A STUDY OF LIPID OXEDATION {N MODEL SYSTEMS 0F COPPER'PROTEINS. NHLK LIPIDS AND MSLK DiALYZATE Thesis for the Degree of Ph. D. MiCHlGAN STATE UNEVERSITY * JASWANT SINGH AULAKH ; 1967 TH E85. m: '4. 5.. 0-.) .‘ This is to certifg that the thesis entitled A Study of Lipid Oxidation in Model Systems of Copper-Proteins, Milk Lipids and Milk Dialyzate presented by Jaswant Singh Aulakh has been accepted towards fulfillment of the requirements for Ph.D. degree in Food Science MMSW Major professor Date January 5, 1968 0-169 ” “ _.' [P a s It" ‘Nr ‘ ABSTRACT A STUDY OF LIPID OXIDATION IN MODEL SYSTEMS OF COPPER-PROTEINS, MILK LIPIDS AND MILK DIALYZATE by Jaswant Singh Aulakh The binding of copper with micellar casein, fat- globule membrane protein, a-casein, B—casein, B-lacto- globulin was studied by the technique of equilibrium dialysis at pH 6.5. Solutions of these proteins, in concentrations of 2.000, 3.000, 2.A36, 1.205, and 1.420 mg/l were placed in washed dialysis bags and dialyzed against each of four copper solutions for 72 hours, an interval considered sufficient for the attainment of equilibrium. The concentrations of the four c0pper solutions outside the dialysis bags were 16x, 32x, 64x, or 80 x 10-6M, respectively. The data were represented in graphs by plotting the moles of bound copper per mole of protein against the logarithm of the concentration of un- bound copper. The moles of copper bound per mole of pro- tein were approximately 67, 2A, 10, 2, 2.5, respectively, for the above proteins at a free COpper concentration of l x 10-5M. The results clearly show the great difference in ability of certain milk proteins to bind c0pper ions. Micellar casein bound the greatest amount of COpper. Jaswant Singh Aulakh The fat-globule membrane protein demonstrated the second greatest affinity for COpper ions. Much lower amounts of copper were bound by a—casein, B-lactoglobulin, and B-casein. Model systems of copper-micellar caseinate, copper—fat-globule membrane protein, copper-sodium-a- caseinate, copperesodium B-caseinate and copper—B-lacto- globulin were incorporated into model systems of washed fat-globules and milk dialyzate. The model system with— out copper was specified as the control. The systems with c0pper and without copper were heated at 1M5°F for 30 minutes and 190°F for 10 minutes. Model systems of copper proteinate of B-lactoglobulin, fat—globule membrane protein and a model system of washed fat-globules and milk dialyzate were also heated individually at lu5°F for 30 minutes and l90°F for 10 minutes and then combined to- gether. Samples were stored at 35°F for 10 days. At 2-, 4-, 6-, 8—, and 10-day intervals, oxidation was followed by determination of thiobarbituric acid and peroxide values. The results indicated that heated sys- tems showed significantly less tendency to oxidize than unheated systems. Systems heated to 190°F for 10 minutes exhibited the least amount of oxidation. The individual heating of copper proteinates and model system of washed fat-globules and milk dialyzate demonstrated lower level of oxidation than the systems in which copper proteinates Jaswant Singh Aulakh were heated and washed fateglobules and milk dialyzate were unheated. The decrease in oxidation in such systems suggests that physical or physico-chemical changes at the fat-globule surface are associated with the beneficial effects of high heat treatment of milk. A STUDY OF LIPID OXIDATION IN MODEL SYSTEMS OF COPPER-PROTEINS, MILK LIPIDS AND MILK DIALYZATE By Jaswant Singh Aulakh A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Food Science 1967 DEDICATION The work in this thesis is dedicated to my be- loved wife, Gurcharan Kaur Aulakh, whose patience and encouragement during this strenuous period of graduate study will never be forgotten. ii ACKNOWLEDGMENTS The author wishes to express his sincere appreciation to Dr. C. M. Stine, Associate Professor of Food Science, for his encouragement and counsel during this study. His philosophy of life as well as his scientific competence will be a perpetual source of inspiration. The author also extends his thanks to Dr. H. A. Lillevik, Dr. J. R. Brunner, Dr. E. J. Benne and Dr. T. I. Hedrick for their suggestions in the preparation of this manuscript. The author further thanks and is grateful to Experiment Station of Michigan State University for financial assistance in the form of an assistantship. iii TABLE OF CONTENTS DEDICATION ACKNOWLEDGMENTS . . . . . . LIST OF TABLES. , . . . . . . . . . LIST OF FIGURES INTRODUCTION LITERATURE,REVIEW. . . . . Distribution of copper in fluid milk. Binding of copper with milk proteins and the methods used in the binding studies. Methods of copper analysis in milk and milk products . . . . . . . Lipid oxidation in fluid milk induced by copper. . . . . . . . . EXPERIMENTAL METHODS AND MATERIALS Preparation of various milk proteins. Procedure for binding of copper with proteins . . . . . . . Preparation of model systems Milk dialyzate Washed fat—globules. Preparation of samples and heat treatment iv Page ii iii vii ix l2 13 22 22 22 23 23 2A 2A Page Oxidation studies . . . . . . . . . 25 Experimentation. . . . . . . . . . 25 Experiment 1. --binding of COpper with micellar casein . . . . . . 25 Experiment 2.--binding of copper with fat-globule membrane protein. . . . 26 Experiment 3. --binding of copper with sodium a- —caseinate . . . . . . 26 Experiment 4. --binding of copper with sodium 8- caseinate . . . . 26 Experiment 5. --binding of copper with B- lactoglobulin . . . . . . 27 Experiment 6.--lipid oxidation in model systems of copper—micellar caseinate, washed fat-globules and milk dialyzate . . . . . . . . 27 Experiment 7.—-lipid oxidation in model systems of copper—fat globule membrane proteinate, washed fat- _ globules and milk dialyzate . . . . 28 Experiment 8.--lipid oxidation in model systems of copper sodium a- caseinate, washed fat-globules and milk dialyzate . . . . . . . . 28 Experiment 9.--lipid oxidation in model systems of copper sodium 8— caseinate, washed fat-globules and milk dialyzate . . . . . . . . 28 Experiment lO.--lipid oxidation in model systems of copper-B-lacto- globulin proteinate, washed fat— globules and milk dialyzate . . . . 28 Experiment ll.-—lipid oxidation in heated and unheated model systems of copper-B-lactoglobulin proteinate, washed fat-globules and milk dialyzate Experiment l2.—-lipid oxidation in heated and unheated model systems of copper-fat-globule—membrane proteinate, washed fat-globules and milk dialyzate RESULTS AND DISCUSSION. SUMMARY. BIBLIOGRAPHY APPENDIX vi Page 29 29 3O 64 66 78 Table 10. 11. LIST OF TABLES Binding of Copper by Micellar Casein Binding of Copper by Fat—Globule Membrane Protein. . . . . . . . . . . Binding of Copper by Sodium a-Caseinate Binding of Copper by Sodium B-Caseinate Binding of Copper by B-Lactoglobulin The TBA Values Obtained From Lipid in Model Systems of COpper-Micellar Caseinate, Washed Fat-Globules and Milk Dialyzate . . . . . The Peroxide Values Obtained from Lipid in Model Systems of Copper—Micellar Caseinate, Washed Fat—Globules and Milk Dialyzate . . . . . . The TBA Values Obtained from Lipid in Model Systems of COpper-Fat-Globule Membrane Proteinate, Washed Fat— Globules and Milk Dialyzate The Peroxide Values Obtained from Lipid in Model Systems of Copper-Fat- Globule Membrane Proteinate, Washed Fat—Globules and Milk Dialyzate. The TBA Values Obtained from Lipid in Model Systems of Copper Sodium a- Caseinate, Washed Fat-Globules and Milk Dialyzate . . . . . . . The TBA Values Obtained from Lipid in Model Systems of Copper Sodium B— Caseinate, Washed Fat-Globules and Milk Dialyzate . . . . vii Page 80 80 81 81 81 82 82 83 83 84 8A Table 12. 13. 1A. 15. 16. The TBA Values Obtained from Lipid in Model Systems of COpper B-Lacto- Globulin, Washed Fat—Globules and Milk Dialyzate . . . . The TBA Values Obtained from Lipid in Heated and Unheated Model Systems of Copper B—Lactoglobulin, Washed Fat— Globules and Milk Dialyzate . The Peroxide Values Obtained from Lipid in Heated and Unheated Model Systems of Copper B-Lactoglobulin, Washed Fat-Globules and Milk Dialyzate. The TBA Values Obtained from Lipid in- Heated and Unheated Model Systems of Copper-Fat Globule Membrane Protein— ate, Washed Fat-Globules and Milk Dialyzate . . . . . . . . The Peroxide Values Obtained from Lipid in Heated and Unheated Model Systems of Copper-Fat-Globule Membrane Pro- teinate, Washed Fat-Globules and Milk Dialyzate . . . . viii Page 85 85 86 86 87 Figure l. 2. 6a. 6b. 7a. 7b. 10. LIST OF FIGURES Binding of Copper with Micellar Casein Binding of Copper with Fat—Globule Mem- brane Protein. . . . . . . . Binding of Copper with Sodium a—Caseinate Binding of Copper with Sodium B-Caseinate Binding of Copper with B—Lactoglobulin Lipid Oxidation in Model Systems of Copper-Micellar Caseinate, Washed Fat- Globules and Milk Dialyzate (TBA) Lipid Oxidation in Model Systems of Copper-Micellar Caseinate, Washed Fat- Globules and Milk Dialyzate (P.V.). Lipid Oxidation in Model Systems of Copper-Fat-Globule Membrane Protein— ate, Washed Fat-Globules and Milk Dialyzate (TBA) . . . . . . Lipid Oxidation in Model Systems of COpper-Fat-Globule Membrane Protein- ate, Washed Fat-Globules and Milk Dialyzate (P.V.). . . . Lipid Oxidation in Model Systems of Copper Sodium a-Caseinate, Washed Fat— Globules and Milk Dialyzate . . Lipid Oxidation in Model Systems of Copper Sodium B—Caseinate, Washed Fat— Globules and Milk Dialyzate . . Lipid Oxidation in Model Systems of Copper-B-Lactoglobulin Proteinate, Washed Fat-Globules and Milk Dialyzate . . . . ix Page 32 33 3A 35 36 A0 Al AA “5 47 49 52 Figure 11a. 11b. 12a. 12b. Lipid Oxidation in Heated and Unheated Model Systems of Copper-B-Lacto- Globulin Proteinate, Washed Fat- Globules and Milk Dialyzate (TBA) Lipid Oxidation in Heated and Unheated Model Systems of Copper-B-Lacto— Globulin Proteinate, Washed Fat- Globules and Milk Dialyzate (P.V.) Lipid Oxidation in Heated and Unheated Model Systems of Copper—Fat-Globule Membrane Proteinate, Washed Fat- Globules and Milk Dialyzate (TBA) Lipid Oxidation in Heated and Unheated Model Systems of Copper-Fat-Globule Membrane Proteinate, Washed Fat- Globules and Milk Dialyzate (P.V.) Page 57 58 6O 61 INTRODUCTION Milk lipids are generally recognized as the major contributors in the development of oxidized flavors in dairy products. The reaction of these lipids with oxygen, resulting in flavor deterioration, creates serious pro- blems in storage stability. The impairment of quality, due to autoxidation, is widespread, and manifests itself in many ways, depending upon the product involved. In certain products, particularly dry whole milk, autoxi- dation has been a major impediment to the development of an acceptable product. Considerable expense and effort have been expended in attempts to prevent autoxidation by use of antioxidants and inert gas packing, but such methods are only partially effective. Copper is capable, to a great extent, of catalyzing these oxidation reactions. Much information on the distri- bution of c0pper in milk is now available in the literature. Copper complexes with the proteins in the fluid milk sys- tems and the copper-protein complex in turn is capable of catalyzing the development of oxidized flavor. In this investigation, binding of copper with various milk proteins was studied. The objective of this study was to determine the effect of low and high heat treatments on the various copper—proteinates suspended in a model system of washed fat-globules and milk dialyzate. LITERATURE REVIEW Oxidized flavor in some dairy products, namely fluid milk, ice cream, and dry whole milk is a distress- ing problem to the dairy industry. No endeavor has been made in this thesis to extensively review the vast liter- ature on oxidized flavor unless it pertained specifically to distribution of copper in fluid milk, the binding of copper with milk protein, and the role of copper in the development of oxidized flavor. Comprehensive general reviews of the subject of oxidized flavor include reviews by Brown and Thurston (9), Day (20), Greenbank (36), Pont (85), Riel and Sommer (88), Stull (106), and Wilkin- son (120). Distribution of Copper in Fluid Milk Elvehjem e£_al., (27) reported earlier that cow's milk contained about 0.15 mg of copper per liter. The copper content of cow's milk could not be increased by feeding the cows sufficient COpper sulfate to increase the copper intake five fold. The average COpper content of fresh cow's milk was found to be 0.0553 ppm : 0.0046 (79). No considerable difference was noted between the copper contents of samples from grazing cows and those from stall fed cows. The average copper content of raw bottled whole milk was 0.0820 ppm : 0.0084, and of heat treated bottled whole milk, 0.0995 ppm : 0.0063. Some— what higher copper values were found in skimmilk, butter- milk and cream, butter and cheese. Koppeljan and Mulder (63) determined the copper content of the milk of a number of cows at different times in a lactation period. Generally a considerably higher amount of copper was found at the beginning of a lactation period than in the second half. The colostrum milk from most cows contained less copper than milk obtained during early