THE EFFECT 0F HEA‘! TREATMEN? ON THE DEBT-REBUTififé OF RESEDUAL AND AbDED COPPER IN FWH) MiLK SYSTEMS Thasis for fhe Elegy-es cw? Mm D. MlCl'fiGfi-‘N STATE UMNERSET‘Y' Jain $15me Sagan?- 1964' THESIS LIBRARY Michigan State University “(MAL 4-. M This is to certify that the thesis entitled THE EFFECT OF HEAT TREATMENT ON TIE. DISTRIBUTION OF RESIDUAL AND ADDED COPPER IN FLUID MILK SYSTEMS presented by John S. E. Sargent has been accepted towards fulfillment of the requirements for Ph. D. degree in Food Science 3 . Major professor Datew 0-169 THE EFFECT OF HEAT TREATMENT ON THE DISTRIBUTION OF RESIDUAL AND ADDED COPPER IN FLUID MILK SYSTEMS By John Samuel Sargent An abstract of a thesis Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Food Science College of Agriculture 1964 ABSTRACT Hhole fluid milk and model milk systems containing from lOO-SOng added Cu/l were subjected to various heat treat- ments in an attempt to investigate the mechanism by which heat imparts resistance to copper-catalyzed lipid autoxida- tien in whole fluid milk. The total copper content of centri- fugally separated cream and skimmilk fractions from 100 ml samples of milk was determined by a carbamate method utili- zing freeze-drying as a step in the analytical procedure. This innovation permitted multiple experimentation on a common sample of milk. Data from experiments with whole milk and model milk systems indicated that added copper is preferentially bound to skimmilk proteins. Storing raw whole milk treated with lOO-SOng Cu/l at 34 F for 18 hr resulted in migration of 2-4% of the added copper to the cream phase. Heating this milk to temperatures greater than 140 F increased the copper content of the cream phase by as much as 600%. The greatest change in extent of heat-induced copper migration to the cream occurred during heat treatments from 165-175 F. Milk having a total capper content of 40-24ng Cu/l exhibited maximum and constant adsorption of capper by the fat fraction following both flash and 10 minute holding at 185-200 F. when the total copper content of the milk was increased to 500-70ng Cu/l, maximum copper adsorption occurred in the cream following both flash heating at approximately 180 F [Ill-ll 1i and 10 minute holding at 170 F; heating the milk to higher temperatures resulted in desorption of capper from the cream phase. Supplemental data from washed cream experiments in- dicated that heat-induced adsorbed copper was most tenaciously bound to the cream phase in whole milk subjected to momentary heating at 180 F. A series of experiments was performed on model milk systems comprised of washed raw cream and a milk dialyzate dispersion of centrifugally separated micellar casein, the whey protein-bearing supernatant from micellar casein, or sodium caseinate. The different heat-induced cepper migration patterns obtained from these milk systems suggests that their individual copper-protein complexes have different reactivities and stability and the overall pattern exhibited in a given milk system is a function of temperature, time of heating, fat gldbule interfacial area, protein composition and quantity. The observation that the heat-induced cepper adsorptionpdesorption phenomenon occurs in the cream phase of a milk system devoid of skimmilk proteins suggests; that this pattern is solely characteristic of the cream fraction. The addition of the sulfhydryl group blocking reagent, N-ethylmaleimide prior to heating of whole milk or model systems resulted in partial inhibition of copper adsorption by the cream phase at all temperatures of heating. Similar results were obtained when Iodoacetamide was added prior to 11 ‘l ‘I II \[.rlx/llll\ [ill fill lill‘ [ ll I'll heating of a model system containing sodium caseinate. The fact that the presence of N-ethylmaleimide partially inhibits copper migration to the cream in a milk system with a protein- free aqueous phase, suggests that copper is complexed in some fashion by heat activated sulfhydryl groups in the fat globule membrane as a result of heat treatment. Indirect evidence suggests that possibly the copper-protein complex(es) adsorbs at the fat globule interface in a manner that inhibits the catalytic ability of copper to breakdown hydroperoxides to free radicals. 111 THE EFFECT OF HEAT TREATMENT ON THE DISTRIBUTION OF RESIDUAL AND ADDED COPPER IN FLUID MILK SYSTEMS BY John Samuel Sargent A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Food Science 1964 icmomncmrs The patience, understanding and encouragement of the author's wife, Pauline, throughout the course of this grad- uate program, are gratefully acknowledged. The author most sincerely thanks Dr. c. M. Stine for his guidance and forebearance during this graduate program and for his aid in preparing this manuscript. The author wishes to acknowledge Dr. L. G. Harmon, Dr. E. J. Benne, Dr. H. A. Lillovik and Dr. F. Markakis as members of his guidance committee. Sincere appreciation is extended to Dr. L. R. Dugan for his assistance in freeze-drying procedures, to Dr. J. R. Brunner for his aid in preparing micellar casein, and to Mr. Jaswant Singh in preparing sodium caseinate for use in model milk systems. The author most gratefully acknowledges the financial assistance of the Campbell Soup Company Fellowship, 1962-64. 11 CONTENTS INTRODUCTION 0 O O O O O O O O O O O O O O O O O Rsvrswos'trrmruas.............. A. B. Copper in Whole Fluid Milk . . . . . . . 1. Level of natural copper . . . . . . . 2. Physical state of capper in milk . . 3. The distribution of copper in milk . Copper Analysis of Milk and Milk Products Copper Catalysis of Lipid Autoxidation in Fluid Milk 0 l O O O O O O O O O O O O l. Solubility of capper in milk . . . . 2. Source and degree of copper contamination 3e Ascorblcacld.......o..o. 4. Processing . . . . . . . . . . . . . EXPERIMENTAL METHODS AND MATERIALS . . . . . . . A. B. whole Milk Systems . . . . . . . . . . . 1. Sampling and preparation of milk for heat treatment . . . . . . . . . . . . . 2. Heat treatment . . . . . . . . . . . 3. Separation . . . . . . . . . . . . . 4. Copper determination . . . . . . . . Model Systems . . . . . . . . . . . . . . 1. Milk dialyzate . . . . . . . . . . . 2. Washed cream . . . . . . . . . . . . 3. Micellar casein . . . . . . . . . . . iii Page (ma-utomm \OCDGDCD 11 16 16 16 16 l6 17 18 18 18 18 Page 4. Purified casein . . . . . . . . . . . . . . 19 5. Beta-lactoglobulin . . . . . . . . . . . . 19 6. Treatment of model milk systems . . . . . . 19 C. Sulfhydryl Group Blocking Agents . . . . . . . 20 D. Experimentation . . . . . . . . . . . . . . . . 20 Experiments 1-4t The effect of flash heat treatment on the distribution of residual and added copper in fluid milk . . . . . . 20-21 Experiments 5 and 6. The effect of heat treatments on the distribution of residual and added copper in fluid milk . . . . . . 21-22 Experiment 7. The effect of 10-minute heat treatments on the distribution of copper added to fluid milk before and after heating 22 Experiment 8. The effect of flash heat treatment on the distribution of two levels of added copper in fluid milk . . . . . . . . 22 Experiment 9. The distribution of added copper in flash heated skimmilk mixed with unheated cram O 0 O O 0 O O O 0 0 O O 0 O 0 0 O 0 0 0 23 Experiment 10. The effect of washing (36 F) on heat-induced adsorbed copper in the cream fraction of fluid milk . . . . . . . . . . . 23 Experiment 11. The effect of washing (80 F) on heat-induced adsorbed cepper in the cream fraction of fluid milk . . . . . . . . . . . 