Acids-Glucose Systems atng 7 As anticipated, the rate of color formation was less intense at pH 7 than at pH 8. Again linear relation- ships were obtained between color and amino acid concentra- tion as shown in Figures 20 and 21. The slopes, the y- intercepts, and the correlation coefficients of these rela- tionships appear in Table 3. C. Amino Acids-Glucose Systems atng 6 At pH 6 the above mentioned amino acids produce very little color in their reactions with glucose. The linear relationship between color and glycine concentration 55 1-0 I 0.8 - .I 06 - a-amlno-n- butyric acla 0'2 Ala - d ‘ Met 0.0 w 0-02 0-03 0-04 AMINO ACID CONCENTRATION I M Figure 20. Color formation by the reaction of a-amino-n-butyrlc acid, glycine. serine, alanine, and methionine (0.02 to 0.04 M ) withglucose (0.4 M I at pH 7, ionic strength 0. 2, for 58 hours at 60° C. 56 0-0 L 0-02 0-03 0-04 GLYCINE CONCENTRATION I M Figure 2|. Color formation by the reaction of glycine with glucose (0.4 M l at pH 6, ionic strength 0. 2 for 58 hours at 60° C. 57 hmmm.o Goo.OI mo.m memm.o ome~.0I hv.ma mcfleohnumz nemm.o mmo.OI ma.~a memm.o oamH.OI mo.mH meadow . . . . . . cflom chumusn ommm 0 sec oI as ma Rama 0 ommo o+ No ma IcI02A2mIa mmmm.o Gmo.0I Am.m mmmm.o NASH.OI 05.8H mascaaa Hmm.o NMH.OI m~.om mmma.o mmao.o- o~.om masosao u uomwmoucfl moon u ummwmmucfl moon afloa oasea 5mm mmo .mpflom OGHEm a may now :oflumupcmocoo pflom OGHEM can A m>mm cufl3 mmoosH@ mo :ofluommu adv Hoaoo coo3umn mcowmmmuomu ummcfla wcu mo AHV mucmfloflmmwoo coflumHmuuoo paw .mummououcflIw .mmmonII.m magma 58 is shown in Figure 21, and can be described by the following regression equation and correlation coefficient, Y = 5.825 X = 0.072 and r = 0.9944. II. Sugars In spite of the wealth of data for glucose, there has been little attention paid to the role of fructose in browning in the past. The role of fructose is important today because of the high utilization and demand for fruc— tose in the food industry. A similar series of experiments to those of glucose and various amino acids was performed with fructose and the same amino acids. The results are shown in Figures 22, 23, and 24. In all cases, the amount of color produced by the fructose—amino acid reaction was lower than that of the corresponding glucose-amino acid interaction (Figure 25). The slopes, the y-intercepts, and the correlation coef- ficients appear in Table 4. Lewis and Lea (1950) reported that the loss in free amino-N in the casein—glucose reac- tion was twice that observed in the casein-fructose reac- tion, at 25°C. III. Lysine and Arginine Interaction with Reducing Sugars Lysine and arginine are discussed separately because (a) of their basic side chains, and (b) their dif- ferent behavior in interacting with reducing sugars. Lysine is a diamino amino acid with its second amino group in A450 59 a-amlno-n-butyrlc acld ' 0.02 0.03 004 AMINO ACID CONCENTRATION I M Figure 22. Color formation by the reaction of glycine, a-amino-n-butyric acid, methionine, and alanine (0.02 to 0.04 M l with fructose (0.4 M l at pH 8, ionic strength 0.2 for 58 hours at 60° C. 60 A450 AMINO ACID CONCENTRATION , NI Figure 23. Color formation by the reaction of glycine, phenylalanine, serine, alanine, a- amino-n-butyric acid, and methionine (0.02 to 0.04 M l with fructose (0.4 M l at pH 7, ionic strength 0.2 for 58 hours at 60° C. 61 0.3 1 0.1 0-02 0.03 , 0-04 GLYCINE CONCENTRATION I M Figure 24. Color formation by the reaction of glycine with fructose (0.4 M l at pH 6, ionic strength 0. 2 for 58 hours at 60° C. 62 .222 an EA 2. secs: use .0 .8 s A .2 3.3 ABS 2:3 283% SE :2 «.2 $862.: 3. Em A 2 3: 88:3 13 8258 8:88 2: E €38.28 8.8 2: so In Co 6m...“ .mN s. :2“. 3C In a n o . . o_< a c «0! Son 2.33-5330 «6 6." 1h: 0. .o BOO 2.33.50560 AS 055v a... ad 63 momm.o memo.o+ NO.H I I I mcflumm . . . . . . afloa oAumusn ameo o Gmmo OI NA m swam o seao o+ mm a IcIocAEaIa mmmm.o Amo.o+ 08.4 I I I acaeaamascmnm mmmm.o omoo.o+ oo.