lactation. Similar results were reported by King and Dunkley (52). They contended that a relatively high concentration of natural copper was present in early lactation milk, and a level of about 0.02—.04 ppm for most cows during the remainder of the lactation period. No difference was observed in the effect of pasture and dry feed on the natural copper concentration of milk from cows on these rations. According to Mulder et_al. (76), the natural copper content of normal cow's milk was 20-40 ug/kg, with somewhat lower or higher values occurring at the beginning of lactation. The influence of a number of factors, e.g. variation during one milking, influence of the season and individual effects on copper content of the milk were also discussed (76). The colostrum con— tained a rather high amount of copper (100 ug/kg). In the first day after calving, the copper content of the milk increased, a maximum being reached after 0-7 days (about 200 ug/kg), and thereafter the copper content decreased to normal values (20-40 ug/kg). Mulder et_al. (76) did not provide a satisfactory explanation of such changes and furnished little information on the influence of the copper content of the cow's feed and of intravenous injections with copper solutions, although many experi— ments on this topic have been reported in the literature (52). The authors concluded that the copper content of the cow's feed had no influence on the copper content of the milk. Rice and Miscall (87) showed that on passing milk containing copper through a separator, the metal dis- tributed itself between the cream and skimmilk fractions approximately in proportion to the water content of these fractions. The copper was considered, therefore, to be dissolved in the water. Their results indicated that the c0pper did not go with the fat and was therefore not dis- solved in the fat. Davies (19) reported that with gravity creaming the copper was distributed in direct proportion to the water in the two fractions or in direct proportion to protein nitrogen. The copper was not distributed ac- cording to the weight of the fat. In the centrifuged portion where the residual fat in the separated milk layer was too small to be determined accurately, the copper/protein nitrogen ratio was approximately three times greater in the cream layer. The high copper—protein nitrogen ratio in the cream layer was apparently due to the adsorption of copper at the surface of the fat— globules. Dills and Nelson (22) and Allen (2) revealed that iron and copper did not exist in the ionic state in milk and related products, but were combined in some way with the lipo—protein complex which surrounds or is at least adsorbed on, the fat-globules. This suggests that the catalytic action of these metals must be considered as the effect of the combined metal proteinate, and not as the effect of a metallic ion. Davies (19) observed that the concentration of the ionic form of heavy metal in milk was of a low order but increase in acidity caused more ionic metal to form. Willard and Gilbert (116) recovered more copper from the cream than from the skimmilk fraction and oxidized flavor development was correspondingly more pronounced in the cream fraction. However, Miller and Tracy (74) showed that copper in milk products was associ— ated with the non-fat portion rather than the fat phase. Butter made by the continuous process contained less copper than the original cream. The centrifugal treat- ment of cream in the process of making butter by the continuous method resulted in a concentration of copper in the non-fat portion. Similarly, butter oils made by the centrifugal process in the continuous method of making butter contained only minute amounts of copper. Mulder and Koppeljan (77) indicated the uneven distribution of the naturally occurring copper between the different phases of milk. A high proportion was re- tained on the surface of the fat—globules, the pro- portion varying with the cow and the stage of lactation. Naturally occurring COpper was found to be associated with fat-globule membrane, whereas the remaining copper in the separated milk was closely combined with the pro— teins (72, 73). King et_al, (54) reported that natural copper was concentrated at the surface of fat-globules but most of the added copper was uniformly combined with skimmilk proteins. Only 2-3% of the total added copper was associated with the fat-globules. Natural copper could not be dialyzed, although added copper in the skim— milk could be slightly dialyzed. However, added copper associated with fat-globules was non-dialyzable. King and William (53) in their later studies used Cu6u in determining the distribution of natural copper in milk at intervals during early lactation. The amount of copper bound with fat-globules was about 15% after 2 and 4 weeks of lactation and 35% after 10 weeks. The change in COpper distribution brought about a change in suscepti- bility to oxidized flavor. Washing of cream and chelation by ethylenediamine tetraacetic acid (EDTA) removed up to 90% of lipid-bound copper. When these treated fat- globules were re-suspended and recovered from the original skimmilk, their copper content increased. Samuelsson (93) very recently investigated the distribution of added and natural copper by adding acti— vated copper chloride to milk. The milk was then fraction- ated and the copper content in these fractions determined by means of activation analysis. Results showed that about 90% of the added COpper was combined with skimmilk proteins and about 10% with fat-globules. About 5% of the COpper could be washed from the fat-globules and about 5% was bound with the actual membrane proteins. Within the skimmilk proteins the whey proteins contained the greatest amount of copper per gram protein. Similarly, Hartman (42) reported that whey proteins had the greatest affinity for copper; 15 to 20% of the natural copper was associated with the fat-globules, 10-15% could be washed off and the actual membrane protein contained about 4% of the total copper but at the same time contained the highest copper content per gram protein (93). Binding of Copper with Milk Proteins and the Methods Used in the Binding Studies Vandevelde (115) studied the precipitation of milk proteins by copper salts. The results indicated that only the cation was bound by the milk proteins, and this in varying quantities, depending upon the concentration of copper solution employed. Sollner (102) used collodion membranes of high ionic selectivity for the potentio— metric determination of K+, Na+, Li+, NHu+, Mg+ and probably other cations. The potential difference which arose between a known solution on the one side of the membrane and the solution of unknown concentration on the other side of the membrane was utilized to determine a particular ion. This method could be used in studies of the activity of counterwions in colloidal systems, for the determinations of the anion and cation binding capacity of proteins, and similar problems. Leviton eg_al. (71) reported the binding of riboflavin and ribo— flavin phosphate by the proteins of milk, particularly a- and B-casein. They determined binding constants and calculated thermodynamic quantities. The binding was of low order of magnitude with an extremely high temperature coefficient. The binding by the calcium caseinate— phosphate complex in milk was equal to the binding by the component protein molecules. They found certain similarities between the binding by the caseins and the binding by flavo—protein enzymes. Other workers (105) studied the stoichiometry of the biuret reaction of alkaline copper tartrate with proteins, including serum proteins, B-lactoglobulin, edestin, pork gelatin, gliadin and zein. In all of these proteins except serum globulin, six pep- tide nitrogens were bound by chelation to each copper. In serum globulin, five peptide nitrogens were combined with each copper. The authors also discussed the contri— butions of cysteine, tyrosine, glutamic acid, aspartic 10 acid, histidine, arginine, and lysine to the color of copper-protein complexes. They proposed methods for determining peptide nitrogen by means of copper binding and for estimating proteins by a titrimetric method. Klotz and Curme (60) and Klotz et_al. (61) studied the extent of binding of cupric ions and methyl orange by bovine serum albumin at a pH of 5.67 and 4.8, respectively. The equilibrium dialysis technique was used in the binding studies. The decrease in binding with decrease in pH and the character of the absorption spectrum of the copper- albumin complex indicated that carboxyl groups on the protein were involved in the bond with the cation. Later on, Klotz (59) reported the binding of cupric ions and calcium ions by bovine serum albumin. Pronounced differ- ences existed in the behavior of Cu++ and Ca++ with albumin near its isoelectric point. Copper was bound strongly and calcium hardly at all. Copper formed much stronger com— plexes with either the carboxyl group of acetate or the nitrogen of ammonia than did calcium or magnesium. The metal—binding activity of conalbumin has also been reported by Fraenkel-Conrat and Feeney (15). Equilibrium dialysis experiments showed that the copper was less firmly bound than iron. Carr (10, ll, 12, 13) and Carr and Woods (14) used membrane electrodes while studying the binding of small ions in protein solutions. To make a measurement, 11. the protein solution containing calcium chloride was placed inside a bageshaped sulfonated polystyrene collodion membrane, and a known amount of distilled water was placed outside. A concentrated standard solu- tion of CaCl was added to the outside solution. After 2 each addition the membrane potential was measured, and this procedure was continued until the potential reversed in sign. The potential measurements and the concene tration of the outside solution were plotted on the semi— logarithmic graph paper. The concentration of the cal- cium in the outside solution could be calculated from such a graph. The amount of calcium bound to the protein was then calculated directly from the difference in the concentration of calcium in the outside solution and the concentration of calcium added to the protein solution. Baker and Saroff (4) used_an apparatus consisting of three chambers in a linear array, separated in pairs by two permselective membranes, one negatively charged and the other positively charged. The protein solution (B-lactoglobulin) was placed in the center chamber and accurately known solutions in the end chambers. A potential was created across each membrane, linearly re- lated to the logarithm of the activity of the free Na+ or 01— in the protein solution. Equilibration of the solutions required from 5 to 10 minutes. The potentials were recorded and a linear interpolation to zero potential 12 on a graph of millivolts versus log Na+ or Cl— gave the log free Na+ or C1—. The difference between Na+ added and free Na+ measured was the Na+ bound to the protein. Methods of Copper Analysis in Milk and Milk Products Most methods for the estimation of COpper are based on dry—ashing procedures, followed by colorimetric measurement of copper as its carbamate (23, 47, 52, 74). Many modifications of carbamate (74, l, 5, 16, 23, 47, 64, 46, 82, 107, 117), and dithizone (75) methods have been reported in the literature. Duin and Brons (25) described a simple and reliable method for the estimation of the COpper content of milk and milk powder. This method was a modification of the rapid method (24) of determining copper by using carbon tetrachloride and tri- chloroacetic acid. The new method was quite accurate and could be used as a routine method for the analysis of milk and milk powder. Several workers (37, 80) recently demonstrated the improved acid precipitating methods. King and Black (55, 56) made a comparison of the deter— minations of copper in butter and butterfat by a wet ashing colorimetric method, by atomic-absorption spectro- scopy and by neutron activation analysis. The wet ashing colorimetric method gave the most consistent results and these were in agreement with those from atomic absorption. Gehrke et a1. (33) developed a rapid, reliable and l3 accurate spectrographic method for the quantitative deter— mination of tin, copper, iron, and lead in milk and evapo- rated milk. The analyses of all four metals could be made simultaneously on a single sample. The precision of the method for the respective elements was from i 5 to i 8%. Willis (119) determined copper in butter and butter oil by atomic absorption spectroscopy. 