24 Experiments 12 and 13. The effect of NEM on heat induced copper migration to the cream ”action or fluid milk 0 O 0 O O O O O O O O 24 Experiment 14. The effect of flash heat treat- ment on the distribution of copper in model systems containing micellar casein and super- natant from spun casein . . . . . . . . . . . 25 Experiment 15. The effect of flash heat treat- ment on the distribution of copper in a model system containing 2.5% purified casein . . . 25 iv Page Experiment 16. The effect of flash heat treat- ment on the distribution of copper in a model system containing beta-lactoglobulin . . . . 25 Experiment 17. The effect of NEM on heat-induced copper migration in a model milk system contain- ing micellar casein . . . . . . . . . . . . . 26 Experiment 18. The effect of NEM on the heat- induced migration of copper in a model milk system containing the supernatant from spun casein O O 0 0 O O O O O O O O 0 O O 0 O O O 26 Experiment 19. The effect of Iodoacetamide on heat-induced migration of copper in a model milk system containing 2.5} purified casein . .26 Experiment 20. The effect of NEM on the heat- induced migration of copper in a model milk system devoid of skimmilk proteins . . . . . 27 RESULT s O O O O O O 0 O O O O O O O I 0 O O O O O O O 2 8 A. Sensitivity of Copper Determination . . . . . . 28 B. Experimentation . . . . . . . .-. . . . . . . . 31 Experiments 1 and 2 . . . . . . . . . . . . . . 31 Experiment Experiment Experiment eeeeeeeeeeeeeeeee 37 O O O O O O O O O O O O O O O O O 41 Experiment e e e e e e e e e e e e e e e e e 43 3 4 5 Experiment 6 . . . . . . . . . . . . . . . . . 39 7 EXperiment 8 9 eeeeeeeeeeeeeeeee45 A Experiment Experiment 10 . . . . . . . . . . . . . . . . . 4? Experiment 11 . . . . . . . . . . . . . . . . . 49 Experiment 12 . . . . . . . . . . . . . . . . . 51 Experiment 13 . . . . . . . . . . . . . . . . . 53 Experiment Experiment Experiment Experiment Experiment Experiment Experiment DISCUSSION . . . SUMMARY AND CONCLUSIONS BIBLIOGRAPHY . . 14 15 16 17 18 19 '20 APPENDIXeeee'eeeeeeeeee .1. Ii. Ill. xv. V. Preparation of Iodoacetamide VI. Nitroprusside Test vi Copper Determination Preparation of Glassware Standard Copper Solution 0 Preparation of Purified Casein Page 55 57 59 61 63 65 67 68 81 84 92 92 92 95 95 96 97 Figure 1 10 11 12 13 14 FIGURES Page The effect of flash heat treatment on the distri- bution of residual and added copper in fluid milk . (Flash heat treatment, 155-195 F) The effect of flash heat treatment on the distri- bution of residual and added capper in fluid milk . (Flash heat treatment, 160-200 F) The effect of flash heat treatment on the distri- bution of residual and added copper in fluid milk . ' (Flash heat treatment, 90-200 F) The effect of flash heat treatment on the distri- bution of residual and added copper in fluid milk . (Total copper, 243ug/l) The effect of heat treatments on the distribution of residual and added cepper in fluid milk . . . . (Added copper, SOOug/l) The effect of heat treatments on the distribution of residual and added copper in fluid milk . . . . (Added copper, 300yg/l) The effect of 10 minute heat treatments on the dis- tribution of copper added to fluid milk before and ‘fter he‘tins O O O O O O O O O O O O O I O O O O O The effect of flash heat treatment on the distribu- tion of two levels of added cepper in fluid milk . The distribution of added copper in flash heated skimmilk mixed with unheated cream . . . . . . . . The effect of washing (36 F) on heat-induced ad- sorbed copper in the cream fraction of fluid milk . The effect of washing (80 F) on heat-induced ad- sorbed copper in the cream fraction of fluid milk . The effect of NEM on heat-induced copper migration to the cream fraction of fluid milk . . . . . . . . The effect of NEM on heat-induced cepper migration to the cream fraction of fluid milk . . .t. . . . . The effect of flash heat treatment on the distri- bution of copper in model systems containing micellar casein and supernatant from spun casein . vii 29 30 32 34 36 38 4O 42 44 46 48 50 52 54 Figure Page 15 The effect of flash heat treatment on the distri- bution of copper in a model system containing 2.5% purified casein . . . . . . . . . . . . . . . . . . 56 16 The effect of flash heat treatment on the distri- bution of copper in a model system containing beta- lictOglflbflln..oo...............58 17 The effect of NEH on heat-induced cepper migration in a model milk system containing micellar casein . 60 18 The effect of NE! on the heat-induced migration of copper in a model milk system containing the super- natant from spun casein . . . . ... . . . . . . . . 62 19 The effect of Iedoacetamide on heat-induced migra- tion of copper.in a model system containing 2.5% purified casein O O O O O O O O O O O O O O O O O O 64 20 The effect of NEM on the heat-induced migration of copper in a model system devoid of skimmilk pPOtaina 0 O 0 O 0 O O O O O 0 O O O O 0 0 0 0 0 O 66 viii INTRODUCTION Milk lipids are recognised as a major contributor in determining consumer acceptability of most dairy products. However, the lipid material in dairy products, like most food fate of vegetable and animal origin, is rather sensitive to deterioration'by oxidative processes. Much data have been gatheredon the conditions which initiate and accelerate these reactions. The role of metals such as copper and iron in accelerating lipid autoxidation by catalyzing the break- down of hydrOperoxides to free radicals is well known. Like- wise extensive research has demonstrated that heat treatment of milk to temperatures greater than 165 F causes the libera- tion or ”activation” of sulfhydryl groups in the non-casein protein fraction. The appearance of such available sulfhydryl groups coincides with increased antioxigenic properties of milk. Exactly why heat treatment increases the resistance of milk lipids to oxidation is not known. This investigation of the effects of heat treatment on the distribution of residual and added capper in fluid and model milk systems was performed in an attempt to elucidate the mechanism of-heat-induced resistance to metal catalyzed autoxidation. In addition, sulfhydryl group blocking agents were employed in milk systems for the purpose of characteriz- ing a possible interrelationship between heat activated sulf- hydryl groups and copper catalyzed autoxidation of milk lipids. REVIEW OF LITERATURE The impairment of flavor quality as a result of autoxi- dation of milk lipids in dairy products is widespread and manifests itself in many ways. No other chemical deteriora- tion has been studied so extensively as autoxidation, nor has any other problem probably been investigated from so many different aspects. No attempt has been made in this thesis to exhaustively review the literature unless it per- tained to the distribution of natural and added copper in whole fluid milk or to the role of copper in the catalysis of development of oxidized flavor in fluid milk. More thor- ough treatment on the autoxidation of dairy products is included in reviews by Brown and Thurston (7), Day (10), Greenbank (27), Kruisheer (45), Pent (64), Riel and Sommer (66), Stull (72), and on lipids in general by Holman (34), Mbrris (57) and Lea (52). A. Copper in whole Fluid Milk. 1. Level of Natural Copper: Milk taken directly from the cow contains a low but variable concentration of natural copper. Earlier works (9, 13) reported values ranging from 200-809ug Cu/l of uncontam- inated raw milk, while more recent research (16, 36, 45) indicated a copper content of 30-20ng Cu/l of fresh raw milk. The difference is attributed to gradual improvements in technique of analysis and sampling (9). 