~ mamm.o omoo.OI mo.a chcaHa Hama.o mmeo.OI mm.m mmmm.o «moo.0I mm.v maAcoAaumz mmma.o oamm.o+ No.m mmmm.o mmm~.o+ om.HH maaosao H umowmmucfl macaw u DQOOMOUGH macaw I voa oaAea And are .mpflom OCHEO AM on» MOM coflumuucmocoo Uflom ocflEm com A on :uflz Omouosuw mo coflvommu my MOHOO cmm3umn mcowmmmummu HMOGHH on» mo Any mMCOHOHmwOOO coflumamuuoo cam .muomonmucfllm .wOQOHmII.v magma 64 epsilon position. Both the o and s-amino groups of lysine are active in the browning reaction. However, the e-amino group contributes more to the brown color of the Maillard reaction than the o-amino group does. In monoamino acids the color development caused by the Maillard reaction is higher when the amino groups are located further away from the carboxylic group. This is true for amino acids containing two to four carbon atoms. However, for four to six carbon atom amino acids the Oppo- site is true (Lento et al., 1957). Lysine is the most reactive amino acid among the ones which were tested here, and perhaps among all amino acids. The rate of color production in the fructose-amino acid systems, as it was observed for fructose-lysine sys- tems, was higher than that of glucose-amino acid systems at the early stage of browning, but at later stages of the reaction, the glucose-lysine dominates fructose. This might be perhaps caused by the shorter induction period that was Observed for fructose (Figure 31). The color produced by the lysine-glucose and lysine-fructose systems is graph- ically presented in Figure 26. The slopes, y-intercepts, and correlation coefficients for the same reactions are shown in Table 5. The reaction of some primary aliphatic and aromatic amines with a number of simple carbohydrate derivates was studied by Beacham and Dull (1951). These authors concluded 65 .o 08 E :22. mm 3% No £9.83. 2:2 .w can .N .9 Ta 5 A .2 3: 326:: can A .2 v.8 382m 5;, 2:3 .6 8:62.: of .3 52252 3.8 .8 95m: 2 . ZO_._.<¢._.szZOO 90¢ 02.21 2:. 8... S... 3... 8... «as 1 6.9 4 G6 6 :q T; Z. . .. 3 h In I a... 23 ”V O . . 3 mmOPODm—u mwOODJO F $4 I . 96 66 Table 5.--S10pes, y-intercepts, and correlation coefficients (r) of the linear regressions between color (A450) and amino acid concentration for the reactions of lysine with glucose and fructose. pH Glucose Fructose slope 12.0 4.5 6 y-intercept +0.3l8 +0.265 r 0.991 0.9949 slope 110.00 15.0 7 y-intercept +0.91 +0.19 r 0.9918 0.9917 slope 133.75 10.25 8 y-intercept +1.44 0.508 r 0.9981 0.9967 that the order of effectiveness in the brown color produc- tion in most cases is parallel to the increase of the basicity of the amines. However, Willits et a1. (1958) suggested that the reactivity of lysine is not due to its basicity, but to a lysine-alkaline pH synergism, since arginine and histidine, other basic amino acids, had no positive effect on browning. Arginine contains a guanidinum group at the 5 position and it is reported that only the a-amino group of arginine is active (Shallenberger and Birch, 1974). Com- paring these two amino acids reveals the following: First, despite the Beachman and Dull (1951) report, there is a significant interaction between reducing sugars and arginine (Figures 27 and 28). The color developed by the reaction of arginine with glucose is second in 67 Lor l T 1 0.4 '- _. PH 7 00 A - p” 6 0.020 0.025 0.030 0.035 0.040 ARGININE CONCENTRATION , M Figure 27. Color formation by the reaction of arginine with fructose (0.4 M) at pH 6, 7, and 8, ionic strength 0.2, for 58 hours at 60° C. 68 4.0 I 1 I 3.0 '- PH 8 / g o v PH 7 < 2 0 - . " . PH 6 ' 1.0 ~ 4 I 0.0 1 I 1 0.020 0.025 0030 0.035 0.040 ARGININE CONCENTRATION . M Figure 28. Color formation by the reaction of arginine with glucose (0.4 M ) at pH 6, 7, and 8, ionic strength 0.2, for 58 hours at 60° C. 69 intensity only to that of lysine—glucose, among the amino acids tested. Second, at pH 8 the color of the arginine-fructose system is more than that of the lysine-fructose system. Third, in the arginine-glucose interaction more color is produced at pH 7 than at pH 6 or 8, under the con- ditions of this experiment (Figure 29). Fourth, in the reaction between fructose and arginine, there was no significant color develOpment at pH 6. The lepes, y-intercepts, and correlation coefficients for the reactions between arginine-glucose and arginine-fructose are shown in Table 6. Table 6.