'The method was accurate in determining low levels of COpper and according to the author this method could be suitable for deter- mining small amounts of copper in milk. Samuelsson (94) determined trace amounts of copper in milk by neutron activation analysis. Smith (97) very recently developed two rapid methods of precipitation for determining the copper content of the milk and compared these.methods with a dry ashing procedure. The methods involved the use of zinc dibenzyl—dithiocarbamate (Arazate) which requires no pH adjustment. In comparison to dry-ashing procedures, considerably less time was involved for the analyses and both methods were precise. Simultaneous micro—determi- nation of copper and iron using mixed phenanthrolines has also been reported in the literature (121, 122). Lipid Oxidation in Fluid Milk Induced by COpper Early investigations (18, 21) reported on the development of an oily metallic off-flavor in raw and 14 heated milk. The off-flavor was traced to the presence in the milk of copper and its compounds and was prominent in milk exposed to copper during flash pasteurization. A protective milk coating which formed on metal at temperatures above 75°C prevented direct contact of milk with metal. Several workers (26, 9) demonstrated that under some conditions, dissolved copper combined with stainless steel in a form that was readily available for contamination of milk. Absorption of dissolved copper on stainless steel occurred greatly in the pH range of 6 to 10. The amount of c0pper that absorbed increased with the concentration of copper in solution, the temperature, and the time of exposure. Krukovsky and Guthrie (66) demonstrated the promotion of the tallowy flavor in milk by adding copper in the presence of ascorbic acid. William and Burgwald (118) studied the development of oxidized flavor in milk when copper at a concentration of 2 ppm was added after pasteurization of the milk at 143°F for 30 minutes. Copper was an effective catalyst for the development of oxidized flavor in milk and the intensity of the flavor was greater when COpper was added after rather than before pasteurization (36). Pont (85) demonstrated oxidized flavor in whole milk containing 1.0 ppm added copper after 2—3 days of storage at 2-5°C. He concluded that oxidized flavor in milk was made up of an oily-metallic component, 15 produced by oxidation of fat, and a cardboard component associated with skimmilk. Mulder §£_al, (78) reported that oxidized flavor was not detected in any milk con- taining no added copper after 2-3 days storage, while a slight off-flavor developed in only a few samples after 6 days storage. The susceptibility to copper-induced oxidized flavor varied markedly from cow to cow. Ascorbic acid plays a crucial role in the develOp- ment of copper catalyzed oxidized flavor in milk. An extensive amount of work concerning this topic has been reviewed in literature by several investigators (31, 32, 39, 40, 42, 43, 68, 83, 84, 99, 104, 113, 114, 57). Corbett and Tracy (17) reported that although ascorbic acid reduced oxidized flavor development in some milks to which copper had been added, it extensively increased the flavor in others. This was confirmed by Olson and Brown (81). They showed that washed cream from suscepti- ble milk did not develop flavor when contaminated with c0pper ions and stored at 40°F but the addition of 40-200 mg/liter of ascorbic acid plus copper caused a strong oxidized flavor to develop. Brown and Olson (7) also found that resistant milk could develop the flavor following addition of copper and a.$orbic acid or dehydroascorbic acid to the milk. Similarly, Krukovsky and Guthrie (66, 67), found that-milk depleted of its natural ascorbic acid failed to develop off—flavor but 16 the flavor could be induced by the addition of reduced or oxidized form of ascorbic acid. Pont (85) reported that cardboard and metallic flavors could be induced in skimmilk, as well as whole milk and washed cream, by addition of ascorbic acid. Recently King and Dunkley (52) demonstrated that the levels of COpper naturally present in milk are suffi- cient to catalyze the development of off—flavor and that there was a strong correlation between the incidence of spontaneous development of oxidized flavor and high levels of natural copper in the milk. Smith and Dunkley (98, 101) emphasized the role of ascorbic acid in re— ducing c0pper to the cuprous state which was considered the active catalyst for phospholipid oxidation. In more recent studies El-Negoumy (28) showed that in the absence of salts, (sodium citrate, NaCl, KCl and CaCl2) ascorbic acid behaved as an antioxidant at concen- trations from 0.50-2.0 mg and a weak pro-oxidant at con— centrations from 2-10 mg per 100 ml. The reverse of this behavior was observed in the presence of salt solu- tions, with variation of oxidative intensity depending upon the composition of the model system. Roland gt_al. (92) and Roland and Trebler (91) developed a method for determining the relative sensi- tivity of milk and cream to oxidized flavor induced by pasteurization in the presence of metallic copper. The 17 sensitivity of standaridized milk and cream to copper- catalyzed oxidized flavor was apparently related to the composition of the products as determined by the fat con- tent. A variation of about one % in the whole milk caused a significant change in the flavor score. The mechanical separation of milk produced a signifi- cant decrease in its sensitivity to copper—catalyzed oxidized flavor. Thurston §£_al, (111) found that homogenization, prolonged agitation at low temperature, and freezing and thawing of milk eliminated susceptibility to development of oxidized flavor. The most likely theory to explain the non-appearance of oxidized flavor in homogenized milk was the increased absorption of protective protein on the surface of the fat-globule (112). Larsen et_al. (69) showed that homogenization of milk had a tendency to stabilize the milk against oxidation but did not increase or decrease the oxidation-reduction potential. The mechanism by which homogenization retarded the develop- ment of oxidized flavor was not associated with oxidation— reduction potential. The homogenized milk was relatively immune to oxidized flavor, since in such milk the fat re- mains scattered and the unstable lipid component of the fat-globule membrane is split and withdrawn from the surface into the interior of the fat—blobules, where it is protected by the antioxidant (65). Guthrie (38) 18 studied the development of oxidized flavors in unhomo- genized and homogenized milk. None of the homogenized samples was judged to be slightly oxidized, but a very few were doubtful. In contrast, one-half of the unhomo— genized samples develOped oxidized flavor by the 5th day of storage. Tarassuk and Koops (109) reported that the site of oxidative reaction that produces oxidized flavor in milk was the surface of fat—globules and the reactants were phospholipids and oxygen. These reactions were catalyzed by the copper-protein complex. The concentra- tion of the phospholipids and the copper-protein complex per unit of surface area was decreased proportionally to the homogenization pressure. This decrease in concen- tration per unit of the newly formed fat—globule surface was apparently the most important factor that retarded the development of oxidized flavor in homogenized milk. Brown e§_a1. (8) demonstrated that the contamination of the milk after pasteurization with amounts of copper varying up to 2.5 ppm caused a more frequent and more in— tense development of flavor than did contamination with identical amounts of copper before pasteurization. When amounts of copper ranging up to 2.5 ppm were added to susceptible milk, the intensity of the resulting oxidized flavor was greater in the raw milk than in the corres— ponding pasteurized milk where the c0pper contamination occurred before pasteurization. Gould and Sommer (35) 19 reported that the cooked flavor was closely related to heat induced inhibition of oxidized flavor if copper contamination occurred before heating, but no relation was found when copper was added after heating. Gjessing and Trout (34) exposed milk to 65°C or to 70°C for 10 minutes, and then added c0pper. A distinct oxidized flavor develOped when copper was added before heating. At the 75, 80 and 85°C exposures little difference was noted in the intensities of the developed oxidized flavor, which required the presence of 0.26 to 0.39 mg of copper per liter to develop a distinct oxidized flavor, regardless of whether the copper was added before or after heating. At 90°C, however, the oxidized flavor was not observed in the milk contaminated with 0.52 or 0.65 mg of copper per liter before heating, but developed slightly when similar amounts of copper were added after heating. Ewbank and Gould (29) found that cream containing 5 ppm of added c0pper and heated at 85°C flash and 90.6°C flash, produced butter oil of stability equal to that of a control pasteurized at 62.8°C for 30 minutes and con- taining no added copper. The stabilizing effect of temperatures of 85°C and 90.6°C was about equal regard- less of the addition of c0pper before or after heating. This heat effect may be the result of the formation of hydrogen sulfide and sulfhydryl groups which inactivated 20 the copper sufficiently to obviate its full catalytic action. Casein-copper and lactalbumin-COpper mixtures were heated at 121°C for 15 minutes by Bernhart and Linden (6). They added both heated and unheated mix- tures of ascorbic acid solution and found that heated mixtures caused a lower rate of loss of reduced ascorbic acid than did the unheated ones. When copper was added to previously heated protein solution, an effect on re- duced ascorbic acid oxidation intermediate between heated and unheated samples occurred. They concluded that high temperature heat treatment retarded pro—oxidant effects of milk copper. Smith and Dunkley (100) showed that the suscepti- bility of raw bulk milk to oxidized flavor was increased by pasteurization at 73°C for 40 seconds and decreased by homogenization and laboratory heat treatment at 75°C for 3C3minutes. Studies with chelating agents indicated that no correlation existed between catalysis of lipid oxidation and copper catalyzed oxidation of ascorbic acid. The effect of heating on the distribution of added copper in milk was studied by Sargent (96). Only 2—4% of the added amount of copper was associated with the fat-globules at 1°C while heating to 60°C enhanced this association by 600%. The extent of this migration of copper in milk was dependent upon the fat content, the total copper content and the temperature to which milk was exposed. In very 21 recent studies, Samuelsson (95) reported that upon heat— ing whole milk, the c0pper content of the cream more than doubled as the 10 minutes heating temperature rose from 50 to 80°C. A particularly sharp increase was observed between 65° and 75°C. When whole milk was heated at 80°C for 10 minutes, the changes in copper content in various fractions were observed. EXPERIMENTAL METHODS AND MATERIALS Preparation of Various Milk Proteins Sodium d- (crude) and B—caseinates were prepared according to the method of Hipp §t_al, (48). The whole casein was dissolved in concentrated aqueous urea solu- tions (6.6M), and separated into the components d- and B—casein by adding water in apprOpriate amounts. Sodium d- and B—caseins were obtained by dissolving a- and B- casein in .1N NaOH solution followed by freeze drying. B-Lactoglobulin was obtained from whole milk following the method of Aschaffenburg and Drewry (3). Fat-globule membrane protein (soluble fraction) was prepared from cow's milk according to the method of Herald and Brunner (45). Micellar casein was prepared by centrifuging skim- milk. Procedure for Binding of Copper with Proteins Seamless visking cellulose tubing of 9— or 10-inch lengths (0.984 inches flat width) were soaked three times in redistilled water. Ten milliliters of protein solu— tion of known concentration was transferred accurately 22 23 through a length of the tubing which had been tied at one end. The bag was suspended in a 25 m1 glass cylinder containing 10 m1 of copper solution. A control dialysis bag containing buffer of pH 6.5 was similarly prepared and suspended in a glass cylinder containing copper solution. The two glass cylinders were placed in cold room at approximately 35°F for a period of 72 hours, an interval sufficient for the attainment of equilibrium (60). The bags were then removed and the external solu- tion analyzed spectrophotometrically for copper (96). The extent of binding of copper ions by the different proteins under investigation was calculated from the difference in the concentration of free copper in a con— trol bag containing buffer and the one containing protein. The degree of binding was determined from a plot of the average number of moles of bound copper per molecule of protein versus the logarithm of the concentration of the. free copper ion (60). Preparation of Model Systems Milk Dialyzate Several seamless visking cellulose tubings 3.14" (flatwidth)were filled with l L of redistilled water. The dialysis tubings were sealed by tying a knot and were suspended in lO—gallon stainless steel cans con- taining raw whole milk. One lO-gallon can contained 24 only one dialysis bag. The cans were kept in a cold room at a temperature of 35°F for 36 hours, an interval suffi- cient for equilibration of water with the milk. Washed Fat-Globules Eight gallons of raw whole milk were separated at 40°C, employing a laboratory model DeLaval cream separator. The cream was washed four times with four volumes of 40°C distilled water. Preparation of Samples and Heat Treatment Four grams of washed fat-globules were added to 100 ml of milk dialyzate (4%g) in order to obtain a model system of a composition similar to normal cow's milk. To ensure thorough mixing, the mixture of fat—globules and milk dialyzate was carefully shaken. Model systems of different copper proteinates under study were suspended in a model system of washed fat-globules and milk dialyzate. The model system without c0pper was designated as the con- trol. The model systems with copper and without copper were transferred to a l L erlenmeyer flask fitted with an immersion thermometer and were subjected to heat treat- ments of 145°F for 30 minutes and l90°F for 10 minutes. During heating the model systems were mixed continuously. After cooling in ice cold water, the samples were stored at 35°F for 10 days. 25 Oxidation Studies At intervals of 2, 4, 6, 8, and 10 days the samples were removed from the cold storage and equilibrated to room temperature by placing them in a warm water bath. Oxidation was followed by performing TBA and peroxide value tests. The optical density (OD) of the TBA re- action mixture was measured at 532 mu by the method of King (58). Blank determinations were made on all the reagents and their 0D values subtracted from those of the oxidized samples. The fat sample for the peroxide test was prepared as described by Pont (84) and analyzed by a modified ferrous thiocyanate test (103). Experimentation Experiment I.