2 Koppejan and Mulder (43) and Monger and Mulder (54) showed clearly the variations in capper content that may be expected in milk samples taken from individual cows at diff- erent stages of lactation. Natural copper is present in high concentrations (up to 2ogug cu/l) during the first days of a lactation period but gradually declines to normal levels (20-4qug Cu/l) during the first two months of lactation. The natural cepper content of milk generally is not considered to be influenced by dietary copper (18, 54). However, Davis (12) stated that the capper content of milk can be almost doubled by a large increase in the dietary in- take by the cow. King and Dunkley (45) increased the natural copper content of cow's milk by drenching the cow with 10 g doses of cepper sulfate. 2. gaggical State of Copper in Milk. In an early study, Rice and Miscall (65) con- cluded that copper in milk was in the ionic state. Results of later investigations (1, 40) indicated that ceppsr does not exist as such, but is complexed with the lipo-protein. layer surrounding the fat globules. Davies (13) presented data from diffusion experiments which demonstrated that the amount of ionic capper in.milk was negligible at the normal pH of milk; however, the concentration of ionic capper in- creased with a decrease in pk. King et a1. (40) found natural and added copper assoc- iated with the fat globule membrane in a non-dialyzable stats. Added copper in the skimmilk was slightly dialyzable; the amount of unbound metal increased as the pH was lowered to 3.0. Diehl (14) postulated that cepper forms a complex with amino and hydroxyacids through coordinating groups so as to form chelate rings. This type of complex would inhibit the ionization of cepper in fluid milk. 3. The Distribution of Copper in Milk. Many investigators have studied the partition of copper between skimmilk and cream; however, the data are both incomplete and controversial. Some workers have re- ported that added cepper is preferentially adsorbed by the fat phase of the milk while others believe copper is primar- ily associated with the skimmilk.. Rice and.Miscall (65) found that copper was uniformly distributed in milk in accordance with total nitrogen or water content. Davies (13) reported added copper was uniform- ly distributed in gravity-separated milk, but was concentrat- ed in the cream of milk separated by centrifugation. He concluded that added copper was distributed in proportion to the surface area of the fat globules. Willard and Gilbert (81) reported that cepper became concentrated in the cream of milk subjected to contamination from copper-alloy equipment. Miller and Tracy (55) recover- ed most of the copper from the skimmilk when cream used in a continuous butter-making operation was exposed to heavy copper contamination. According to data of Hartman (30), added capper was preferentially adsorbed by the whey proteins. By fractional precipitation with ammonium sulfate, Dills and Nelson (16) isolated a cepper containing protein (0.19% Cu, 15% H) from skimmilk. This fraction had no ascorbic acid oxidase activity and the copper was non-dialyzable at pH 6.5. No attempt was made to relate this copper-proteinate to haemo- cuprein (0.34% Cu) in blood. Allan (1) studied the kinetics of ascorbic acid oxida- tion in milk and butter serum. He concluded that both added and natural copper were nonionic and were associated with the fat globule membrane. Koppejan and Mulder (43) investigated the variation and distribution of natural copper in individual milks. They found that 4-43% of the total copper was associated with the fat globules. Their limited data indicated that added copper does not accumulate at the fat globule surface. Monger and Mulder (54)likewise reported that added copper became primar- ily associated with the skimmilk proteins, and that natural copper was primarily associated with the fat globule membrane. 6 In a more recent investigation, King et a1. (40) em- ployed a radioactive tracer to determine the distribution of cepper in milk. Naturally occurring copper was concen- trated (ca. 10-35% of the copper) at the surface of the fat globules; whereas, most of the added capper was uniformly associated with the skimmilk proteins. Only 2-3% of the total added copper was associated with fat globules. Natural and added capper in association with the fat gldbules were recovered with the fat globule membrane proteins. King and williems (#1) employed Gu6‘ to determine the distribution of natural cepper in milk at intervals during early lactation. The amount of copper associated with fat globules was about 15% of the total copper 2-4 weeks after parturition and 35% after 10 weeks. A combination of cream washing and chelation by ethylenediamine tetraacetic acid (EDTA) removed up to 90% of the lipid-bound copper. However, following mixing of this washed cream with the original skim- milk and reseparation, the dapper content of the cream increased. Administration of the isotope by infusion or by drench yielded similar result with respect to distribution, but the rate of appearance in milk was slower by drenching. B. Copper Analzsis of Milk and Milk Products. Many modifications of the carbamate procedure (17, 31, 32, 44, 53, 55, 61, 74, 83) and dithizone procedure (35, 58) have been applied to the analysis of capper in dairy products. Gehrke (22) developed a spectrographic method for determining the capper content of milk and dairy products. Procedures for the simultaneous microdstermination of copper and iron in biological systems are also available (50, 84, 85, 86). .Muldsr and KoppeJan (43) proposed a mathematical calcu- lation of the quantity of copper present in milk plasma and the quantity of dapper retained by the butterfat globules. They based their calculations on tho assumptions ”that copper is equally distributed over all globules and that the differ- ence in size between the fat globules may be neglected. The calculation was carried out as follows: Suppose that 1 g of plasma is found to have 'azug of copper, and the 1 g of butterfat globules retains 'bzug of copper. Hhen the fat content of the initial milk is 4%, and that of the cream is 20%, and of its skimmilk 1%, the follow- ing equation will result: Copper in 1 kg of milk = 40b + 960a Copper in 1 kg of cream = 200b + 800s Copper in 1 kg of skimmilk = 10b + 990a. The values of 'a and 'b' can be calculated from these equations." These investigators disregarded the immense surface area presented by the small fat globules present in the skimmilk fraction. C. Copper Catalysis of Lipid Autoxidation in Fluid Milk. 1. Solubility of Copper in Milk. Miscall et a1. (56) and Gebhardt and Sommer (21) in- vestigated the factors affecting the solubility of copper in fluid milk. in increase in titratable acidity of milk de- creased tho solubility of metallic copper. Removal of milk gases or addition of carbon dioxide decrease the ability of milk to dissolve copper, whereas, the presence of oxygen in milk increased the solubility of cepper (21, 56). Maximum solubility of copper in milk occurred following a 30-120 minute exposure to copper at 158 F (21) or flash heating to 140 F (56). The solubility of copper at the boiling tempera- ture of milk was of the same magnitude as that observed at room temperature. Preheating of milk to various temperatures above 158 F decreased the solubility of copper in milk. This effect of preheating was greater as the temperature and time of preheating was increased (21). Pasteurized milk was shown to dissolve more metallic capper than raw milk at the same temperature (56). 