--S10pes, y-intercepts, and correlation coefficients (r) of the linear regressions between color (A450) and amino acid concentration for the reactions of arginine with glucose and fructose. pH Glucose Fructose 310pe 132.2200 - 6 y-intercept -2.3934 - r 0.9953 - slope 89.1000 38.2200 7 y-intercept +0.0032 +0.1808 r 0.9989 0.9952 slope 77.6200 14.8000 8 y-intercept -0.2460 -0.2512 r 0.9993 0.9979 The complete analysis of variance is shown in Tables 7 and 8. This analysis indicates that not only all independent variables (amino acids, concentrations of amino 70 .2 3d 328 8:5 .2 ed 383 "cozgcmocoo .o 08 S .c :2. mm 3% 5:32-328: 65583388: .33.: -3836 5553? 38:3 2533. 2: E E88238 .68 co Ia co 382m. .8 8 :2... In. Oflv 3303.: 3302a m4 9.:— 71 hoooo.o mvoo.o om Houum mmmm.m nm.mv¢ vamo.o na¢~.o m umfi mmmm.~ mm.¢am omao.o homa.o m om vunm.¢ mm.vmomm mmmm.o whom.ma m Ufl Hmvm.m m~.¢oa mHHo.o mmvo.o v m4 aeem.a me.amo~m -e~.~ mvmv.a m loo mm 2: Hmvm.m nm.mao~ mava.o vmom.o v cofiumuucmocoo Uwod onwad . . . . Adv pend onwE< ammo h mm vmmm vmma o vmma o H «o ousumz Away oflumm mmumsvm moumsvm Eocmmuh :oflumHHm> cannula um com: mo meow mo common mo condom .oCHGflmum cam ocflmwa nufl3 omouosum mo cofiuomou mnu an pmosooum Hoaoo map so mcofluomumucfl uflonu 0cm .mm .coflumuucoocoo cflom OGflEm .cflom ocean mo macaw: on» NO uoommm may mcflsosm cosmenm> mo mwm>HMG¢II.> magma 72 l I maoo.o mmmoa.o om uouum mmmm.m ~>.mhm hamm.o mvmv.m m um« mmmm.~ mm.hm ohma.o mooN.H m om vnnm.v oa.mvama mamv.vm hmmm.mo m 04 Hmem.m mm.¢ «moo.o mmmo.o e m< vhnm.v ow.hmmmm bmmm.mm mahm.maa m AUV mm Amy Hmvm.m mm.om>m neom.o mmam.hm v cofluwuucoocou Uflod ocead . . . . Adv nfio< czafid ammo h «w mmmna ohma Hm ohma Hm H mo musumz Away CaummIm mmnmsvm moumsvm Eopooum cofiuMflum> manmulm . now: no mEdm Mo ooumoo mo mounom .ocflcflmum cam ocwmma cufl3 mmoosHm mo aofluummu may CH mcoHuomuoucfl Macaw 6cm .mm .coflumuucoocoo pflom OCHEM .Uflom ocfiEm mo ousum: 039 Mo poowmo on» mcfl3onm mUCMflHm> mo mammamc ooH> 0cm :ofiuwmom onEmm onu ca poooam mm3 ouuo>oo xcoan onu .ommo menu CH .oHQEMm ona away once Es one no Connemno AocoHo Homsm wouoozv xcoan onu ponu mcooE Alva hm0.01 0vH.0I 000.0 N00.0+ 0N0.0+ 0v0.0 NN0.0I me.0I wm0.0 NH0.0+ m00.0l 0m0.0 HH0.03 NBH.0I Nm0.0 0v0.0+ Hm0.0I Nm0.0 000.0 NmH.0I 0N0.0 m00.0+ 000.0: 0N0.0 HN0.0+ mom.0t vN0.0 000.0+ «v00.0n «No.0 000.0+ 0Hv.0+ 000.0 000.0+ 0Hv.0+ 000.0 omaa .mm< oma< .mmm z omaa .sao omva .sao z + omoosaw + omouosum .mm¢ + omoosao + omouooum .sao .0 00 mo coauoHOm Hommsn CH oeflu cofluooou muson mm com 0000 um .Az v.00 omoosam 0cm .2 v.00 omouosum wo cofluouflHoEouoo onu :0 mwfloo oauuoamo 02o UHEousam mo chwuoHucoocoo msofluo> mo uoommmil.0H oHnoB 80 Figure 33. Molecular structure of (a) glutamic and (b) aspartic acids. iv. Another possible cause for the resistance of glutamic or aspartic acid to browning might be due to hydro- gen bonding, or dipole-dipole interaction, or both, between the charged amino group and the carboxylic groups. VI. Inhibition of Sugar-Amine Browning by Glutamic and Aspartic Acids A. Glutamic Acid Figure 34 shows the inhibitory effect of glutamic acid on lysine-glucose and lysine-fructose systems. The molar ratio, lysine:glutamic acid, was the same (1:1) in both the fructose and glucose-containing systems. At higher concentrations of both amino acids, the color development A450 FRUCTOSE 81 0 5 v 2 5 . GLUCOSE q 2 3 0.4 - . . O -' 2.1 LU 0.3 _ FRUCTOSE o 8 8 . 1.9 a 0 ° 3 .2 b - 1.7 < 0-0 ' 1.5 lays 0.01 0.03 0.04 "10.01 0.03 0.04 LYSINE AND GLUTAMIC ACID [1'1 M RATIO]. M Figure34. Effect of glutamic acid concentration (0.02 to 0.04 M l on the retardation of the browning reaction between lysine (0. 