—-Bindingof Copper with Micellar Casein Ten milliliters of copper solutions of the follow- 6 6M, 32 x 10—6M, ing concentrations, 80 x 10— M, 64 x 10— and 16 x 10-6M were accurately measured successively into four 25 ml graduated glass cylinders. A dialysis bag Icontaining 10 ml of micellar casein solution (pH 6.5) at a concentration of 2.000 mg/l was placed in each cy- linder. Control dialysis bags containing 10 ml of buffer (pH 6.5) were also kept in glass cylinders containing copper solutions of above-mentioned concentrations. Both sample and control cylinders were placed in cold room at a temperature of 35°F for 72 hours. The bags 26 were then removed and the external solution analyzed for copper ions. The data were represented in a graph and the moles of copper bound per mole of micellar casein were determined. Experiment 2.——Binding of Copper with Fat—globule Membrane Protein The concentration of-the fat-globule membrane protein solution used in this experiment was 3.000 mg/l. The rest of the procedure was as described in the pre- ceding experiment. The data were presented in a graph and the moles of copper bound per mole of fat-globule membrane protein were read from the graph. Experiment 3.-—Binding of Copper with Sodium a-Caseifiate The concentration of the sodium a—caseinate solu— tion used in this experiment was 2.436 mg/l. The rest of the procedure was similar to that of Experiment I. The data were represented in a graph and the moles of copper bound per mole sodium a-caseinate were determined. Experiment 4.—-Binding of Copper with Sodium B—Caseinate The concentration of the sodium B-caseinate solu- tion employed in this experiment was 1.205 mg/l. The remaining procedure was as described in Experiment I. The data were shown in a graph and the moles of copper 27 bound per mole of sodium B—caseinate were read from the graph. Experimentp5,--Binding,of Copper with geLactoglobulin The concentration of B-lactoglobulin solution used in this experiment was 1.420 mg/l. The rest of the pro— cedure was as described in Experiment I. The data were represented in a graph and the moles of copper bound per mole of B-lactoglobulin were determined from the graph. Experiment 6.--Lipid Oxidation in Model Systems of Copper- Micellar Caseinate, Washed Fat-Globules and Milk Dialyzate The copper proteinate solution of micellar casein (3.5% V/V) was suspended in the model system of washed fat-globules and milk dialyzate. The model system with- out copper was specified as the control. The model sys- tems with and without copper were given the heat treat- ments of 145°F for 30 minutes and 190°F for 10 minutes. After cooling, the samples were stored at 35°F for 10 days. At 2-, 4-, 6-, 8-, and lO—day intervals, oxidation was followed by TBA and peroxide valde. Data were shown in graphs by plotting the TBA peroxide values against the storage time. 28 Experiment 7.--Lipid Oxidation in Model Systems of Copper-Fat- Globule Membrane Proteinate, Washed Fat—GIobules and Milk Dialyzate The preceding experiment was repeated with the exception that the fat-globule membrane protein was substituted for micellar casein. Experiment 8.--Lipid Oxidation in Model Systems of Copper Sodium a-Caseinate Washed Fat— Globules and MilkIDialyzate The copper proteinate of sodium a-caseinate was suspended in the model system of washed fat—globules and milk dialyzate. The remaining procedure was similar to that of Experiment 6. Experiment 9.-—Lipid Oxidation in MOdeI Systems of copper Sodium B-Caseinate, Washed Fat- Globules and Milk DiaIyzate The copper proteinate of sodium B-caseinate was suspended in the model system of washed fat-globules and milk dialyzate. The rest of the procedure was as de— scribed in Experiment 6. Experiment 10.--Lipid Oxidation in Model Systems of Copper-B- Lactoglobulin Proteinate, Washed Fat-Globfiles and Milk Dialyzate The copper proteinate of B-lactoglobulin was sus- pended in the model system of washed fat-globules and milk dialyzate. The remaining procedure was similar to that of Experiment 6. 29 Experiment ll.--Lipid Oxidation in Heated and Unheated Model Systems ofICopper-B-Lactoglobulin Proteinate, Washed‘Fat-Globules and'MiIk Dialyzate Model systems of B-lactoglobulin, washed fat- globules and milk dialyzate were heated separately at 145°F for 30 minutes and 190°F for 10 minutes. After cooling, the model system of copper proteinate of B-lactoglobulin (3.5% V/V) was suspended in a model system of washed fat-globules and milk dialyzate. An unheated model system of copper proteinate of B- lactoglobulin was also incorporated into an unheated model system of washed fat-globules and milk dialyzate. The samples were stored at 35°F for 10 days. At 2-, 4-, 6-, 8-, and lO-day intervals, oxidation was studied by running TBA and peroxide value tests. Experiment 12.--Lipid Oxidation in Heateduand Unheated Model Systems of Copper—Fat-Globule- Membrane Proteinate, Washed Fat— Globules and Milk Dialyzate The procedure of the preceding experiment was re- peated with the exception that the fat-globule membrane protein was substituted for B-lactoglobulin. RESULTS AND DISCUSSION The copper—protein complex catalyzes lipid oxida— tion in fluid milk systems. The influence of copper- protein ratio on lipid autoxidation may account for some inconsistent statements in the literature regarding the catalytic activity of copper when in combination with proteins. In experiments with a model system of linole— ate, addition of 10'” M albumin decreased the catalytic activity of 10‘“ M copper in the presence of ascorbic acid (98). Lea (70) used c0pper concentrations of 0.4 x 10‘6 to 2.4 x 10‘6 M with 6 x 10'” M albumin (4%) and found that at these high concentrations of copper, the COpper catalysis of fat oxidation was inhibited. In contrast, Tappel (108) and Jenness and Patton (50) con— sidered that the combination of c0pper with protein in- creased the catalytic activity of copper in milk. In View of these somewhat conflicting reports, and also be— cause of precise data regarding the exact amounts of copper bound by individual milk proteins were lacking, appropriate experiments were designed to determine the binding of copper with micellar casein, fat-globule membrane protein, sodium d-casein, sodium B-casein and B—lactoglobulin. 30 31 The extent of binding of copper by the above— mentioned proteins is illustrated in Figures 1 to 5. The degree of binding is represented in graphs of the average number of moles of copper bound per mole pro- tein versus the logarithm of the concentration of un- bound copper. These data were obtained by the technique of equilibrium dialysis at a pH of 6.5 to approximate conditions of hydronium ion concentration in cow's milk. The number of moles of copper bound per mole of micellar casein is shown in Figure 1. One mole of micellar casein bound 67 moles of copper. In order to compare relative binding capacities, the number of moles of copper bound per mole of protein was determined at a free copper con- centration of 1 x 10"5 M which is equivalent to 3 ppm of copper ion surrounding the dialysis bag following equili- brium for 72 hours. The number of moles of copper bound per mole of fat-globule membrane protein is presented in Figure 2. One mole of fat—globule membrane bound 24 moles of c0pper. The number of moles of copper bound per mole of sodium a-caseinate is given in Figure 3. One mole of sodium d—caseinate bound 10 moles of copper. The number of moles of copper bound per mole of sodium 8- caseinate is depicted in Figure 4. One mole of sodium B—caseinate bound two moles of copper. The number of moles of COpper bound per mole of B—lactoglobulin is shown in Figure 5. One mole of B-lactoglobulin bound 2.5 moles of copper. 32 MOLES BOUND COPPER I MOLES TO'IAL PROTEIN I | l ’6 -5 -4 L06 COPPER FREE Figure 1. Binding of COpper with micellar casein. MOLES BOUND COPPER / MOLES TOTAL PROTEIN 33 Figure 2. it -3. L06 COPPER FREE Binding of copper with fat-globule membrane protein. 34 MOLES BOUND COPPER / MOLES TOTAL PROTEIN Figure 3. O—L 1— I -5 -4 L06 COPPER FREE Binding of copper with sodium a-caseinate. 35 2 E 3.. E ,_ g t9 6. f3 8 s. 2 \ a: 4__ E 8 3.. Q o s 2~ g ._ 2 ht? i it -3 Figure 4. LOG COPPER FREE Binding of copper with sodium s—caseinate. 36 g .L E g 7. l9 .. ‘ét’ g 5- E n g o 8 ._. 2 D 8 2" i3 ._ _| 2 OLTE S: i -3 LOG COPPER FREE Figure 5. Binding of copper with B—lactoglobulin. 37 The results clearly demonstrate the great differ- ence in ability of certain milk proteins to bind copper ions. Micellar casein, obtained by high speed centrifu- gation of skimmilk bound the greatest amount of OOpper. The fat-globule membrane protein, which of course, is intimately associated with the lipids of the fat-globule, demonstrated the second greatest affinity for copper ions. Much lesser amounts of copper were bound by other pro- teins studied: sodium d—caseinate, B-lactoglobulin and sodium B-caseinate. These results agree with the obser- vation of King g£_a1_(54) who observed that most of the added copper (75%) was associated with the casein fraction. King and William (53) later on also reported that the OOpper binding activity of the fat-globule membrane was greater than that of some other copperébinding substances in skimmilk. This seems to be consistent with the much greater concentration of natural copper in combination with the fat-globule membrane material as compared to the other protein—containing fractions of milk. Koops (62) showed that affinity of the proteins to copper at pH 6.8 was decreased in the following order: fat-globule mem— brane protein, sodium caseinate, B-lactalbumin and 8- lactoglobulin. According to Fiess and Klotz (30), the affinity of the proteins to copper at pH 6.5 was found to decrease in the following sequence: d-casein, B- casein, bovine serum albumin, B-lactoglobulin and 38 lysozyme. Menger and Mulder (73) reported that added copper was primarily associated with skimmilk proteins. Herald gp_§1. (44) found that many of the trace elements (copper, iron, zinc, silver, etc.) reported in the whole milk were concentrated on the fat-globule membrane possi- bly as constituent elements of various metabolic com- plexes. Samuelsson (93) reported that copper is associ- ated to a certain extent with the fat-globule membrane protein, but most copper was found in the serum. This implies that the substances which accompany the fat, namely the membrane proteins have a great capacity for binding copper. Certainly this may well be important in the oxidation of milk as the phospholipids are present on the surface of the fat-globule. The beneficial effect of heating fluid milk in excess of temperatures normally used in pasteurization on the resistance of the resulting dry whole milk to the development of tallowy flavor was first demonstrated by Holm g£_§1. (49) in 1926. This observation has been corroborated by many other workers in the field. Later Gould and Sommer (35) found a relationship between high heat treatment and the appearance of cooked flavor when copper was added before heat treatment. Further, their study showed hydrogen sulfide to be correlated with the cooked flavor and to be involved in the heat retardation of the development of the oxidized flavor. These findings 39 have been verified by the work of Josephson and Doan (51). Greenbank (36) reported that the thermal inhibi— tion of oxidized flavor acts through a lowering of the oxidation reduction potential. This has been confirmed by both the work of Gould and Sommer (35) and Josephson and Doan (51) who found that the sulfide compounds, liberated when a cooked flavor develops, lowered the potential of milk. According to Thomas (110) the heat— liberated sulfhydryl groups presumably function as an antioxidant in the liquid system through preferential oxidation to some undetermined product. In order to determine the effect of low and high heat treatments on various copper—protein complexes sus- pended in a model system of washed fat-globules and milk dialyzate, a series of experiments was conducted. A model system of washed fat-globules and milk dialyzate was used in all of these experiments because the use of such model system will overcome some of the inherent problems associated with the complexity of milk. The extent of lipid oxidation in the model systems of copper micellar caseinate, washed fat-globules and milk dialyz- ate is shown in Figures 6a and 6b. Maximum TBA values corresponding to an optical density of 0.32 and peroxide values of 11.0 milliequivalent of oxygen per kg of fat were obtained in this experiment. Results presented in Figure 6a indicate that the unheated model systems con- taining copper micellar caseinate and micellar casein 40 ,4 CONTROL CDPPER PROTEINATE O 0 UN HEATED A A HEATED TO l45°F FOR 30MIN. D I PEATED TO I90°F FOR IO MIN. .3 .320_ ’07 '0 3 .2 ). g .240 g .200t. .J < 0 160 F H— 8 .20- at t t 4. DAYS AT 35°F Figure 6a. Lipid oxidation in model systems of copper- micellar caseinate, washed fat—globules and milk dialyzate (TBA). 