2. Source and Degpee of Copper Contamination. The first milk which passes through dairy plant processing equipment after cleaning and sanitizing treatments is most heavily contaminated with metals. Explanations pro- posed for this greater initial contamination are based on the formation of a protective coating during product flow that retards solution of the metal (11, 15) and on the forms-- tion between processing runs of a film of soluble metal oxides this is more easily removed by the first milk (79). Krukovsky and Guthrie (46) found that 0.1 ppm of added copper caused tallowy flavors in milk not depleted of its ascorbic acid when held for 24 hours at 0-5 C. Williams and Burgwald (82) induced the typically oxidized flavor in fluid milk with the addition of 2 ppm of copper after pasteuri- zation. Pent (63) reported that the addition of 1.0 ppm of copper produced a well-defined oxidized flavor in pasteurized whole and skimmilk following storage at 3 C for 2-3 days. Under identical conditions, 0.1 ppm of capper produced a slight off-flavor described as ”cardboard” in whole milk. Hollander and Tracy (33) produced an oxidized flavor in dried whole milk when as little as 0.5 Ppm 0f copper was added during the preheat treatment of the milk. King and Dunkley (4l)(45) observed a highly significant correlation between the concentration of natural copper in milk and the incidence and intensity of spontaneous oxidized flavor. They concluded that copper present in milk as it comes from the cow is an important catalyst of oxidized flavor. 3. Ascorbic Acid. This compound plays an ill-defined role in the 10 development of cepper-catalyzed oxidized flavor in fluid milk. Considerable experimentation has been performed in an attempt to elucidate its contribution to the off-flavor in whole milk (2, 5, 19, 20, 29, 46, 47, 48, 50, 62, 63, 69, 71, 72, 80, 82). ' Guthrie and Krukovsky (28, 46) made the interesting ob- servation that milk was protected against oxidized flavor when all of the reduced ascorbic acid was destroyed. These findings were confirmed by Tobias and Herreid (78); in addi- tion, they noted that a degradation product of ascorbic acid, gulonic acid, was also destroyed before inhibition of oxi- dized flavor was obtained. With the destruction of this do- gradation product, oxidized flavor did not develop for five to seven days in milk samples treated with 1 ppm copper. In a later publication, Krukovsky and Guthrie (47) pre- sented data indicating that the oxidation of the lipid fraction of fluid milk is coupled to that of ascorbic acid when a certain equilibrium between the reduced and oxidized forms has been established. They attributed the protective influence of large amounts of ascorbic acid added to milk to the ex-‘ haustion of dissolved oxygen prior to the establishment of the favorable equilibrium between the oxidized and reduced forms of Vitamin C. Smith and Dunkley (69) concluded that at the copper 11 concentrations normally present in milk, ascorbic acid is an essential reactant in spontaneous, as in copper-induced, oxidized flavor. At capper levels greater than 1.0 ppm, ascorbic acid was not necessary for lipid peroxidation, pre- sumably because at higher copper/protein ratios, the protein rather than ascorbic acid reduces capper to the cuprous state (69). The pro-oxidative activity of ascorbic acid may depend upon its ability to reduce capper to its lower valence and to form a specific association with capper which in some unexplained manner increases the pro-oxidative properties of the milk (42). 4. Processipg. a. Separation. Roland and Trebler (67, 68) reported that the sensitivity of standardized milk and cream to copper-induced oxidized flavor appears to be related to fat content. Skim- milk exposed to metallic capper failed to produce the typical oxidized flavor. Mechanical separation of milk produced a marked decrease in sensitivity to copper-induced oxidized flavor. b. Homogenization. Homogenization retards the development of copper- induced oxidized flavor (51, 76, 77). Thurston (77) postulated the resistance of homogenized 12 milk to develop oxidized flavor is due to increased adsorption of protective protein on the surface of the fat globule. Larsen et a1. (51) reported the mechanism by which homogeni- zation prevents or retards development of oxidized flavor does not appear to be associated with the redox potential of the milk. The migration of unstable lipid components, presumably phospholipids, from the fat globules into the serum phase as a result of homogenization has been suggested as a possi- ble explanation for the retardation of oxidized flavor in homogenized milk (49, 76). King (38) recently proposed that homogenization induces an irreversible change in the struc- tural configuration of the capper-protein-lipid complex in such a way that ascorbic acid is no longer able to initiate the formation of lipid free radicals. An investigation by Tarrassuk and Keeps (75) has shown the concentration of the phospholipids and the copper-protein complex per unit of surface area to be decreased proportion- ately to the homogenization pressure. They concluded this decrease in concentration per unit of newly formed fat globule surface appears to be the most important factor that retards the development of oxidized flavor in homogenized milk. c. Heat Treatment. The addition of 2.5 ppm of capper to milk follow- 13 ing pasteurization tends to cause a more frequent and more intense development of oxidized flavor than does contamina- tion with identical amounts of capper prior to pasteurization (3, 4, 23). Similar results were obtained when copper was added to whole milk prior to spray drying (70). Gould and summer (24) and Josephson and Doan (37) de- monstrated that heat treatment of milk at temperatures high enough to produce cooked flavor resulted in milk more resis- tant to oxidation. This was attributed to production of volatile sulfides or sulfhydryl compounds which could serve as antioxidants. Gould and Sommer (24, 25) reported a correlation be- tween the development of cooked flavor and oxidized flavor in whole milk. The temperature range at which cooked flavor appeared was 167-172 F for momentary heating and 158-162 F for a 30 min. holding time. The momentary heating required to produce the cooked flavor was raised to 183-187 F when 1 ppm cepper was added after heating. When 1 ppm of cOpper was added to the milk prior to momentary heating to 183-187 F no oxidized flavor deve10ped in the milk; heating below these temperatures did not prevent the development of oxidized flavor. COpper added to milk following heating accelerated the development of oxidized flavor in all instances, even when the milk was heated to 194 F. 14 Brown and Olsen (8) performed experiments with washed cream and found that heating of the milk at 180 F for 5 min. prior to washing of the cream did not affect the susceptibi- lity of the cream to oxidized flavor when contaminated with COPPOI‘ e Forrester and Sommer (l9) attempted to relate the oxi- dation of milk protein and susceptibility of milk to oxidized flavor development. They postulated that ascorbic acid may serve as a hydrogen carrier