02 to 0.04 M l-fructose (0.4 M l and lysine (0.02 to 0.04 M l-glucose (0.4 M l at pH 8, ionic strength 0.2, after 58 hours reaction time at 60° C. 82 was lower than that observed at lower concentrations of amino acids. Apparently, this becomes more evident from the results shown in Figures 35 and 36, in which increasing concentrations of glutamic acid over constant concentrations of lysine result in decreasing browning. B. Aspartic Acid Results similar to those of glutamic acid were observed when aspartic acid was added to lysine-sugar sys- tems (Figures 37 and 38). The mechanism of this inhibition is unknown. One may speculate that aspartic and glutamic acids interfere with the reducing sugar-lysine reaction in a way similar to that of the inhibition of caramelization by these amino acids. Aspartic and glutamic acids may react with one or more of the many intermediate products formed during melanoidin formation (Figure 4) and thereby prevent the production of the final brown compounds. C. Inhibition of Non-Enzymatic Potato Browning by Aspartic and Glutamic Acid When the ampoules containing potato flour with or without glutamic acid were examined after overnight heating at 110°C, it was observed that the ampoules which did not contain glutamic acid had exploded, while the others were intact. The eXplosion was probably due to the formation of C02, which is regularly produced in sugar-amine reactions. The ampoules containing glutamic acid were less dark and must have had little or no gas, as no explosion occurred. 83 0.2 l l r 0.9 .. LYSINE , 0.04 M O o in V < 0'5 ' o ‘ . o LYSINE ,0.024 M . o o 0.3 ’ 0,0 l l 1 0.00 0.01 0.02 0.03 0.04 GLUTAMIC ACID, M Figure35. Effect of lysine concentration on color development in the glutamic acid- fructose system at pH 8, ionic strength 0.2, after 58 hours reaction time, and at 60° C. 84 7-0 6.0 LYSINE ,0.04 M 5-0 A450 2.0 cvsme . 0.02 M 1-0- I 1 fl 0.00 0.01 0.02 0.03 0.04 GLUTAMIC ACID. M Figure 36. Effect of lysine concentration on color development in the glutamic acid- glucose system at pH 8, ionic strength 0.2, after 58 hours reaction time, and at 60° C. A450 85 LYS'NE ’ 0.04 M 0-6 '7 - 0-4 "' _, LYSINE 0.024 M 0.2 " 0.0 1 .4— 1 0.00 0.01 0.02 0.03 0.04 ASPARTIC ACID . M Figure 37. Effect of lysine concentration on color development in the aspartic acid- fructose system at pH 8, ionic strength 0.2, after 58 hours reaction time, and at 60° C. 86 7.0 1 i l 6. 0 LYSINE , 0-04 M 4.0 A450 3.0 LYSINE 4 0-024 M 2.0 " 1.0 ” "‘ 0.0L 1 1 L 0.00 0.01 0.02 0.03 0.04 ASPARTIC ACID. M Figure 38. Effect of lysine concentration on color development in the aspartic acid- glucose system at pH 8, ionic strength 0.2, after 58 hours reaction time, and at 60° C. 87 Apparently, glutamic acid inhibited the browning reaction in the heated potato flour. The results of the color measurement of the sliced fried potatoes treated or nontreated with glutamic acid and aspartic acids were summarized in Table 11 and Figure 39. In Table 11 the L-value, which is a measure of the lightness in color, clearly indicates that the samples treated with glutamic or aspartic acid are less dark than the untreated samples. The reflectance spectra depicted in Figure 39 show that the treated samples reflect more light in almost all of the visible region, confirming the results obtained by the Hunter method of color measurement. It must be noted that usually these color differences were small immediately after frying, but they were amplified after 24 hours of standing at room temperature. The instrumental color evalu- ation was performed after the 24 hours of storage. Should glutamic and aspartic acids prove to inhibit the sugar-amino reaction in a number of commodities in which non-enzymatic browning is undesirable, a rather novel way of preventing this browning may be in sight. These amino acids not only are non-toxic, as 802 is suspected to be when added to foods, but they are also nutrients. Further- more, monosodium glutamate is known to be a flavor enhancer. 