41 CONTROL COPPER PROTEINATE l3. 0 O UNI-£ATED A A HEATED TO I45‘F F0? 30 MIN. I20 0 I HEATED TO IQO‘F FOR IOMIN ILO I00 I RD 20 $0.. 50.. MILLIEOUIV. Ozl Kg. FAT. 40.. 10.. 20 8 3 D3 Bi ‘3’ i E t .‘ .3 DAYS AT 35°F Figure 6b. Lipid oxidation in model systems of copper— micellar caseinate, washed fat—globules and milk dialyzate (P.V.). 42 exhibited the greatest amount of oxidation of the six model systems. The systems heated to 145°F for 30 minutes showed less oxidation than unheated systems. The systems heated to 190°F for 10 minutes demonstrated the least -amount of oxidation compared to unheated systems. The extent of oxidation in the model system containing copper micellar caseinate increased considerably with the in— crease of storage period. The maximum increase in oxi- dation in such systems was observed after 10 days storage. This of course is similar to the pattern of oxidation which occurs in normal whole cow's milk when copper is present. Milk containing OOpper in a quantity sufficient to catalyze lipid autoxidation (.05 - 5.0 ppm) will in- variably be objectionably oxidized within 10 days storage at 35-40°F. The systems subjected to both low and high heat treatments exhibited significant amount of oxidation after 10 days of the storage period. Model systems con- taining only micellar casein showed very slight oxidation during the storage period of 10 days. Results presented in Figure 6b demonstrate that peroxide values for unheated model system containing copper micellar casein showed an abrupt fall after the 6th day of storage. Similar results were obtained from un- heated model systems which contained only micellar casein. However, the heated systems with and without copper did not show a similar fall in the peroxide values. 43 Wilkinson (120) reported a similar phenomenon. Accord— ing to him the relationship of peroxide value to off- flavor does not always hold well in dairy products in which oxidation has been catalyzed by added copper, the peroxide value often being lower than that of control samples to which no copper has been added. Ritter and Nussbaumer (89) found that OOpper added to butter-fat accelerated peroxide formation in the presence of light. Later they reported (90) that small amounts of copper palmitate caused a large increase of peroxide value, but large amounts induced little or no change. The reason for this phenomenon is that OOpper not only accelerates the assimilation of oxygen to increase the peroxide num— ber, but also hastens the further decomposition of the fat peroxides to free radicals. This suggests why the peroxide number may decrease in the presence of greater amounts of copper and why this value taken at a given time in a stored sample is not an infallible index of acceptability. Results of lipid oxidation in model systems of copper-fat-globule membrane proteinate, washed fat- globules and milk dialyzate are shown in Figures 7a and 7b. Maximum TBA values of .115 Optical density and peroxide values of 3.78 milliequivalents of oxygen per kg of fat were obtained in this experiment. The results presented in Figure 7a are similar to those shown in 44 J20 - “0.. OPTICAL DENSITY (532) S .OKL. CONTROL COPPER PROTEINATE O O UNHEATED A A HEATED T0 I45’F FOR 30MIN. D I HEATED T0 I90‘F FOR IO MIN. Figure 7a. Nb- .L 0H- 5.. DAYS AT 35° F Lipid oxidation in model systems of copper-fat- globule membrane proteinate, washed fat—globules and milk dialyzate (TBA). 45 CONTROL COPPER PROTEINATE 4.00.. O 0 WHEATED A A HEATED TO I45°F Fm 30MIN. D I HEATED T0 l90°F FOR I0 MIN. 360- 3.20- |...: E 2.8m, d x \ 2.40, >0 2.00_ 8 lg 160.. 2 LZOL .BOL. Nin— b— 0)— 6h- DAYS AT 35°F Figure 7b. Lipid oxidation in model systems of copper-fat— globule membrane proteinatt washed fat—globules and milk dialyzate (P.V.). 46 Figure 6a. However, in unheated model systems containing copper-fat-globule membrane proteinate, the oxidation was significantly lower than those containing copper micellar caseinate. Both low and high heat treatments inhibited the extent of oxidation. Model systems subjected to high heat treatment indicated the least amount of oxidation. Control model systems heated to 145°F for 30 minutes and to 190°F for 10 minutes did not show any considerable difference in the intensity of oxidation. The extent of oxidation increased with increasing time of storage in the model systems containing copper-fat-globule membrane protein. In unheated and heated control model systems, the extent of oxidation accelerated up to the 6th day in storage; thereafter a decline was noted. Results shown in Figure 7b demonstrate that peroxide values, for unheated model systems containing copper-fat-globule membrane pro- teinate agree closely with those containing copper micellar caseinate (Figure 6b). The curves for peroxide values ob- tained from heated systems containing copper-fat-globule membrane proteinate are characterized by an initial linear portion, followed by a gradual decrease in slope, termi- nating in a rather abrupt leveling off at some maximum value. Peroxide values obtained from heated control model systems showed minimal changes throughout the period of storage. Figure 8 indicates the results of lipid oxidation in model systems of copper sodium d—caseinate, washed OPTICAL DENSITY (532) 47 JZO_. .IIO_ ‘g I .090__ .080 ORO_. 040.. .030__ £20.. INO_. CONTROL COPPER PROTEINATE 0 0 UNHEATED A A HEATED T0 I45°F FOR 30MIN. D I I‘EATED T0 90°F FOR IOMN. NL— I I I I Figure 8. 4 6 f IO DAYS AT 35°F .Lipid oxidation in model systems of copper sodium a—caseinate, washed fat—globules and milk dialyzate. 48 fat-globules and milk dialyzate. Maximum TBA values corresponding to an optical density of about 0.77 were Obtained in this experiment. The data presented in Figure 8 are also somewhat similar to those shown in Figures 6a and 7a. The extent of oxidation in unheated model systems with and without OOpper increased up to 8th day in storage, afterwards the extent of oxidation decreased in intensity. The unheated model system con- taining copper sodium a-caseinate (Figure 8) was less strongly oxidized than systems containing copper micellar caseinate or copper-fat-globule membrane proteinate (Figures 6a and 7a). However, in systems containing no added copper the unheated model system containing sodium a-caseinate was more strongly oxidized than those con- taining micellar casein and fat-globule membrane protein. The low and high heat treatments caused a decrease in oxidation in model systems of sodium d-caseinate with and without copper. The extent of oxidation increased with increasing storage period in the heated model system con- taining copper sodium d-caseinate. In case of the heated model system containing sodium a-caseinate, the extent of oxidation increased up to 6th day of storage period, after which a decrease occurred. Figure 9 shows the results of lipid oxidation in model systems of OOpper sodium B—caseinate, washed fat- globules and milk dialyzate. Maximum TBA values OPTICAL DENSITY (532) 49 .QO_. HO_. IOO_. .Ofll_ .050 __ .040 ._ .030 _. .020 ,. CONTROL COPPER PROTEINATE O O UNHEATED A A HEATED TO I45°F FOR SOMIN. D I HEATED TO |90°F FOR IO MIN. Figure DAYS AT 35°F 9. Lipid oxidation in mOdel systems of COpper sodium B-caseinate, washed fateglobules and milk dialyzate. 50 corresponding to an optical density of about 0.107 were obtained in this experiment. The results shown in Figure 9 indicate some similarity to those presented in Figures 6a, 7a, and 8. The extent of oxidation was directly prOportional to the storage period in unheated and heated model system with and without COpper. Unheated model system containing copper—sodium B—caseinate (Figure 9) was less strongly oxidized than those containing OOpper micellar caseinate and copper-fat-globule membrane protein- ate (Figures 6a and 7a). However the unheated model system containing copper sodium B-caseinate was somewhat more oxidized than that containing sodium d-caseinate (Figure 8). The high heat treatment caused the greatest decrease in oxidation in model system containing copper sodium 8— caseinate. As shown by the curves in Figure 9, the effect of low and high heat treatments on the lipid oxidation in model systems without copper was over—lapping. The over- lapping effects of high and low heat treatments varied with the storage period of the samples. Possibly the lessened effects of heat on the model systems of control (containing no added copper) d— and B- caseinate are due to the fact that these proteins as isolated and prepared are not typical of their natural state in milk. They have in effect been fractionated from the much larger natural casein micelle by treatment with urea and their binding affinities are much lower. Thus, when resuspended in the milk dialyzate, they now 51 exist more or less as a synthetic entity and are less effective in binding metals or functioning as catalysts. The results of lipid oxidation in model systems of copper B-lactoglobulin, washed fat-globules and milk dialyzate are demonstrated in Figure 10. Maximum TBA values corresponding to an Optical density Of about 0.92 were Obtained in this experiment. The results presented in Figure 10 are similar to those shown in Figures 6a, 7a, 8, and 9. The extent of oxidation was directly pro- portional to the storage period in heated and unheated model systems with and without OOpper. The unheated model system containing cOpper B-lactoglobulin proteinate (Figure 10) was less strongly oxidized than those contain- ing OOpper micellar caseinate, copper-fat-globule mem- brane proteinate and OOpper-sodium B—caseinate (Figures 6a, 7a, and 9). The unheated model system containing copper-B-lactoglobulin was more strongly oxidized than that containing OOpper-sodium d-caseinate. Both low and high heat treatments decreased the extent of oxidation in model systems containing copper-B—lactoglobulin pro- teinate. The high heat treatment caused the greatest amount of decrease in the extent of oxidation in model systems with and without copper. The data presented (Figures 6a, 6b, 7a, 7b, 8, 9, and 10) thus far have indicated that unheated model sys- tems containing OOpper micellar caseinate and copper- fat—globule membrane proteinate exhibited the greatest OPTICAL DENSITY (532) J30 CONTROL (DPPER PROTEINATE r- (3 II IflMHEAWED’ A A HEATED T0 I45‘F FOR 30MIN Cl I HEATED T0 I90'F Fm IOMIN. % I % Jr is DAYS AT 35°F Figure 10. Lipid oxidation in model systems of copper—B— lactoglobulin proteinate, washed fat-globules and milk dialyzate. .1 lililjl‘lllu"1tl]llll.llll|lj 53 amount of oxidation in their systems. This was attri— buted to greater copper binding capacity of micellar casein and fat—globule membrane protein. The results obtained from experiments involving lipid oxidation in model systems clearly show that the model systems sub- jected to low and high heat treatments exhibited less tendency to oxidize than unheated controls. The model system subjected to the highest heat treatment demon— strated the least amount of oxidation. The decrease in the extent of oxidation suggests that heating of copper proteinates, washed fat—globules and milk dialyzate at 145°F for 30 minutes causes the formation of reducing substances. The reducing substances in turn reduce the pro—oxidant activity of copper and lessenthe extent of oxidation in the model systems. Heating of above systems at 190°F for 10 minutes results in the formation of greater quantities of reducing substances, probably the most important of which are sulfhydryl groups. The heat- activated sulfhydryl groups may complex with copper (as does mercury to form mercaptides) during high heat treat— ment and bring about further reduction in the pro-oxidant properties of copper. Systems of copper proteinates, washed fat-globules and milk dialyzate heated at 145°F for 30 minutes would of course show much lower levels of heat activation. Similar Observations have been demon- strated by several workers (41) who reported that high 54 temperature, short time heat treatments above 212°F imparts greater reducing prOperties to fluid milk than holder treatment at lower (145°F) temperatures. It is quite Obvious from examination of the data in this investigation and also from the literature al- ready published