in the oxidation of lipids and proteins, and that reactive sulfhydryl groups must be present to initiate protein oxidation. A.heat treatment of 170-180 F for 5-10 min. would initiate more active sulfhydryl groups for the oxidation of milk proteins. The available sulfhydryl groups would exhaust the oxygen present in the milk, thereby preventing lipid oxidation with its accompanying flavor de- velopment. Capper in the cuprous state promotes oxidized flavor by blocking the reactive sulfhydryl groups so that protein oxidation cannot proceed (19, 26). The amount of capper required to initiate development of the flavor would be related to the number of free sulfhydryl groups in the' milk protein. Such a postulation is one attempt to explain why the concentration of capper which must be added to pro- duce an oxidized flavor varies so greatly from one milk to another (19). In a study of the distribution of added cepper and iron 15 in fluid milk systems, Stine (70) demonstrated that heat treatment of milk had a pronounced effect on the displace- ment of iron from the fat globule interface. Data obtained from experiments wherein cOpper was added to milk prior to heat treatment were more inconclusive, ranging from positive to negative displacement of capper from the interface. These data were calculated using the simultaneous equations of Koppejan and Mulder (43) and possibly, the inconsistencies noted may be due in part to the assumptions and limitations of these equations. A thorough search of the literature available failed to disclose other possible effects of heat on the distribu- tion of natural and added copper in milk. EXPERIMENTAL METHODS AND MATERIALS A. whole Milk Systems. 1. Sampling and Preparation of Milk for Heat Treatment. Individual raw milk samples were obtained following the milking of cows picked at random from the University Holstein, Jersey and Brown Swiss herds. In order to avoid copper contamination, the milk was obtained directly from milking parlor glass ”weigh Jars“ and delivered to four liter erlenmeyer flasks that had been scrupulously cleaned with concentrated nitric acid. (See Appendix). The milk was tempered to 80 F and treated with standard copper solution (Appendix) in amounts ranging from 100 to SOO’ug Cu/l of milk. 2. Heat Treatment. Representative control and copper treated milk sam- ples in velumes slightly in excess of 100 ml were delivered to a two-neck 300 m1 round bottom flask fitted with standard taper adapters and an immersion thermometer. The milk was then flash heated under constant agitation in a boiling water bath to temperatures of 140 to 200 F in 5 and 10 degree in- crements. The heated milk was either cooled immediately or maintained at a controlled temperature for 10 minutes before cooling to 80 F in an ice bath. 3. Separation. A 100 m1 aliquot of the cooled milk was accurately 16 l7 pipeted into 125 m1 International Centrifuge separatory funnels and-either separated immediately at 27c rcf for 1 hour at 34 F, or stored 18 hours at this temperature prior to separation. The skimmilk fractions were delivered to 250 m1 beakers and the cream fractions were quantitatively rinsed from the separatory funnels into 150 m1 beakers with three 30 ml aliquots of redistilled water (conductively less than 0.9’u mho) tempered to 120 F. Representative samples of the copper treated whole raw milk and skimmilk.fractions were analyzed for fat content by the Roese-Gottlieb method (Mojonnier modification). 4. Copper Determination. The skimmilk and cream fractions were frozen to -30 F for 12 hours and subsequently freeze-dried for 48 hours in a Stokes Freeze Dryer at a pressure of 80-100 microns of mercury with water at 110 F circulating in the plates. The freeze-dried material was removed quantitatively from the beakers with the aid of a polypropylene spatula and delivered to 20 m1 platinum crucibles. The material was ashed in a muffle furnace at 550 C for 10 hours and the capper content of the resulting ash determined colorimetrically by a car- bamate method (39) (See Appendix) employing carbon tetra- chloride as the extracting solvent for the copper-carbamate complex. 18 B . Model Systems . 1. Milk Dialyzate. Strips of “Visking” seamless cellulose dialysis tubing (1 7/8" diameter) were filled with 1500 ml volumes of redistilled water. The sealed dialysis membranes were then supported in stainless steel perforated cheese strainers and completely immersed in milk (33-35 F) stored in a farm bulkemilk tank. The water was allowed to equilibrate with the bulk milk supply under intermittent agitation for a per- iod of 36 hours. 2. Washed Cream. Five gallons of Brown Swiss milk were centrifugally separated in a laboratory model De Laval cream separator. The cream was washed with equal volumes of distilled water at 120 F and reseparated a total of five times. ‘ 3. Micellar WEE-l" Fresh whole milk was obtained from individual Brown Swiss cows by the sampling method previously described. One liter volumes of the milk were centrifuged at maximum speed in an International Centrifuge at 34 F for 1 hour. The skimmilk fraction was siphoned into polypropylene centri- fuge vessels and spun in a Beckman Model L preparative centri- fuge at 41,000 rcf for 6 hours at a temperature of 32-40 F. The supernatant liquid containing the whey proteins was de- canted and stored at 34 F for use in a model milk system. The remaining micellar casein pellets were dispersed in 19 fresh milk dialyzate (32-40 F) by high speed mechanical agi- tation for 20 seconds. The redispersed micellar casein was made up to a volume of 1 l.with fresh milk dialyzate. This dispersion was then centrifuged in the preparative centrifuge as previously described for the original skimmilk. The casein pellets were once again dispersed in fresh milk dialy- zate and reprecipitated at 41,000 rcf. Following the second washing, the micellar casein pellets were dispersed by mech- anical agitation in 900 ml of fresh milk dialyzate in a cold room at 34 F for 24 hours. This final dispersion was filter- ed prior to incorporation in a model milk system. 4. gurified Casein. Sodium caseinate was prepared from isoelectric casein obtained by a method described in Biochemical Preparations (8a) (See Appendix). 5. Beta-Lactoglobulin (3X) was obtained from Nutritional Biochemicals Corporation, Cleveland, Ohio. 6. Treatment of Model Milk Systems. The prepared model systems were flash heated, cooled, separated, and the copper content of the subsequently freeze- dried fractions determined in the manner described for whole milk systems. The reduced protein content of the aqueous fractions of model milk systems necessitated the determination of capper in this fraction by difference. 