88 Table 11. Hunter color difference values for fried potatoes treated or untreated with aspartic and glutamic acid prior to frying. 3:332:43. grams. L L L 39.0 44.2 47.6 38.3 44.7 44.6 38.2 43.0 45.3 39.0 43.0 47.0 37.9 43.7 46.8 36.7 43.9 46.6 39.5 43.6 47.2 Average 38.4 i 0.57** 43.7 i 0.57 46.4 i 1.00 *The values for each replicate is the average of 3 numbers obtained by rotating each sample three times at an angle of 120°. **The differences of treated from untreated samples are significant at the 99% probability level. The differ- ence between treated samples is not significant. 89 am“: I‘m "DI!" .8355 A .2 0 005 .m 8228 8.... .328 2:88 2... 8.523.... 5.2.. 8.8: can 8.8.2.: .o 8.83 8:88.82 .3 e :2“. E... . Ih02m4w><3 4’81) 1414):-“ I” I Z- I: A' '(‘lfil x-‘ .-('a:) I. H D, H I. N 3 S (— —.. m o .A - .2 1 A l x .._._‘- -1- .._.a Filo-0...!!!» n.‘.nn.28¢<.(v flHh¢l°hl°DI- IIOJ I 30-51. BONVLOBMBH SUMMARY AND CONCLUSION The interaction between sugars, amino acids and pro- teins may result in undesirable discoloration in foods (brown- ing) and concommitant loss of nutritional value. In order to better understand the nature of this discoloration, the interaction between glucose and fructose on one hand, and alanine, arginine, aspartic acid, glutamic acid, glycine, lysine monohydrate, methionine, phenylalanine, serine, and a-amino-n-butyric acid on the other hand was studied through model systems. These systems were solutions of sugars and amino acid in phosphate buffers, exposed to higher than ambient temperature for defined periods of time. The main findings of this study are as follows: A. Sugars l. Fructose contributes less than glucose to brown- ing while reacting with amino acids, except in the model systems containing less glycine than 0.031 M, at pH 8.0. Fructose also shows a shorter induction period compared to glucose, so that the rate of discoloration at the initial stages of browning is higher than that of glucose; as the reaction advances, however, the browning of glucose systems is greater than that of the corresponding fructose systems. 90 91 2. Fructose is more readily caramelized compared to glucose. B. Amino Acids 1. The rate of the browning reactions in most cases was increased by increasing the pH value from 6 to 8. How- ever, the relationship between reaction rate and pH was not linear. In the arginine-glucose system, the reaction pro- ceeded the fastest at pH 7. 2. Among the amino acids tested, lysine and arginine were the most active in the production of brown color when reacting with glucose and fructose. However, the activity of arginine at pH 6 with fructose was nearly nil. 3. At pH 7 the reactivity of arginine with fructose is greater than that of lysine with fructose. 4. Glutamic acid and aspartic acid not only do not contribute to the browning in the particular reaction con— dition, but also retard caramelization and the Maillard reaction. Impregnating freeze-dried potato slices with aspartic or glutamic acid solutions resulted in less brown discoloration upon frying the slices. C. Activation Energy The activation energy for the reaction between glucose and fructose with lysine was 13,331 cal/mole for glucose-lysine and 10,210 cal/mole for fructose-lysine. The Qlo for the same reactions were 1.87 for glucose-lysine and 1.61 for fructose-lysine. BIBLIOGRAPHY BIBLIOGRAPHY Adrian, P. J. 1972. La Reaction de Maillard vue sous l'angle nutritionnel. I. Donaine de la reaction de Maillard. Ind. Alim. Agr. 89:1281-1289. Adrian, P. J. 1973. La Reaction de Maillard vue sous l'angle nutritionnel. III. Substances prejudiciables engendrees par la reaction de Maillard. Ind. Alim. Agr. 90:449-455. Adrian, P. J. 1973. La Reaction de Maillard vue sous l'angle nutritionnel. IV. Aromes engendre's par la reaction de Maillard. Ind. Alim. Agr. 90:559- 564. Adrian, P. J. 1974. World Review of Nutrition and Die- tetics. 