on the subject, that heating of fluid milk induces an ill defined change in the structure of some of the milk components with the result that milk fat becomes more resistant to oxidation. Radema (86) has demonstrated that heating high fat milk or cream to 85°C or higher for periods up to 30 minutes brings about a shift in some of the phospholipid material from the fat-globule surface. This indicates a definite alter- ation of the fat-globule surface upon high heat treat- ment. It is recognized that significant amounts of copper and iron in milk are combined with the fat- globule membrane or are absorbed on the fat—globule surface as a metal protein complex, and the ionic con— centrations of these metals is extremely low. Mulder and Koppeljan's data (77), and also this research, show that concentration of copper at the fat globule inter- face may be quite high. This proximity of copper and fat in milk may enable the copper to function effectively as a catalyst in lipid oxidation. It is probable that the effect Of heat on milk may be the displacement of OOpper from the fat globule surface to a position or configuration where it would be less able to catalyse 55 oxidation. There is also a possibility that heat may alter the metal in denatured protein particles and make it less active as a catalyst. Copper is a well-known catalyst of oxidized flavor in milk and dairy products. Tappel (108) has reported that "catalysis of unsaturated fat oxidation by copper proteins may be the cause of oxidative rancidity in butter and other dairy products." Virtually no ionic copper is present in milk; however, both copper and iron catalyze the oxidation of fat in dairy products. Dills and Nelson (22) isolated a copper-bearing protein from cow's milk and found that the copper in the protein was non—ionic and could not be removed by dialysis at pH 6.5. Copper-proteins formed by the binding of copper ions to conalbumin, serum albumin or caseinate appear to be more effective catalysts for linoleate oxidation than OOpper alone. The increase in oxidation of linoleate when catalyzed by copper proteins as compared to the catalytic effect of ionic copper is attributed to in- creased ease of formation and increased stability of the intermediate complex of linoleate peroxide-OOpper-protein (108). According to Tappel (108), OOpper proteins proba- bly catalyze the oxidation Of linoleate by reacting with linoleate peroxides and catalyzing their decomposition into chain-initiating free radicals. The overall reaction might be represented as follows: 'Illlsl‘ (III I 56 Cu++ - protein + ROOH ——-—-> 1300‘ — Cu++ — protein * * ++ -————+ R0 + OH + Cu - protein In the present investigation, copper-proteins formed by the binding of copper to micellar casein, fat—globule membrane protein, sodium a-casein, sodium B-casein, and B-lactoglobulin seem to be strong catalysts for lipid oxidation in the model system of washed fat—globules and milk dialyzate. Additional experiments were carried out in order to find the effect of low and high heat treatments on lipid oxidation in model systems (copper proteinate, and washed fat-globules and milk dialyzate) in which each of these components was individually heated. The heated components were then combined as a model system. The intent of these experiments Was to determine whether any alteration in proxidant activity of copper-protein occurred when the constituents of the model system were heated separately. Complex interactions could conceiv— ably occur between metal protein and lipo-protein when an entire model system was heated. Such interactions would be minimized if constituents were individually heated and then combined. The results Of lipid oxidation in heated and unheated model systems of OOpper B-lactoglobulin, washed fat-globules and milk dialyzate are shown in Figures 11a and 11b. The data presented in Figure 11a were quite II.‘ II 57 o UNHEATED _,,o__ O CU-PROTEINATE HEATED TO I45°F FOR 30MIN. A cu- -PROTEINATE HEATED TO ISO‘F FOR IOMIN. o CU-PROTEINATE AND MILK DIALYZATE HEATED .l20_ TO I45°F FOR BOMIN I cu NAT ”0L To Esngém co m? MILK DIALYZATE HEATED OPTICAL DENSITY (532) 8 . I I. P 4. DAYS AT 35°F Figure 11a. Lipid oxidation in heated and unheated model systems of copper-B-lactoglobulin proteinate, washed fat—globules and milk dialyzate (TBA). 58 200 C>IJMHEATED ' " ID CU-FTKHTBNWTE IHEATEDI'KJI459F FORENDNHN. A CU-PROTEINATE HEATED T0 I90’F FOR I0 MIN. 0 CU-PROTEINATE AM) MILK DIALYZATE HEATED Lao. TO I45°F FOR 30 MIN. ll OU-PREHTBNATE INN) NNLK DUMJHZATE IfliATED TOIEKPT’ROR KDNMN. LSO... ° 9.: 4 IL I.40+. 6 X . \ I20- 0| 0 O a >' '°°I- ° 8 . A g .80... / _] . =1. / 2 mp . /. 40.. .20_. I I I J I 2 4 6 8 IO DAYS AT 35°F Figure 11b. Lipid oxidation in heated and unheated model . systems of copper-B—lactoglobulin_ proteinate, washed fat—globules and milk dialyzate (P.V.). 59 different from those shown previously in Figures 6a, 7a, 8, 9, and 10. The results shown here (Figure 11a) on the basis of TBA values indicated that unheated model systems demonstrated lesser amounts of oxidation than those containing heated (low and high temperature) copper proteinate. A maximum decrease in the oxidation was found in the model systems of copper B-lactoglobulin, washed fat-globules and milk dialyzate when they were heated separately to 190°F for 10 minutes and then com— bined. The results shown in Figure 11b indicated from peroxide values that the unheated model systems were more strongly oxidized than those subjected to low and high heat treatments. When both copper proteinate, and the model system of washed fat-globules and milk dialyzate were heated individually to 190°F for 10 minutes, the extent of oxidation was decreased. Such heated systems did not show any significant increase in the extent Of oxidation with increasing storage period. Figures 12a and 12b indicated the results of lipid oxidation in heated and unheated model systems of copper- fat-globule membrane proteinate, washed fat—globules and milk dialyzate. Most of the results presented in Figure 12a were similar to those shown in Figure 11a. However, higher TBA values were obtained (Figure 12a) from model systems containing heated copper proteinate than those shown in Figure 11a. The higher TBA values could be _ related to higher amount of oxidation in these systems. 60 OPTICAL DENSITY (532) £30.. £20 110 O UNHEATED 0 CU- PROTEINATE HEATED TO I45°F FOR 30MIN. A CU-PROTEINATE HEATED TO I90°F FOR IOMIN. U CU-PROTEINATE AND MILK DIALYZAT H A TOP-15°F Fm 30MIN. E E TED I CU PROTEINATE AND MILK Y A T TOISO’FFCRDMIN. MZTEHEAED Figure 12a. t I a. t 3. DAYS AT 35°F Lipid oxidation in heated and unheated model systems of copper-fat—globule membrane proteinate, washed fat—globules and milk dialyzate (TBA). 61 400 O UNHEATED " O CU-PROTEINATE HEATED 1'0 I45°F FORSOMN. A CU-PROTEINATE I-EATED TO l90°F Fm IO MIN. 0 CU-PROTEINATE AND MILK DIALYZATE HEATED 3.60 .. TO I45°F FOR SOMIN. I CU PFD'II’EINATE AND MILK DIALYZATE HEATED TO ISO’F FOR IO, MIN. ...I_ 5.: E . 280.. O X E 240.. y 200 8 |_l_-| LOO .I =1 2 I220 .80.. .40.. 1 I E 4 I 1 f T5 DAYS AT 35" F Figure 12b. Lipid oxidation in heated and unheated model systems of copper—fat—globule membrane proteinate, washed fat—globules and milk dialyzate (P.V.). 62 Unlike the results shown in Figure 11a, the extent of oxidation decreased in both heated systems (copper protein- ate, and model system of washed fat-globules and milk dialyzate) after the termination of the storage period (Figure 12a). The results indicated in Figure 12b are essentially similar to those presented in Figure 11b in that peroxide values of systems containing heated metal proteinate and heated fat—globules exhibited minimal oxidation. The results shown here agreed with those pre- sented in Figures 6a, 7a, 8, 9, 10, 11a, 11b and 12a in that the high heat treatments for short time had the maximum inhibiting effect on the extent of oxidation. Although minimal oxidation occurred in systems in which both copper proteinate, and fat-globules and milk dialyz- ate had been heated, the systems containing heated metal proteinate and unheated fat-globules did not show any decrease in oxidation. Inspection of the data in Figures 11a and 12a indi- cates that there is beneficial antioxidant effect of heating the milk dialyzate and fat—globules. Heating the metal proteinate itself without subjecting the fat- globules and milk dialyzate to similar heat treatment did not reduce oxidation in the system. Such evidence would appear to substantiate the data of Sargent (96) and others (77) who found that heating fat-globules in their natural environment caused a shift in trace metal distribution in III l.‘ I I I‘ll" |I Ill-IIJII' . Ii] I It 63 the milk. Heating fat—globules apparently causes a significant alteration in either the configuration or concentration of autoxidizable material at the fat- globule interface. The decrease in oxidation in both heated systems (copper proteinate, and washed fat— globules and milk dialyzate) is also attributed to the reduction in the pro-oxidant activity of copper protein- ates. This decrease in the pro—oxidant properties Of copper-proteinate added to heated model system of washed fat-globules and milk dialyzate as compared to this property when copper proteinate was added to the unheated model system, is probably due to binding of copper by some reducing substances formed during the heat treat- ment. The individual heating Of copper proteinates and model system of washed fat-globules and milk dialyzate did not seem to yield results significantly different than those obtained from the experiments where both systems (copper proteinate, and washed fat-globules and milk dialyzate) were heated together. l'i‘ii.lllilli; iiliiii I‘ll-lull SUMMARY The binding of OOpper by d-casein, B—casein, B- lactoglobulin, fat-globule membrane protein, and centri- fuged micellar casein was studied by equilibrium dialysis at pH 6.5. Solutions of the above proteins, in concen— trations of 2.436, 1.205, 1.420, 3.000, and 2.000 mg/l, respectively, were placed in washed dialysis bags and equilibrated for 72 hours against each Of four copper solutions. The concentration of copper outside the mem— brane was 16x,32x,64x,or 80 x 10-6M. Plots of moles of copper bound per mole of protein versus the log of the unbound copper yielded typical curves. The results indi- cated the maximum copper binding capacity of micellar casein and fat-globule membrane protein. The affinity of the proteins to copper at pH 6.5 was decreased in the following order: centrifuged micellar casein, fat— globule membrane protein, sodium d-caseinate, B—lacto- globulin and sodium B-caseinate. Model systems Of copper proteinate of micellar casein, fat—globule membrane protein, sodium d—casein, sodium B-casein and B-lactoglobulin were suspended in a model system of washed fat-globules and milk dialyzate. The model system without copper was designated as the 64 65 control. The model systems with copper and without copper were subjected to heat treatments of 145°F for 30 minutes and 190°F for 10 minutes. After cooling, the samples were stored at 35°F for 10 days. At 2—, 4—, 6-, 8— and lO—day intervals oxidation was studied by con— ducting TBA and peroxide value tests. Data were shown in graphs by plotting the TBA and PV values against the storage time. Results indicated that the rate of oxi— dation was higher in unheated systems than in those heated to 145°F for 30 minutes. The systems heated to 190°F for 10 minutes exhibited the least amount of oxi— dation. The extent of oxidation in most of the model systems under investigation was directly proportional to their storage period. Likewise, in systems in which the protein or metallo-protein were heated separately and then combined with a system of washed fat globules and milk dialyzate which also had been heated separately, lower levels of oxidation, as measured by TBA values, were noted in heated systems. In systems in which the metallo-protein had been heated, but the fat—globules and milk dialyzate had not been heated, such antioxidant properties were not observed. These results indicate that much of the beneficial effects of heat treatment arise from the effects of heat on materials adsorbed at the fat—globule. BIBLIOGRAPHY 66 BIBLIOGRAPHY Alifax, R. and Bejambes, Marie. 1956. Tests for Simultaneous Determination of Iron and Copper in Milk and Dairy Products. Ann. Inst. Nat. Rech. Argon. Paris Ser. E., 2:619 (cited from Dairy Sci. Abstr. 19:775, 1957). Allen, J. E. 1950. The Oxidation of Ascorbic Acid in Dairy Products. J. Dairy Res., llz54. Aschaffenburg, R. and Drewry, J. 1957. Improved Method for the Preparation of Crystalline B- Lactoglobulin and d-Lactalbumin from Cow's Milk. Biochemical J., §§:273. Baker, H. P., and Saroff, H. A. 1965. Binding of Sodium Ions to B-Lactoglobulin. Biochem., 4:1670. Barska, Celina, Bednarczyk, W., and Luczak, Maria. 1959. A Rapid Method of Determining Copper and Iron in Milk, Cream and Butter. Proc. 15th. Intern. Dairy Congress, 3:1754. Bernhart, F. W., and Linden, Elizabeth. 