20 C. Sulfhzdryl Group BlockingAgents. The blocking agents used in whole milk and model systems were added to the milk immediately before heat treatment. lodoacetamide, prepared by a method according to Anson (la) (See Appendix) and N-ethylmaleimide (NEM) were dissolved in distilled water before addition to the raw milk. Raw milk containing 500’ug added copper per liter, (1) 100 ppm NEM or (ii) 100 ppm Iodoacetamide, was subjected to the flash heat treatments employed in experiments with whole and model milk systems. A representative sample of each heat treated milk was then subjected to a nitroprusside test (37) in order to determine the extent of reaction of the sulfhydryl group blocking agents with heat activated sulfhydryl groups. D. merinentat ion. ‘gxpgriment 1 and 2. The effect of flash heat treatment on the distribution of residual and added copper in fluid milk. Morning milk was obtained from a Holstein cow in her fourth month of lactation. This milk was treated with 100 lug Cu/l, flash heated to temperatures ranging from 155 to 190 F (l) and 160 to 200 F (2) in ten degree increments, cooled, and stored 12 hours at 34 F prior to separation. Experiment 3. The effect of flash heat treatment on the distribution of residual and added copper in fluid milk. 21 The morning milk obtained from a Holstein cow in her fourth month of lactation was treated with loo/mg Cu/l. The milk was subjected to flash heat treatment to temperatures ranging from 90-200 F in five and ten degree increments, cooled and separated immediately at 270 rcf for 1 hour at 34 F. Experiment 4. The effect of flash heat treatment on the distribution of residual and added copper in fluid milk. Morning milk obtained from a Jersey cow in her fourth month of lactation was treated with 200’ug Cu/l. The milk was then flash heated to temperatures ranging from 110 to 200 F in five and ten degree increments, cooled and stored for 18 hours at 34 F prior to separation. ‘Egperiment 5; The effect of heat treatments on the distribution of residual and added copper in fluid milk. The morning milk of a Holstein cow in her fourth month of lactation was treated with 500/ug Cu/l. This milk was subjected to the following heat treatments: (A) Flash heating from 140 to 200 F in five and ten. degree increments, zero hours of storage at 34 F prior to separation, (B) Ten minute holding time at each of the flash heat treatments, zero hours of storage at 34 F prior to separation, and (C) Ten minute holding time at each of the flash heat 22 treatments followed by rapid cooling and storage for 18 hours at 34 F prior to separation. Experiment 6. The effect of heat treatments on the dis- tribution of residual and added cOpper in fluid milk. The morning milk from a Jersey cow in her second week of lactation was treated with 300/ug Cu/l and subjected to the heat treatments as described in the preceding experiment. EEperiment 7. The effect of 10 minute heat treatment; on the distribution of copper added to fluid milk before and after heating. The morning milk from a Jersey cow in her fourth month of lactation, was treated with 400’ug of Cu/l before (B) and following a 10 minute heat treatment (C) at temperatures ranging from 140 to 200 F in five and ten degree increments. A control series (A) of milk containing only residual copper was also subjected to these heat treatments. Experiment 8. The effect of flash heat treatment on the distribution of two levels of added cepper in fluid milk. The morning milk from a Jersey cow in her first month of lactation was treated with 200’ng (A) and 500’ug (B) Cu/l respectively. This milk was then subjected to flash heat treatments ranging from 150 to 200 F in five and ten degree increments. 23 meriment 2. The distribution of added copper in flash heated skimmilk mixed with unheated cream. 'The morning milk from a Brown Swiss cow in her fourth month of lactation was divided into two equal portions. One volume was treated with 500 )1g Ola/1(a) and flash heated from 160 to 200 F in five and ten degree increments. The second portion of milk containing only residual copper was delivered in 100 ml aliquots to centrifuge separa- tory funnels and centrifuged at 270 rcf for 1 hour at 34 F. The skimmilk fractions were combined and treated with 500,ng Cu/l. One hundred ml portions of this skimmilk were flash heated to temperatures from 160 to 200 F, cooled to 80 F and mixed with the unheated cream (B) from the original 100 ml sample of raw milk. This recombined milk was allowed to stand in the centrifuge separatory funnels for 2 hours at 34 F prior to separation at this temperature. Experiment 10. The effect of washing (36 F) on.heat- induced adsorbed copper in the cream fraction of fluid milk. The morning milk from a Brown Swiss cow in her fourth month of lactation was treated with 500/ug Cu/l. The treated milk was subsequently flash heated to temperatures ranging from 160 to 200 F, cooled, and separated at 34 F. The skim- milk fraction was drawn from the separatory funnel and the copper content determined. The remaining cream fraction was 24 5 washed twice with two 100 m1 aliquots of redistilled water at 36 F, followed by reseparation in an International centri- fuge for 1 hour at 34 F. The washed cream fraction was then analyzed for cOpper content. Experiment 11. The effect of washing (80 F) on heat- induced adsorbed copper in the cream fraction of fluid milk. The preceding experiment was repeated using redistilled water at 80 F to wash to cream fraction. ‘Experiment 12. The effect of NEM on heat-induced copper migration to the cream fraction of fluid milk. The morning milk from a Brown Swiss cow in her fourth month of lactation, was treated with 500/ug Cu/l (A). A portion of this milk was treated with 200 ppm N-ethylmalei- mide (B). The milks were then flash heated to temperatures ranging from 160 to 200 F, cooled to 80 F and separated at 34 F. Experiment 13. The effect of NEM on heat-induced copper migration to the cream fraction of fluid milk. . The preceding experiment was repeated with milk from a Jersey cow in her fourth month of lactation. The effect of heat treatment on the distribution of residual cOpper in the fluid milk was also examined. Experiment 14. The effect of flash heat treatment on 25 the distribution of copper in model systems containing mi- cellar casein and supernatant from spun casein. Model milk systems were prepared from milk dialyzate, washed cream, 500Ing added Cu/l, (A) micellar casein and (B) supernatant from the spun casein. These two model systems were subjected to flash heat treatment of temperatures rang- ing from 160 to 200 F, cooled, and centrifugally separated at 34 F. Experiment 15. The effect of flash heat treatment on the distribution of cOpper in a model system containing 2.5% purified casein. A model system was prepared using 2.5% purified casein (sodium caseinate), washed cream, milk dialyzate and 500lng Cu/l. This system and a control system containing no added cOpper were flash heated to temperatures from 160 to 200 F, cooled and separated at 34 F. Experiment 16. The effect of flash heat treatment on the distribution of copper in a model system containing _ Beta-lactoglobulin. A.model system comprised of 0.2% Beta-lactoglobulin (N.B.C., 3x), milk dialyzate, washed cream and 500’pg Cu/l was flash heated to temperatures from 160 to 200 F followed by cooling and separation at 34 F. 26 A control system comprised of all components but Beta- 1actoglobulin was subjected to the same series of flash heat treatments. Experiment 1]. The effect of NEM on heat-induced copper migration in a model milk system containing micellar casein. A.model system prepared from milk dialyzate, washed cream, micellar casein and 500‘pg Cu/l was flash heated to temperatures from 160 to 200 F, cooled and separated at 34 F. This procedure was repeated with 200 ppm NEM added to the system prior to flash heat treatment. Experiment 18. The effect of NEM on the heat-induced migration of cOpper in a model milk system containing the supernatant from spun casein. The preceding experiment was repeated with the exception that the supernatant from the spun casein was substituted for micellar casein dispersed in milk dialyzate. Experiment 19. The effect of Iodoacetamide on heat- induced migration of copper in a model milk system containing 2.5% purified casein. A model milk system was prepared from milk dialyzate, 'washed cream, 2.5% purified casein and SOD/pg Cu/l. A series of control samples (B) containing no sulfhydryl group block- ing agent and samples to which had been added 200 ppm 27 Iodoacetamide (A) were flash heated to temperatures from 160 to 200 F and subsequently cooled and separated at 34 F. §§periment 20. The effect of REM on the heat-induced migration of copper in a model milk system devoid of skimmilk proteins. A control model system (B) composed only of milk dialy- zate, washed cream and 500’ug Cu/l was flash heated to temp- eratures from 160 to 200 F, cooled and separated at 34 F. This model system containing 200 ppm NEM (B) was subjected to the flash heat treatments employed by the control series. RESULTS A. Sensitivity of Copper Determination. The carbamate method applied to 326 freeze-dried fractions of heated milk (Experiments l-8,inclusive) resulted in an average copper recovery of 97.9% with a standard deviation of 1'. 