19:74. Ambrose, A. M.; Robinns, D. J.; and Eds, F. 1961. Acute and subacute toxicity of aminohexose—reductones. Prec. Soc. Exp. Biol. Med. 106:656-659. Anderson, F. H.; Moran, D. H.; Huntley, T. E.; and Holahan, J. L. 1963. Responses of cereals to antioxidants. Food Technol. 17:1587-1592. Baliga, G. P.; Bayliss, M. E.: and Lyman, C. M. 1959. Determination of free lysine epsilon amino groups in cottonseed meals and preliminary studies on relation to protein quality. Arch. Biochem. 84:1-6. Beacham, H. H., and Dull, M. F. 1951. Some observations on the browning reaction. Food Research. 16:439-445. Bondarovich, H. A.; Friedel, P.; Drampl, V.; Renner, J. A.: Shepard, F. W.; and Gianturee, M. A. 1967. Volatile constituents of coffee pyrazines and compounds. J. Agric. Food Chem. 15:1093-1099. Branes, H. M., and Kaufman, E. W. 1947. Industrial aspects of browning reaction. Ind. Eng. Chem. 39:1167-1170. British patent, 1966. l, 169, 538. 92 93 Bronislaw, W. L., and McWeeny, D. J. 1974. Non-enzymic Browning Reaction of Ascorbic Acid and their Inhibi- tion. The identification of 3-Deoxy-4-suph0pentosulose in dehydrated, sulphited cabbage after storage. J. Sci. Food Agric. 25:589—592. Burton, H. S., and McWeeny, D. J. 1963. Non-enzymatic browning reactions: Consideration of sugar stability. Nature (London) 197:266-268. Cantor, S. M., and Peniston, Q. P. 1940. The Reaction of Aldoses at the Dropping Mercury Cathode: Estimation of the aldehyde structure in aqueous solutions. J. Am. Chem. Soc. 62:2113-2121. Cera, J. D.. and Tappel, A. L. 1976. Do carbonylamine browning reactions occur in vivo? J. Agric. Food Chem. 24:74-77. Clark, A. V., and Tannenbaum, S. R. 1973. Studies on limit- peptide pigments from glucose-casein browning system. Using radioactive glucose. J. Agric. Food Chem. 21: 40-43. Cook, B. 8.: Morgan, A. P.; Singer, 8.; and Parker, J. 1951. The effect of heat treatment on the nutritive value of milk proteins. J. Nutr. 44:62-81, 217-235. Cooney, P. M.: Hodge, J. E.; and Evans, C. D. 1958. Devel- opment of color in fats stabilized with amino-hexose— reductones. J. Am. Oil Chem. Soc. 35:167-171. Cotton, A. F., and Wilkinson, G. 1966. Advanced Inorganic Chemistry, 2nd ed. 545 pp. Day, E. A.; Bassette, R.; and Keeney, M. 1960. Identifi- cation of volatile carbonyl compounds from cheddar cheese. J. Dairy Sci. 43:463-474. Erickson, J. G. 1953. Reactions of aliphatic amines with sugars. J. Am. Chem. Soc. 75:2784. Evans, R. J., and Butts, H. A. 1948. Studies on heat inactivation of lysine in soybean oil meal. J. Biol. Chem. 175:15-20. Evans, R. J., and Butts, H. A. 1949b. Studies of heat inactivation of methionine in soybean oil meal. J. Biol. Chem. 178:543-548. 94 Evans, C. D.; Moser, H. A.; Cooney, P. M.; and Hodge, J. E. 1958. Amino-hexose-reductones as antioxidants. I. Vegetable Oils. J. Am. Oil Chem. Soc. 35:84-87. Friedman, L., and Kline, O. L. 1950. The amino acid-sugar reaction. J. Biol. Chem. 184:599. Friedman, L., and Kline, O. L. 1950. The relation of the amino acid-sugar reaction to the nutritive value of protein hydrolysates. J. Nutr. 40:295-307. Greaves, E. O., and Morgan, A. F. 1933. Nutritive value of raw and heated casein with and without added amino acids. Proc. Soc. Exp. Biol. Med. 31:506-507. Griffith, T., and Johnson, J. A. 1957. Relation of the browning reaction to storage stability of sugar cookies. Cereal Chem. 34:159-169. Hagar, S. S. N.; Horn, M. J.; Lipton, S. H.; and Wemak, M. 1970. Fructose-glycine as a source of nonspecific nitrogen for rats. J. Agric. Food Chem. 18:273-275. Hannan, R. S., and Lea, C. H. 1952. Studies of the reaction between proteins and reducing sugars in the "dry" state. VI. The reactivity of the terminal amino groups of lysine in model systems. Biochem. Biophys. Acta. 9:293-305. Hansen, L. P., and Millington, R. J. 1979. Blockage of protein enzymatic digestion (carboxy peptidose-B) by heat-induced sugar-lysine reactions. J. Food Sci. 44:1173-1179. Haugaard, L.; Tumerman; and Silvestri, G. 1951. A study of the reaction of aldoses and amino acids. J. Am. Chem. Soc. 73:4594-600. Hedin, P. A. 1962. Gastrointestinal gas production in rats as influenced by some animal and vegetable diets, sulfiting, and antibiotic supplementation. J. Nutr. 77:471-476. Hendel, C. E.: Silveira, V. G.: and Harrington, W. O. 1955. Rats of nonenzymatic browning of white potato during dehydration. Food Technol. 