1950. The Effect of Heat Treatment on the Pro—oxidant Activity of Copper in Milk. J. Dairy Sci., 33:166. Brown, W. C., and Olson, F. C. 1942. Oxidized Flavor in Milk. XII. Further Studies of Ascorbic Acid Mechanism in the Production of Oxidized Flavor in Milk. J. Dairy Sci., g§:1041. Brown, W. 0., Thurston, L. M., and Dustman, R. B. 1936. Oxidized Flavor in Milk. III. The Time of Copper Contamination During Production and Process- ing, and Aeration Versus No Aeration as Related to Oxidized Flavor Development. J. Dairy Sci., 19:753. Brown, W. C., and Thurston, L. M. 1940. A Review of Oxidation in Milk and Milk Products as Related to Flavor. J. Dairy Sci., g3z629. 67 10. ll. 12. 13. 14. 15. 16. 17. 18. 19. 2C. 68 Carr, 0. W. 1953. Studies on the Binding of Small Ions in Protein Solutions with the Use of Membrane Electrodes. III. The Binding of Chloride Ions in Solutions of Various Proteins. Arch. Biochem. and Biophys., 4§:417. Carr, C. W. 1953. Studies on the Binding of Small Ions in Protein Solutions with the Use of Membrane Electrodes. II. The Binding of Calcium Ions in Solutions of Bovine Serum Albumin. Arch. Biochem. and Biophys., 43:147. Carr, C. W. 1953. Studies on the Binding of Small Ions in Protein Solutions with the Use of Membrane Electrodes. IV. The Binding of Calcium Ions in Solutions of Various Proteins. Arch. Biochem. and Biophys., 4§:424. Carr, C. W. 1952. Studies on the Binding of Small Ions in Protein Solutions with the Use of Membrane Electrodes. I. The Binding of Chloride Ion and Other Inorganic Anions in Solutions of Serum Albumin. Arch. Biochem. and Biophys., 49:286. Carr, C. W., and Woods, K. R. 1955. Studies on the Binding of Small Ions in Protein Solutions with the Use of Membrane Electrodes. V. The Binding of Magnesium Ions in Solutions of Various Proteins. Arch. Biochem. and Biophys., Sizl. Fraenkel-Conrat, Heinz, and Feeney, Robert E. 1950. The Metal—Binding Activity of Conalbumin. Arch. Biochem., 32:101. Conn, L. W., Johnson, A. H., Trebler, H. A., and Karpenko, V. 1935. Determination of Minute Amounts of Copper in Milk. Ind. Eng. Chem., 1:15. Corbett, W. J., and Tracy, P. H. 1941. Experiments on the Use of Certain Antioxidants for Control Of Oxidized Flavor in Dairy Products. Food Res., Qz445. Dahlberg, A. C., and Carpenter, D. C. 1936. The Influence of Method of Sterilizing Equipment Upon Development of Oxidized Flavor in Milk. J. Dairy Sci., 19:541. Davies, W. L. 1933. The Mode of Combination and Distribution of Traces of Heavy Metals in Dairy Products. J. Dairy Res., 4:255. Day, E. A. 1960. Autoxidation of Milk Lipids. J. Dairy Sci., 4;:1360. - 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 69 Digeser, A. 1939. A Study of the Occurrence and Prevention of an Oily Metallic Off—Flavor in Milk. Molkerei—Z, 33:407 (cited from Chem. Abstr., 33: 5077, 1939). Dills, W. L., and Nelson, J. M. 1942. Isolation of a COpper Bearing Protein from Cow's Milk. J. Amer. Chem. Soc., 33:1616. Drabkin, D. L. 1939. Report on Copper. J. Assoc. Offic. Agr. Chem., 33:320. Duin, H. Van, and Brons, C. 1963. A Method for the Determination of the Copper Content Of Butter. Neth. Milk and Dairy J., 33:323. Duin, H. Van, and Brons, C. 1965. A Method for the Rapid Determination of the Copper Content of Milk and Milk Powder. Neth. Milk and Dairy J., 33:298. Dunkley, w. L., and King, R. L. 1959. Adsorption fig Copper on Stainless Steel. J. Dairy Sci., 33: 0. Elvehjem, C. A., Steenbock, H., and Hart, E. B. 1929. The Effect of Diet on the Copper Content of Milk. J. Biol. Chem., 33:27. El-Negoumy, A. M. 1965. Relation of Composition of the Aqueous Phase to Oxidized Flavor Development b§ Dialyzed Globular Milk Fat. J. Dairy Sci., 4 :1406. Ewbank, F. C., and Gould, I. A. 1943. Oxidation of Butter Oil as Influenced by Previous Heat Treatment of the Oil, Butter or Cream. J. Dairy Sci., 33:409. Fiess, H. A., and Klotz, I. M. 1952. The Thermo- dynamics Of Metallo-Protein Combinations. Com— parison of Copper Complexes with Natural Proteins. J. Am. Chem. Soc., 13:887. Forster, T. L., and Sommer, H. H. 1951. Manganese, Trypsin, Milk Proteins and the Susceptibility of Milk to Oxidized Flavor Development. J. Dairy Sci., 33:992. Garrett, 0. F. 1941. Some Factors Affecting the Stability of Certain Milk Properties. V. Inter- relations of Certain Metals and Metallic Ions and the Development of Oxidized Flavor in Milk. J. Dairy Sci., 33:103. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 7O Gehrke, C. W., Runyon, C. V., and Pickett, E. E. 1954. A Quantitative Spectrographic Method for the Determination of Tin, Copper, Iron, and Lead in Milk and Milk Products. The Effect of Storage on the Concentration of These Metals in Evaporated Milk. J. Dairy Sci., 33:1401. Gjessing, E. C., and Trout, G. M. 1940. Ascorbic Acid and Oxidized Flavor in Milk. II. The Effect of Various Heat Treatments of Milk Upon the Stability of Ascorbic Acid and Upon the Development of the Oxidized Flavor. J. Dairy Sci., 33:373. Gould, I. A., and Sommer, H. H. 1939. Effect of Heat on Milk, with Special Reference to the Cooked Flavor. Mich. Agr. Expt. Sta. Tech. Bu11., 164:48. Greenbank, G. R. 1948. The Oxidized Flavor in Milk and Dairy Products. A Review. J. Dairy Sci., 33:913. Grubb, Hanne, and Jorgensen, K. 1966. An Improved Method of Analysis for the Determination of the Copper Content of Milk. Proc. 17th. Intern. Dairy Congress, 3:221. Guthrie, E. S. 1956. Development of Oxidized Flavors in Unhomogenized and Homogenized Milk. J. Dairy Sci., 33:219. Guthrie, E. S. 1955. Role of Ascorbic Acid in Development of Oxidized Flavors in Dairy Products. Mem. Cornell Agr. Expt. Sta., 340. Guthrie, E. S., and Krukovsky, V. N. 1949. Effect of the Quick and Complete Elimination of Vitamin C on the Development of the Oxidized Flavors in Homogenized Milk, with Special Reference to the Action of Daylight. J. Dairy Sci., 33:786. Harland, H. A., Coulter, S. T., and Jenness, R. 1952. The Interrelationships of Processing Treat- ments and Oxidation—Reduction Systems as Factors Affecting the Keeping Quality of Dry Whole Milk. J. Dairy Sci., 33:643. Hartman, S. 1944. De Oxydative Virkningen i Smor og maelk. Denmark Statens Forogsmer jeri Beretning (cited from J. Dairy Sci., 33:780, 1959). 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 71 Hartman, G. H., and Garrett, O. F. 1943. Some Factors Affecting the Stability of Certain Milk Properties. VI. Relations of the Concentration of Dissolved Oxygen to the Oxidation of Ascorbic Acid and to the Development of Oxidized Flavor in Milk. J. Dairy Sci., 33:337. Herald, C. T., Brunner, J. R., and Bass, S. T. 1957. A Spectrographic Analysis of Two Protein Fractions of the Fat—Globule Membrane of Normal Cow's Milk. J. Dairy Sci., 33:446. Herald, C. T., and Brunner, J. R. 1957. The Fat- Globule Membrane of Normal Cow's Milk. 1. The Isolation and Characteristics of Two Membrane— Protein Fractions. J. Dairy Sci., 33:948. Herder, P. C. Den, and Krol, B. M. 1950. Routine Determinations of Copper, Iron and Manganese in Butter. Neth. Milk and Dairy J., 3:42. Hetrick, J. H., and Tracy, P. H. 1945. The Deter- mination of Copper in Milk Powder by a Direct Carbamate Method. J. Milk Technol., 3:5. Hipp, N. J., Groves, M. L., Custer, J. H., and McMeekin, T. L. 1952. Separation of d—, B- and y-Casein. J. Dairy Sci., 33:272. Holm, G. E., Greenbank, G. R., and Deysher, E. F. 1926. Results of Preliminary Experiments Upon the Effect of Separating, or Clarifying, and Pasteuriz— ing of a Milk Upon the Keeping Quality of its Milk Powder. J. Dairy Sci., 3:512. Jenness, R., and Patton, S. 1959. Principles of Dairy Chemistry. John Wiley and Sons, New York. Pp. 378—379. Josephson, D. V., and Doan, F. J. 1939. Observa— tions on Cooked Flavor in Milk——Its Source and Significance. Milk Dealer, 33 (2):35. King, R. L., and Dunkley, W. L. 1959. Relation of Natural Copper in Milk to Incidence of Spontaneous Oxidized Flavor. J. Dairy Sci., 33:420. King, R. L., and William, W. F. 1963. COpper Distribution in Milk During Early Lactation. J. Dairy Sci., 33:11. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 72 King, R. L., Luick, J. R., Litman, I. I., Jennings, W. G., and Dunkley, W. L. 1959. Distribution of Natural and Added Copper and Iron in Milk. J. Dairy Sci., 33:780. King, L. F., and Black, R. G. 1963. Determination og Cfipper in Butteroil. Aust. J. Dairy Tech., 13:1 5. King, L. F., and Black, R. G. 1964. Estimation of Copper in Butter and Butteroil. Aust. J. Dairy Tech., 33:27. King, R. L. 1963. Oxidation of Milk Fat Globule Membrane Material. 11. Relation of Ascorbic Acid and Membrane Concentrations. J. Dairy Sci., 46:267. King, R. L. 1962. Oxidation of Milk Fat Globule Membrane Material. 1. Thiobarbituric Acid Reaction as a Measure of Oxidized Flavor in Milk and Model Systems. J. Dairy Sci., 33:1165. Klotz, I. M. 1952. Some Metal Complexes with Pro- teins and Other Large Molecules. P. 428. Modern Trends in Physiology and Biochemistry. Academic Press, Inc., New York. ‘ Klotz, I. M., and Curme, H. G. 1948. The Thermo- dynamics Of Metallo-Protein Combinations. COpper with Bovine Serum Albumin. J. Amer. Chem. Soc., 70:939. Klotz, I. M., Walker, F. M., and Pivan, R. B. 1946. The Binding of Organic Ions by Proteins. J. Amer. Chem. Soc., 33:1486. Koops, J. 1963. Verslagen van het Nederlands Instituut voor Zuivelonderzoek nr 80, P. 131. Koppeljan, C. A., and Mulder, H. 1953. The Copper Congents of Milk. Proc. 13th Intern. Dairy Congress, 1:4 1. Krol, B. M., and Herder, P. C. Den. 1955. A Routine Method for the Determination of Copper in Milk and in Dried Milk. Neth. Milk and Dairy J., 3:56. Krukovsky, V. N. 1952. The Origin of Oxidized Flavors and Factors Responsible for their DevelOp- ment in Milk and Milk Products. J. Dairy Sci., 35:21. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 73 Krukovsky, V. N., and Guthrie, E.‘ E 1945. Ascorbic Acid Oxidation a Key Factor in the Inhibition or Promotion of the Tallowy Flavor in Milk. J. Dairy Sci., 33:565. Krukovsky, V. N., and Guthrie, E. S. 1946. Vitamin C, Hydrogen Peroxide, Copper and the Tallowy Flavor in Milk. J. Dairy Sci., 33:293. Krukovsky, V. N., Loosli, J. K. , and Whiting, F. 1949. The Influence of Tocopperols and Cod Liver Oil on the Stability of Milk. J. Dairy Sci., 33:196. Larsen, P. B., Gould, I. A., and Trout, G. M. 1941. Oxidation-Reduction Potentials and Oxidized Flavor in Homogenized Milk. J. Dairy Sci., 33g789. Lea, C. H. 1936. Antioxidants and the Preservation of Edible FatS. J. Soc. Chem. Ind., 33:293. Leviton, Abraham, and Pallansch, M. J. 1960. Binding of Riboflavin and Riboflavin Phosphate by the Pro- teins of Milk. J. Dairy Sci., 33:1713. Menger, J. W. 1957. Natural and Added Copper in Milk. Landbouwk. Tijdschr. ,'s Grav. , 69:111. (Cited from Dairy Sci. Abstr., 33:598,1957). Menger, J. W., and Mulder, H. 1956. Natural and Added Copper in Milk. Zuivelbereid. en Handel, 33:528 (Cited from Dairy Sci. Abstr., 33:146, 1957). Miller, D. E., and Tracy, P. H. 1952. The COpper Content of Butter Made by a Continuous Method. J. Dairy Sci., 33:292. Morrison, 8. L., and Paige, H. L. 1946. Modified All-Dithizone Method for Determination of Traces of Copper. Anal. Chem., 33:211. Mulder, H., Menger, J. W., and Merjers, P. 1964. The Copper Content of Cow's Milk. Neth. Milk and Dairy J., 33:52. Mulder, H., and Koppeljan, C. A. 1953. The Distri- bution of Copper and Iron Over the Phases of Milk. Proc. 13th. Intern. Dairy Congress, 3:1402. Mulder, H., Geurts, T. J., and Langurer, J. 1966. An Observation on the Spontaneously Developed Oxidized Flavor in Milk. Proc. 17th. Intern. Dairy Congress, 3:313. 79. 80. 81. 82. 83. 8M. 85. 86. 87. 88. 89. 90. 7M Munch-Petersen, S. 1950. On the Copper Content of Milk Fresh from the Cow, in Bottled Whole Milk and in Other Milk Products. Nord. Vet. Med., 2:697 (Cited from Dairy Sci. Abstr., 33:633, 195?). Olling, C. J. 1963. A Routine Method for the Estimation of Copper in Liquid Milk and Milk Powder. Neth. Milk and Dairy J., 33:295. Olson, F. C., and Brown, W. C. 1942. Oxidized Flavor in Milk. XI. Ascorbic Acid, Glutathione, and Hydrogen Peroxide as Mechanisms for the Pro- duction of Oxidized Flavor. J. Dairy Sci., 33:1027. Perrin, D. R., Lightfoot, F. R., and Moir, G. M. 1951. Copper and Iron in Dairy Products. Improved Methods of Analysis. J. Dairy Res., 33:77. Peterson, R. W., and Walton, J. H. 1943. The Auto— xidation of Ascorbic Acid. J. Amer. Chem. Soc., 33:1212. Pont, E. G. 1955. A De—emulsification Technique for Use in the Peroxide Test on the Fat of the Milk, Cream, Concentrated and Dried Milks. Aust. J. Dairy Technol., 33:72. Pont, E. G. 1952. Studies on the Origin of Oxidized Flavor in Whole Milk. J. Dairy Res., 33:316. Radema, L. 1956. The Effect ofansteurization on the Partition of Phosphatides Between the Fat and the Aqueous Phase of Cream. Proc. 114th Intern. Dairy Congress, 3:HO3. Rice, F. E., and Miscall, J. 1923. Copper in Dairy Products and Its Solution in Milk Under Various Conditions. J. Dairy Sci., 3:261. Riel, R. R., and Sommer, H. H. 1955. Isolation and Characterization of Some Oxidation Products Associated with Oxidized Flavor. J. Dairy Sci., 33:1215. Ritter, W., and Nussbaumer, T. 1938. The Oxidation of Butterfat. I. Peroxide Formation in Butter Fat. Schweiz. Michztg., 6M:59 (Cited from Chem. Abstr., gg:71u8, 1938). "' Ritter, W., and Nussbaumer, T. 1938. The Oxidation of Butterfat. II. Influence of Copper on the Proxide Number of Butter Fat. Schweiz. Milchztg., 33:215 (Cited from Chem. Abstr., 33:71u8, 1938). 