2.0%. Calculation of the recovery of added c0pper from approximately 1800 samples showed little deviation from this degree of accuracy. The innovation of the freeze-drying step in the carbamate procedure permitted multiple experimentation on a common sample of milk. 28 29 1“ 12m ~~~~~~~~ FAT IN MILK, 462% FAT IN SKIMMILK , O'll°/e ||0~ MILK STORED l2 Hr. AT 34°F IOO~ 9C%- 8C)- 0 ADDED COPPERUOOyg/l.) 3 70- : ONATURAL COPPER : 60- CREAM \ 0 3 (r 5C)* m o. o. O o 40- &‘~ 30— ~~~~~~~ ~~o\ o r ¢SKIM 4 20* ,,,, d—fi LLCREAM 4 0",: ZZZZZZ 0,. l0- 0 L L 1 1 L 1 80 l55 |65 I75 l85 I95 TEMPERATURE , F FIGURE 1. The effect of flash heat treatment on the distribution of residual and added copper in fluid milk. (Flash heat treatment, 155-195 F) COPPER (pg/I. of milk) 30 I20- e- ------------- - a FAT IN MILK, 3-7I /. HO FAT IN SKIMMILK, 0-I0°/. MILK STORED l2Hr. AT 34°F I00- 90. . SKIM 4 BOP 0 ADDED COPPER (lOOpg/l.) 70’ 0 NATURAL COPPER 60- CREAM 4 50- 40~ 0“ 3O” ““““““ ex A A SKIM o 20» / ,_ CREAM I’,”’ ’-O‘___—_______i u— \7 fl ‘0 ext/”x, 'OF cr” 0 l 1 I l l I 80 ISO !70 ISO I90 200 TEMPERATURE, F FIGURE 2: The effect Of flash heat treatment on the distribution of residual and added copper in fluid milk. (Flash heat treatment, 160-200 F) 31 B. I§§perimentation. (£223: The number on the graph corres- ponds to order of the performed experiment). ‘Egpgriment Nos. 1 and 2. A momentary heat treatment of Holstein milk (residual copper, 45/ng/1; added cOpper, lOOng/l) to 155 F induced a detectable migration of residual and added cOpper from the skimmilk to the cream fraction. Maximum and constant migra- tion of copper to the cream was noted in milk flash heated from 180 to 200 F. The copper content in the cream fraction of milk flash heated in this temperature range increased by as much as 420 per cent. Storage of the unheated milk for 12 hours at 34 F resulted in a 2-3 per cent increase in the copper content of the cream fraction. (Figure l and 2). | I ||'l‘|l IIJIII [fr it'll. III! {I 32 Am com-om .ucmfiueeuu umms nmcamv 33.2: 3an a.“ Monaco peeve use Hangman mo aoHusfluumwp on“. :o ucefiumouu and: E33. mo vacuum 05. .m “.55on n. . mmahITIIWINI>IS NI n3 t Lil 5. 3 O .h 0! O >I‘III‘N 3‘0 'IS/bn 1 sec . q "" 360 '0 I I I I I I I I I I I I ‘340 80 I40 I50 I60 I70 I80 I90 200 TEMPERATURE, F FIGURE 5. The effect of heat treatments on the distribution of residualland added copper in fluid milk. (Added copper, SOOpg/l) 37 Emperiment No. 5. Figure 5 illustrates the effects of three separate heat treatments of Holstein milk containing 532.P5 Cu/l. Flash heating of this milk with no storage at 34 F prior to separa- tion (A) resulted in maximum migration of copper to the cream in milk heated to 185-190 F. The total cOpper content of the cream fraction increased by 410 per cent in this temperature range. when the milk was subjected to a 10 minute holding period at each of the flash heat temperatures (B), maximum adsorption of copper by the cream was detected in milk held at 170 F. .Storage of the milk for 18 hours prior to separa- tion (0) resulted in an increase in the extent of copper migra- tion to the cream at the 170 F heat treatment. At this temp- erature, the total capper content of the cream incrased by 580 per cent. In each of the three heat treatments of the milk, desorp- tion of copper from the cream was noted as a result of heating the milk to temperatures greater than that temperature inducing maximum adsorption of copper by the cream. Cu IN CREAM, pg/l. 0F MILK of residual and added copper in fluid milk. (Added copper, 300pg/1) L“ FAT IN MILK, 4-7I°/., SKIMMILK, 0-ll% J480 ‘ ORIGINAL Cu, 220/9“. q Q; ‘ v ADDED Cu, 30qu/I. _460 0 ‘ C ...I_ W _ “s“ 2 r \ T440 (D l60_ J 5 ~ 420 Z 3 I40... OCREAM [— 3R L 4400» \-.\ \ C 380 ~— IOOL B \ R « O , ‘n L ( 360 80L . g 60— "l’ ' ‘I - 4 320 4opufifiii' ........ . _ A- MILK FLASH HEATED, OHr. STORAGE AT 34°F. 4 300 20 _ B— HE AT TREATMENT FOR IOMIn..,OHr.ST0RAGE AT 34°F -- -I C- HEAT TREATMENT FOR IOMIn., ISHr. STORAGE AT34°F 280 O I I J I I I L I I I I I 80 I40 I50 l60 I70 ISO I90 200 TEMPERATURE, F. FIGURE 6. The effect of heat treatments on the distribution 39 gaperiment No. 6. The preceding experiment was repeated with milk from a, Jersey cow in her second week of lactation. The cream frac- tion demonstrated maximum adsorption of GOpper following flash heating (A) or 10 minute holding at 180 F (B). When the milk was stored 18 hours at 34 F prior to separation (C), maximum migration of copper to the cream was observed in milk heated at 170 F for 10 minutes, as previously noted in Experi- ment No. 6. In addition, the greatest change in the extent of'heat-induced copper migration occurred in the milk sub- jected to heat treatments of 160 to 170 F. (Figure 5 and 6). 4O FAT IN MILK,6-06°/e-, SKIMMILK, 048%. RESIDUAL Cu, 55°2yg/l. ; ADDED Cy, 400pg/I. HEAT TREATMENT, IO Mins. A" CONTROL 8 " Cu ADDED BEFORE HEATING C ‘ Cu ADDED AFTER HEATING 9 '40 v AWAM' L E 5 ' E x — _, Iso r- 5 , .X LL. '—v—SK|MM|LK t O a c» O 1 'n “ Z E F “J x a: 0 5° _ Z ' 200 C 5 e |60 e 2° MA -I20 0_ I I I I I L I I I J I 80 30 I40 I50 l60 I70 I80 I90 200 T EMPERATURE, F. FIGURE 7. The effect Of 10-minute heat treatments on the distribution of copper added to fluid milk before and after heating. 41 gngriment No. 7. Figure 7 illustrates the effects of heating Jersey milk for 10 minutes at temperatures ranging from 140 to 200 F. The control sample (A) containing only residual copper demon- strated maximum and cOnstant adsorption of copper by the cream phase following 10 minute heat treatments at 180 F and above. Treating the milk with 400,ug Cu/l prior to heating (B) resulted in a maximum cOpper adsorption peak in cream phase at 170 F. 0n the other hand, when the same amount of copper was added to the milk following heat treatment (C), maximum adsorption of capper by the cream phase, but to a lesser extent than treatment B, was noted in the range of 175 ot 185 F. The desorption phenomenon was again evident in the cream fractions of copper treated milk following heating at temperatures greater than 170 F. (Curve B). CU IN CREAM,IIg/I. OF MILK 42 an mo— , 560 C M0. 32 _vV—SKIMMILK 520 (j) X '30 —.o.—CREAM g _ 480 g r. ”X _ ID \. —4OO"" O ‘ ‘TI 3mg; F X an mm FLASH HEAT TREATMENT 7° RESIDUAL COPPER, Ieepg/I. ‘ ADDED COPPER , ZOOIJg/IIA) - 24° 60 500ug/I.