9:433-438. Herz, W. J., and Schallenberger, R. S. 1960. Some aromas produced by simple amino acids, sugar reactions. Food Res. 25:491-494. 95 Hodge, J. E. 1959. Chemistry of browning reaction in model systems. J. Agric. Food Chem. 1:928-943. Horn, J. M.; Linechtenstein, H.; and Womack, M. 1968. A methionine-fructose compound and its availability to microorganisms and rats. J. Agric. Food Chem. 16:741-745. Hunter, I. R.; Walden, M. K.; McFadden, W. H.; and Pence, J. W. 1966. Production of breadlike aromas from proline and glycerol. Cereal Sci. Today 11:493-501. Ingles, D. L. 1959. Chemistry of non-enzymatic browning. V. The preparation of aldose-potassium bisulphite addition compounds and some amine derivates. Aust. J. Chem. 12:97—101. Ingles, D. L. 1959. Chemistry of non-enzymic browning. VI. The reaction of aldoses with amine bisulphite. Aust. J. Chem. 12:275-279. Ingles, D. L. 1966. Prevention of non-enzymic browning. Food Preserv. Quart. 26:39-44. Ingles, D. L. 1963. Thoil Interactions in Sugar-Amine Systems. Chem. Ind. (London):1901-1902. Itah, H.; Kawashima, K.; and Chibata, I. 1975. Anti- oxidant activity of browning products of triose sugar and amino acid. Agric. Biol. Chem. 39:283-284. Japanese Patent. 1976. Mitsui Totasu Chemicals, Inc. 5, 118, 508 (Ja). Johnson, J. A.; Rooney, L.; and Salem, A. 1966. Chemistry of bread flavor. Advances in Chemistry Series. American Chemical Society publication, Washington, D.C. 153-173. Joslyn, M. A., and Braverman, J. B. S. 1955. Adv. Food Res. 5:97-160. Kearsley, M. W. 1975. Action of aqueous sodium hydroxide on glucose syrups. Food Chem. 2:27-41. Kearsley, M. W. 1978. The control of hygroscopicity, browning and fermentation in glucose syrups. Food Technol. 13:339-348. 96 Kirigaya, N.; Kato, H.; and Fuzimakis, M. 1968. Studies on antioxidant activity of nonenzymatic browning reaction products. Part I. Relation of Color Inten- sity and Reductones with Antioxidant Activity of Browning Reaction Products. Agr. Biol. Chem. 32: 287-290. Labuza, T. P.; Tannenbaum, S. R.; and Karel, M. 1970. Water Content and Stability of Low-Moisture and Intermediate-Moisture Foods. Food Technol. 24:543- 550. Lento, H. G.; Underwood, J. C.; and Willits, C. O. 1957. Browning of Sugar Solutions. II. Effect of Position of Amino Group in the Acid Molecule in Dilute Glucose Solutions. Food Res. 23:68-71. Lewis, V. M., and Lea, C. J. 1950. A note on the relative rates of reaction of several reducing sugars and sugar derivates with casein. Biochem. et BiOphys. Acta. 4:532-534. Linko, Y. Y., and Johnson, J. A. 1963. Changes in amino acids and formation of carbonyl compounds during baking. J. Agric. Food Chem. 11:150-152. Maga, J. A., and Mente, W. C. 1977. The Effect of Butylhydroxyanisole (BHA) on Amino Acid and Carbohydrate Concentration During Heating. Lebensm-Wiss. U.-Technol. 10:130-140. Markakis, P. 1979. Private communication. Mauren, J.; Mettu, F.; Bujar, E.; and Egli, R. H. 1955. The availability of lysine, methionine and tryptophan in condensed milk and milk powder. In vitro studies. Arch. Biochem. 59:435-451. McDonald, F. J. 1966. Available lysine content of dried milk. Nature (London) 209:1134. McWeeny, D. J., and Burton, H. S. 1963. Some possible glucose/glycine browning intermediates and their reactions with sulphites. J. Sci. Food Agric. 14: 291-302. McWeeny, D. J., and Burton, H. S. 1962. Sulphites and Carbonyl-amino ChromOphore Development. Nature (London)196:4l-42. 