91. 92. 93. 9h. 95. 96. 97. 98. 99. 100. 101. 102. 75 Roland, C. T., and Trebler, H. A. 1937. The Effect of Fat Content on Oxidized Flavor in Milk and Cream. J. Dairy Sci., 33:3“5. Roland, C. T., Sorensen, C. M., and Whitaker, R. 1937. A Study of Oxidized Flavor in Commercial Pasteurized Milk. J. Dairy Sci., 333213. Samuelsson, E. G. 1966. The Copper Content in Milk and the Distribution of COpper to Some Various Phases of Milk. Milchwissenschaft, 33:335. Samuelsson, E. G. 1966. Determination of Trace Amounts of Copper in Milk by Neutron Activation Analysis. Milchwissenschaft, 33:31. Samuelsson, E. G. 1966. Copper Distribution in Milk and the Effect of Different Heat Treatments on It. Svenska Mejeritidn, 58:663 (Cited from Dairy Sci. Abstr., 33:231, 196777 Sargent, J. S. 1964. The Effect of Heat Treatment on the Distribution of Residual and Added Copper in Fluid Milk Systems. Ph.D. thesis, Michigan State University, East Lansing. - Smith, A. C. 1967. Rapid Methodesfor Determining COpper Content of Milk. J. Dairy Sci., 33:664. Smith, G. J., and Dunkley, W. L. 1962. Pro-oxidants in Spontaneous DevelOpment of Oxidized Flavor in Milk. J. Dairy Sci., 33:170. Smith, G. J., and Dunkley, W. L. 1961. The Mechanism of Spontaneous Peroxidation in Milk. J. Dairy Sci., 33:1152. Smith, G. J., and Dunkley, W. L. 1962. COpper Bind- ing in Relation to Inhibition of Oxidized Flavor by Heat Treatment and Homogenization. Proc. 16th Intern. Dairy Congress, 3:625. Smith, G. J., and Dunkley, W. L. 1962. Ascorbic Acid Oxidation and Lipid Peroxidation in Milk. J. Food Sci., 33:127. Sollner, K. l9u3. A New Potentiometric Method to Determine Cations and Anions with Collodion and Protamine-collodion "Membrane Electrodes." J. Amer. Chem. Soc., 33:2260. 103. 10“. 105. 106. 107. 108. 109. 110. 111. 112. 113. 76 Stine, C. M., Harland, H. A., Coulter, S. T., and Jenness, R. 195”. A Modified Peroxide Test for Detection of Lipid Oxidation in Dairy Products. J. Dairy Sci., 33:202. Stotz, E., Harrer, C. J., and King, 0. G. 1937. A Study of Ascorbic Acid in Relation to Copper. J. Biol. Chem., 119:511. Strickland, R. D., Freeman, M. L., and Gurule, F. T. 1961. Copper Binding by Proteins in Alkaline Solution. Anal. Chem., 33:5U5. Stull, J. W. 1953. The Effect of Light on Acti— vated Flavor Development and on the Constituents of Milk and Its Products. J. Dairy Sci., 33:1153. Swartling, P., and Mattsson, S. 195“. Determination of Copper and Iron in Butter, Milk, and Cream. Meddelande “2, Alnarx Institute, Sweden. 1955. Catalysis of Linoleate Oxi- Tappel, A. L. dation by Copper-Proteins. J. Amer. Oil Chem. Soc., 33:252. Tarassuk, N. P., and Koops, J. 1960. Inhibition of Oxidized Flavor in Homogenized Milk as Related to the Concentration of Copper and Phospholipids per Unit Fat Globule Surface. J. Dairy Sci., 33:93. Thomas, Alan. 195“.. A Study of the Mechanism of Sulfhydryl Action in Milk. Ph.D. thesis, Uni- versity of Minnesota, Minneapolis. Thurston, L. M., Brown, W. C., and Dustman, R. B. 1936. Oxidized Flavor in Milk. II. The Effects of Homogenization, Agitation and Freezing of Milk on its Subsequent Susceptibility to Oxidized Flavor Development. J. Dairy Sci., 33:671. Thurston, L. M. 1938. Theoretical Aspects of the Cause of Oxidized Flavor Particularly from the Lecithin Angle. J. Dairy Sci., 33;257. Tobias, J., and Herreid, E. O. 1959. Determination of Degradation Products of Ascorbic Acid. J. Dairy Sci., 333428. 113. 115. 116. 117. 118. 119. 120. 121. 122. 77 Trout, G. M., and GJessing, E. C. 1939. Ascorbic Acid and Oxidized Flavor in Milk. I. Distri- bution of Ascorbic Acid and Occurrence of Oxidi- zed Flavor in Commercial Grade A Raw, in Pasteurized Irradiated and in Pasteurized Milk Throughout the Year. J. Dairy Sci., 33:271. Vandevelde, A, J. J. 1923. The Precipitation of Milk Proteins by Copper Salts. Lait, 3:337 (Cited from Chem. Abstr., 33:3211, 1923). Willard, H. S., and Gilbert, C. S. 1953. Influence of Dairy Equipment on Copper Content and Intensity of Oxidized Flavor in Milk, Cream and Skimmilk. Milk Dealer, 33 (6):“6. William, W. 1931-1932. The Determination of Copper and Iron in Dairy Products. J. Dairy Res., 3:93. William, E. E., and Burgwald, L. H. 1931. The Effects of the Direct Addition of Carotene and Mixed TOCOpherols on the Development of Oxidized Flavor in Milk. J. Dairy Sci., 33:539. Willis, J. B. 1963. The Determination of Copper in Butter and Butter Oil by Atomic Absorption. Aust. J. Dairy Technol., 33:70. Wilkinson, R. A. 1963. Theories of the Mechanisms of Oxidized Flavour Development in Dairy Products. Internal Report No. 3, C.S.I.R.0., Melbourne. Zak, B., and Ressler, Newton. 1956. Simultaneous Micro-Determination of Copper and Iron Using Mixed Phenanthrolines. Anal. Chem., 33:1158. Zak, B., and Ressler, Newton. 1958. Serum Copper and Iron on a Single Sample. Clinical Chem., 3:43. APPENDIX 78 79 APPENDIX A. Copper-Free Glassware All glassware used for copper analysis was immersed in concentrated nitric acid for two minutes, followed by five thorough rinsings with double-distilled water. The electrical conductivity of redistilled water was always less than 1.0u mhos. Reagents needed for copper analysis and standard cOpper solution were prepared by employing the method of King and Dunkley (52), followed by Sargent (96). B. Copper Determination Preparation of standard curve 1. Ten milliliters of copper solutions of the follow- 6 6 ing concentrations 16 x 10- M, 32 x 10' M, 6 M, and 80 x 10"6 M were pipetted succes- 63 x 10- sively into four 125 m1 glass extraction flasks. Ten milliliters of ammonium citrate solution was added into each extraction flask. Two milliliters of 2% aqueous sodium diethyldithio- carbamate solution was added; the flasks were shaken vigorously for 2 minutes, and 15 minutes was allowed for completion of reaction. Ten milliliters of reagent grade carbon tetrachloride was added, the flasks were shaken for 5 minutes and set aside for 10 minutes. 80 5. About 3 m1 of the carbon tetrachloride layer was transferred to a cuvette, and the optical density at 335 mu was determined with a Beckman Model DU-2 spectrophotometer, slit width 0.03 mm. This procedure was followed in the various COpper—protein binding experiments in the present investigation. TABLE 1.--Binding of copper by micellar casein.a *7. Moles bound Cu Moles total protein Log cOpper free 133.5 77 —3.50 7 100.0 -3.66 57.5 -5.19 25.0 -5.70 aPlotted in Figure 1. TABLE 2.--Binding of copper by fat-globule membrane protein.a Moles bound Cu Moles total protein Log copper free 70 -3.33 57.5 -U.39 25 —U.89 15 -5.36 aPlotted in Figure 2. 81 TABLE 3.—-Binding of copper by Sodium can-caseinate.a Moles bound Cu Moles total protein Log copper free 13.1 11-“.53 11 _u.71 7 -5.u6 3'5 -507O aPlotted in Figure 3. TABLE H.-—Binding of copper by sodium B—caseinate.a Moles bound Cu — Moles total protein Log copper free 7 3.95 —u.u6 ' 3.30 —u.56 2.50 —5.15 1.00 —5.7 aPlotted in Figure u. TABLE 5.—-Binding of COpper by B-lactoglobulin.a Moles bound Cu ' Moles total protein Log copper free 77 4.63 ' ' -u.ui 3.00 —u.5u 3.00 -5.10 1.30 —5.52 aPlotted in Figure 5. 82 TABLE 6.--The TBA values obtained from lipid in model systems of copper—micellar caseinate, washed fat-globules and milk dialyzate.a Sample Treatment Days at 35°F 2 u 6 8 10 Control unheated 0.056 0.063 0.075 .068 0.069 Control at lu5°F 0.029 0.035 0.0uu .039 0.0u3 for 30 min. Control at 190°F 0.015 0.018 0.019 .019 0.019 for 10 min. Cu-P unheated 0.075 0.138 0.25M .233 0.311 Cu-P at lu5°F 0.050 0.07M 0.105 .092 0.167 for 30 min. ' Cu—P at 190°F 0.0u8 0.053 0.082 .080 0.11“ for 10 min. aPlotted in Figure 6a. Cu-P = copper proteinate. TABLE 7.--The peroxide values obtained from lipid in model systems of copper—micellar caseinate, washed fat—globules and milk dialyzate.a Days at 35°F Sample Treatment 2 u 6 8 10 H Control unheated 1.0665 .1U9 1.360 0.328 0.8U4 Control at 1u5°F 0.659 0.938 0.938 0.609 1.u06 for 30 min. Control at 190°F 0.ulu 0.u36 0.u57 0.uul 0.661 for 10 min. Cu-P unheated 3.706 9.189 9.02“ u.805 3.633 Cu-P at 145°F 1.195 3.1ul 5.016 7.032 11.300 for 30 min. Cu-P at 190°F 0.891 1.781 2.55M 3.u22 7.688 for 10 min. fl. T #7 aPlotted in Figure 6b. Cu-P = copper proteinate. 83 TABLE 8.--The TBA values obtained from lipid in model systems of copper-fat-globule membrane proteinate, washed fat-globules and milk dialyzate.a Days at 35°F 2 ‘u 8 6 8 10 Sample Treatment ‘7, Control unheated 0.023 0.025 0.03M 0.033 0.036 Control at lfl5°F 0.017 0.018 0.029 0.022 0.023 for 30 min. Control at 190°F 0.015 0.016 0.027 0.021 0.022 for 10 min. 1 Cu-P unheatgd 0.060 0.079 0.108 0.109 0.115 Cu—P at 185 F 0.030 0.0u7 0.063 0.062 0.070 for 30 min. Cu-P at 190°F 0.025 0.0N2 0.0A6 0.0H5 0.055 for 10 min. aPlotted in Figure 7a. Cu-P = COpper proteinate. TABLE 9.-—The peroxide values obtained from lipid in model systems of copper—fat—globule membrane proteinate, washed fat- globules and milk dialyzate.a f .7?— Days at 35°F 2 u T 6 8 10 Sample Treatment ,— fi vfi .T—Y Control unheated 0.621 0.663 0.488 0.518 0.A92 Control at lu5°F 0.u29 0.u06 0.u2u 0.490 0.u33 for 30 min. Control at 190°F 0.U2u 0.377 0.377 0.u01 0.u08 for 10 min. Cu-P unheated 1.711 3.770 3.586 2.H14 2.719 Cu-P at 1u5°F 0.77u 1.52u 1.992 2.086 2.297 for 30 min. Cu-P at 190°F o.u92 0.820 1.055 1.9u6 1.922 for 10 min. fi—fiw fife T aPlotted in Figure 7b. Cu—P = COpper proteinate. 84 TABLE lO.—-The TBA values obtained from lipid in model systems of copper sodium-a—caseinate, washed fat-globules and milk dialyzate.a if Days at 35°F Sample Treatment 2 A 6 8 10 Control unheated 0.036 0.033 0.046 0.049 0.044 Control at 145°F 0.020 0.025 0.038 0.039 0.036 for 30 min. Control at 190°F 0.018 0.023 0.040 0.037 0.034 for 10 min. Cu-P unheated 0.0“0 0.060 0.061 0.077 0.074 Cu—P at 1N5°F 0.039 0.038 0.039 0.056 0.062 for 30 min. Cu-P at 190°F 0.038 0.039 0.041 0.044 0.048 for 10 min. aPlotted in Figure 8. Cu-P = copper proteinate. TABLE ll.-—The TBA values obtained from lipid in model systems of COpper sodium-B-caseinate, washed fat-globules and milk dialyzate.a Sample Treatment Days at 35°F 2 4 6 8 10 Control unheated 0.026 0.033 0.036 0.040 0.043 Control at 145°F 0.017 0.020 0.024 0.024 0.029 for 30 min. Control at 190°F 0.018 0.020 0.022 0.025 0.028 for 10 min. Cu-P unheated 0.040 0.058 0.085 0.095 0.107 Cu—P at 145°F 0.029 0.035 0.044 0.044 0.054 for 30 min. Cu-P at 190°F 0.028 0.036 0.043 0.042 0.046 for 10 min. aPlotted in Figure 9. Cu—P a copper proteinate. 85 TABLE l2.--The TBA values obtained from lipid in model systems of copper B-lactoglobulin, washed fat—globules and milk dialyzate.a Days at 35°F Sample Treatment 2 4 6 8 10 Control unheated 0.020 0.025 0.025 0.022 0.028 Control at 145°F 0.014 0 021 0.020 0.020 0.025 for 30 min. Control at 190°F 0.013 0.020 0.020 0.019 0.026 for 10 min. Cu—P unheated 0.044 0.062 0.082 0.076 0.080 Cu—P at 145°F 0.029 0.035 0.040 0.038 0.049 for 30 min. Cu-P at 190°F 0.025 0.029 0.035 0.030 0.041 for 10 min. aPlotted in Figure 10. Cu—P = copper proteinate. TABLE l3.--The TBA values obtained from lipid in heated and unheated model systems of copper B—lactoglobulin, washed fat- globules and milk dialyzate.a Days at 35°F Sample Treatment 2 4 6 8 10 Unheated systems 0.043 0.042 0.045 0.056 0.073 Cu-P at 145°F 0.047 0.050 0.041 0.057 0.075 for 30 min. Cu-P at 190°F 0.052 0.062 0.049 0.061 0 075 for 10 min. Cu—P and milk dialyzate at 145 F for 30 min. Cu-P and milk dialyzate at 190 F fer 10 min. 0.036 0.034 0.037 0.047 0.065 0.039 0.032 0.028 0.042 0.058 aPlotted in Figure 11a. Cu-P = copper proteinate. 86 TABLE l4.--The peroxide values obtained from lipid in heated and unheated model systems of COpper B-lactoglobulin, washed fat-globules and milk dialyzate.a Sample Treatment Days at 35°F 2 4 6 8 10 Unheated systems 0.567 0.727 1.031 1.125 1.641 Cu-P at 145°F 0.497 0.727 0.844 1.078 1.360 for 30 min. Cu-P at 190°F 0.483 0.717 0.797 0.938 1.266 for 10 min. Cu-P and milk dia yzate at . 1&56F for 0.548 0.624 0.891 0.938 1.172 30 min. Cu-P and milk dia yzate at 190$F for 0.441 0.605 0.516 0.563 0.563 10 min. aPlotted in Figure 11 b. Cu-P = COpper proteinate. TABLE 15.--The TBA values obtained from lipid in heated and unheated model systems of copper-fat globule membrane protein- ate, washed fat-globules and milk dialyzate.a Sample Treatment Days at 35°F 2 4 6 8 10 Unheated systems 0.048 0.058 0.077 0.072 0.079 Cu-P at 145°F 0.061 0.062 0.085 0.078 0.088 for 30 min. Cu-P at 190°F 0.061 0.074 0.082 0.087 0.084 for 10 min. Cu-P and milk dia yzate at 1458F for 0.035 0.039 0.050 0.053 0.047 30 min. Cu-P and milk dia yzate at 1908F for 0.046 0.041 0.050 0.058 0.051 10 min. aPlotted in Figure 12a. Cu-P = COpper proteinate. 87 TABLE l6.--The peroxide values obtained from lipid in heated and unheated model systems of COpper-fat-globule membrane proteinate, washed fat-globules and milk dialyzate.a Days at 35°F Sample Treatment 2 4 6 8 10 l—J Unheated systems 0.609 .547 2.344 2.532 2.578 Cu-P at 145°F 0.637 1.641 2.625 2.672 2.485 for 30 min. Cu—P at 190°F 0.646 1.735 2.344 2.578 2.391 for 10 min. Cu-P and milk dialyzate at 145 F for 30 min. 0.403 0.703 0.891 0.984 1.031 Cu-P and milk dialyzate at 190°F for 10 min. 0.267 0.375 0.375 0.422 0.469 aPlotted in Figure 12b. Cu—P = COpper proteinate.