97 Merritt, C. J.; Bazinet, M. 8.; Sullivan, J. H.; and Robertson, D. H. 1963. Mass spectrometric determi- nation of the volatile components from ground coffee. J. Agric. Food Chem. 11:152-155. Mills, F. D.; Baker, B. G.; and Hodge, J. E. 1969. Amadori Compounds as Nonvolatile Flavor Precursers in Pro- cessed Foods. J. Agric. Food Chem. 17:723-727. Obanu, Z. A.; Ledward, D. A.; and Lawrie, R. A. 1977. Reactivity of glycerol in intermediate moisture meats. Meat Science 1:177-183. Palamidis, N. 1979. Ph.D. Thesis, Michigan State Univer- sity. Patton, A. R.; Hill, E. G.; and Forman, E. M. 1948. Amino Acid Impairment in Casein Heated with Glucose. Science 107:623-24. Patton, A. R.; Hill, E. G.: and Forman, E. M. 1948. The effect of browning on the essential amino acid content of soy globulin. Science 108:959-660. Payne-Botha, S., and Bigwood, E. J. 1959. Amino acid con- tent of raw and heat-sterilized cow's milk. Brit. J. Nutr. 13:385-389. Pokorny, J.; Con, N. H.; and Janicek, 1973. Nonenzymic Browning. V. Effect of Hydrogen Peroxide on the Destruction of Brown Pigments in Model System. Z. Lebensm. Unters.-Forsch 151:305-307. Pomeranz, Y.; Johnson, J. A.; and Shellenberger, J. A. 1962. Effect of various sugars on browning. J. Food Sci. 27:350-354. Ramshaw, E. H., and Dunstone, E. A. 1969. The flavor of milk protein. J. Dairy Res. 36:203-213. Reynolds, T. M. 1959. Chemistry of nonenzymic browning. III. Effect of bisulphite, phosphate, and malate on the reaction of glycine and glucose. Aust. J. Chem. 12:265-274. Rosenberg, H. R., and Rohdenburg, E. L. 1951. The loss of lysine during baking. J. Nutr. 43:593-598. Rubenthaler, G.; Pomeranz, Y.; and Finney, K. F. 1963. Effects of sugars and certain free amino acids on bread characteristics. Cereal Chem. 40:658-665. 98 Safar, O. 1976. L'importanza dei Ceppi di lievito nel prevenire I'imbrunimento dei vini bianchi. Ann. Microbiol. 26:101-107. Schonberg, A.; Moubacher, R.; and Mostafa, A. 1948. Degradation of a-amino acids to aldehydes and ketones by interaction with carbonyl compounds. J. Chem. Soc. (London):176-182. Schroeder, L. J.; Iacobellis, M.; and Smith, A. H. 1961. Influence of heat on the digestibility of meat pro- teins. J. Nutr. 73:143-150. Self, R.: Rolley, H. L. J.; and Joyce, A. E. 1963. J. Sci. Food Agric. 14:8-14. Shallenberger, R. S., and Birch, G. G. 1974. Nonenzymic Browning Reactions. Sugar Chemistry. The AVI Publishing Company. Simon, M.; Wager, J. R.; Silveira, V. G.; and Hendel, C. E. 1955. Calcium Chloride as a Non-enzymic Browning Retardant for Dehydrated White Potatoes. Food Technol. 9:271-275. Stegnik, L. D.; Freeman, J. B.; Meyer, P. D.: Filer, L. J.; Fry, L. K.; and Denbesten, L. 1975. Excessive trace metal ion excretion due to sugar amino acid complexes during total parenteral nutrition. Food Proc. 34: 931. Stegnik, L. D., and Pitkin, R. M. 1977. Placental transfer of glucose-amino acid complexes present in parental solutions. Am. J. Clin. Nutrition 30:1087-1093. Stegnik, L. D.; Shapherd, J. A.; Fry, L. K.; and Filer, Jr., L. J. 1974. Sugar-amino acid complexes in paren- teral alimentation. Pediat. Res. 8:396. Underwood, J. C.; Lento, Jr., H. G.; and Willits, C. O. 1959. Browning of sugar solutions. 3. Effect of pH on the color produced in dilute glucose solutions containing amino acids with the amino group in dif- ferent positions in the molecule. Food Research 24:181-184. Willits, C. 0.: Underwood, J. C.; Lento, Jr., H. G.; and Ricciuti. 1958. Browning of sugar solutions. I. Effect of pH and type of amino acid in dilute sugar solutions. Food Research 23:61-67. 99 Wiseblatt, L., and Koh, F. E. 1960. Some volatile aromatic compounds in fresh bread. Cereal Chem. 37:55-56. Wolfrom, M. L.; Kashimura, N.; and Horton, D. 1974. Factors affecting the Maillard browning reaction between sugars and amino acids. Studies on the nonenzymatic browning of dehydrated orange juice. J. Agric. Food Chem. 22:796-800. Wolfrom, J. L.; Schuetz, R. D.; and Calvalieri, L. F. 1948. Discoloration of sugar solutions and 5-(hydroxymethyl) furfural. J. Am. Chem. Soc. 70:514-517. Zabrodskii, A. G., and Viktovskaya, V. A. 1960. The effect of melanoid substances on malt amylase. Chem. Abst. 54:1635.