(BI FAT IN MILK, 508%; SKIMMILK,0°067°/e '80 5-. I I I I I I I I I I 30 I50 I60 I70 I80 I90 200 TEMPERATURE, F. FIGURE 8. The effect of flash heat treatment on the distribution of two levels of added copper in fluid milk. 43 Experiment No. 8. Milk from a Jersey cow in her first month of lactation was treated with 200 and 500,ug Cu/l respectively. At the lower level of added copper, curves labeled A on Figure 8, maximum adsorption of cOpper by the-cream occurred following flash heat treatments of 170 to 180 F. The subsequent de- crease in GOpper content of the cream at higher temperatures indicated heat-induced desorption of copper. The same milk containing the higher level of added cOpper, curves labeled B on the graph, exhibited maximum adsorption by the cream phase at 180 F. Once again, heating the milk to temperatures in excess of 180 F resulted in a lower cOpper content in the fat fraction. OF MHJ< Cu IN CREAM, )Jg/I. 44 A-Cu ADDED TO WHOLE MILK BEFORE FLASH HEAT TREATMENT B-Cu ADDED To SKIMMILK BEFORE FLASH HEATING AND MIXING WITH CREAM. FAT IN MILK, 377%, SKIMMILK, 008% RESIDUAL Cu, 44-7hg/I, ADDED CU, SOOpg/l. I20L .4500 Q R 2 U) I00_ v A 430 5 Z T 5. go_ _ CREAM 460 Q P _ SKIMMILK 440% (D >_ 420 0 ‘N 400 g I: .. T X 0 I I I I I I I I 380 80 I60 I70 I80 I90 200 TEMPERATURE,F FIGURE 9. The distribution of added copper in flash heated skimmilk mixed with unheated cream. .l 'l III III.’ III I 45' ‘Egperiment No. 9. Raw skimmilk treated with 500/ug Cu/l was flash heated from 160 to 200 F and mixed with unheated cream. A slight but maximum cOpper adsorption was detected in cream mixed with skimmilk heated to 170 F (Figure 9, Curve B). The con- trol samples (A) containing 545/ug Cu/l exhibited the char- acteristic cOpper adsorption peak in the cream of milk flash heated 180 F followed by the desorption phenomenon at higher temperatures. 46 A-TOTAL Cu IN SKIMMILK,IIq/I. OF MILK B-TOTAL Cu IN CREAM, yg/I. 0F MILK C-TOTAL Cu IN WASHED CREAM (temp. ofwcsh water, 36F) FAT IN MILK,3-97 °/e;SKIMMILK, O'IOO/o RESIDUAL Cu, 45'3}Jg/l.;ADDED Cu,5OO}Jg/I I20_ -—5IO o C ” :2 I00__ 490 U, 5 E r g u. 7: o P - 1" 60- 450‘c a. ‘3 3. _ - - o 2 40__ 430 “n :5 S “i I; 0 20... _4I0 ; —II 3 U 0 I I I I I I I I 390 80 I60 ITO I80 I90 200 TEMPERATURE,F FIGURE 10. The effect of washing (36 F) on heat-induced adsorbed copper in the cream fraction of fluid mi 1k. 47 Expgriment No. 10. The cOpper adsorption-desorption phenomenon was again observed in Brown Swiss milk treated with 500/ng Cu/l and exposed to the flash heat treatments (B). The total copper content of the cream fraction increased by 420 per cent and attained a maximum value in milk heated to 180 F. Washing the cream fractions with redistilled water at 36 F (C) re- moved a relatively constant quantity of adsorbed copper from the cream of milk flash heated from 160 to 180 F; however, washing the cream of milk flash heated to 180 to 200 F removed as much as 43 per cent of the total cOpper content. Copper was most tenaciously adsorbed in the washed cream of milk flash heated to 180 F. (Figure 10). Cu IN CREAM, pq/I OF MILK 48 A-TOTAL Cu IN SKIMMILK, pg/I. OF MILK B-TOTAL Cu IN CREAM, IIg/I. 0F MILK C-TOTAL Cu IN WASHED CREAM (temp. of wash water, 80 F) FAT IN MILK, 382%, SKIMMILK, 0-II°/e RESIDUAL Cu, 658th, ADDED CU, 500,1.Ig/l. 2 I20 -«'5I0 -z- ' (I) a 2 I00 490; I. 3‘ 80 470‘: O 2 60 45091 x i 40 430 20 4l0 O I I I I I I l I 390 80 ISO I70 I80 I90 200 TEMPERATURE, F FIGURE 11. The effect of washing (80 F) on heat-induced adsorbed copper in the cream fraction of fluid milk. 49 ‘gyperhment No. 11. The preceding experiment was repeated using redistilled wash water at 80 F (Figure 11). washing the cream with water at a higher temperature failed to alter the characteristic pattern of heat-induced copper adsorption in the cream fraction; however, the loss of total cOpper in the cream as a result of washing increased throughOut the series of heated milks. As much as 53 Per cent of the total cOpper was washed from the cream of milk heated to 180 F (C). OF MILK IN CREAM, )Ig/l. Cu 50 A- MILK TREATED WITH 500m Cu/l BEFORE FLASH HEATING B-MILK TREATED WITH 500% Cu/l, IOOppm N-ETHYL MALEIMIDE BEFORE FLASH HEATING RESIDUAL CU, 46-0pg/l. FAT IN MILK, I-73°/., SKIMMILK, 006% SOIF‘ A fl5I0 .L a 50 /I500 40 490 30 480 20_ 470 B I/ _SKIMMILK — ICE _CREAM ~460 0 I I I I I I I I 450 80 I60 I70 l80 I90 200 TEMPERATURE, F FIGURE 12. The effect of NEM on heat-induced copper migration to the cream fraction of fluid milk. 'I/brr ‘>I1IININI>IS NI no )ITIIIN :IO 51 ggperiment.No. 12. Figure 12 illustrates the effects of a sulfhydryl group blocking agent on heat-induced copper migration to the cream. The control sample (A) containing no NEM exhibited the copper adsorption peak in the cream fraction of milk flash heated to 180 F. when the milk was treated with 200 ppm NEM prior to flash heating at 180 F, the extent of migration of copper to the cream was reduced by approximately 40 per cent. The fat content of the Brown Swiss employed in this ex- periment was abnormally low as a result of inclusion of frozen silage in the cow's ration. 52 A-MILK TREATED WITH 50011:: Cu/l. BEFORE FLASH HEAT TREATMENT N-ETHYL MALEIMIDE BEFORE FLASH HEAT TREATMENT C-CONTROL_ RESIDUAL Cu, 8l-0IIg/l. FAT IN MILK, 623%; SKIMMILK, 0-|5°/e B-MILK TREATED WITH 500» Cu/I.,200ppm IOOK 4520 ‘ _ e _ (D 3 ~ A . ._ = 2 80— ' 35002 u. " T O) o — J 5 _. - ; v ; Z a - ' S :60: - . E480; 5' ” ' ‘ :5 40— A . ' SKIMMILK ~460 0: B _' - E: 0 ._ _CREAM _ >_ z A a O " 20 an 44 O :3 £;-_. .LJI e—éb‘T’A”’T‘*r 0 'j -T‘ C) r- (3 v _— g; _ _ r- O I I I I I I l I 420x 80 ISO I70 7 I80 I90 200 TEMPERATURE , F FIGURE 13. The effect of NEM on heat-induced copper migration to the cream fraction of fluid milk. 53 ggpgriment No.;:2. The preceding experiment was repeated with Jersey milk. This milk, treated with 500 ,ug Cu/l,’ (Figure 13, Curve A) demenstrated maximum copper adsorption in the cream phase at 175 F rather than 180 F as previously noted. when the milk was treated with 200 ppm NEM prior to heating (B), the extent of migration of copper to the cream as a result of flash heating at 175 F was reduced by 65 per cent. The control sample (C) containing only residual copper exhibited the characteristic constant migration of copper to the cream in milk flash heated from 180 to 200 F. An aliquot of the heated milk containing NEM was sampled following each of the flash heat treatments and subjected to the nitroprusside test. In each case, a negative nitropruss- ide test indicated that heat activated sulfhydryl groups were blocked by NEM. The same milk treated with 200 ppm iodoaceta- mide prior to flash heating from 160 to 200 F also yielded a negative nitroprusside test. OF MILK CU IN CREAM FRACTION,IIg/I. 54 MODEL SYSTEM: Milk Dialyzate, Washed plus; A- MICELLAR CASEIN B- SUPERNATANT FROM SPUN CASEIN FAT IN MILK,4-32°/e; SKIMMILK, 0.07% TOTAL Cu IN SYSTEM, 570 pg/I. Cream, 500mg Added Cull, 8" 5K) O C '2- (D 5 Z 3 I— 490 X ._ CREAM _ 5‘ ‘_ SKIMMILK _420 > (BY DIFFERENCE) 0 d O 450 z E 420 ‘Q 0 4IO '11 — - s 20L —390 I; 01 I I I I I I I I 370 80 I60 I70 I80 I90 200 TEMPERATURE, F FIGURE 14. The effect of flash heat treatment on the distribution of copper in model systems containing micellar casein and supernatant from spun casein. 55 Epperiment No. 14. Curves labeled A on Figure 14 illustrate the effect of flash heat treatment on a model milk system of milk dialyzate, washed cream, micellar casein obtained at 41,000 rcf for 6 hours, and 500,pg added Cu/l. Maximum adsorption of cOpper by the fat fraction was Observed when the system was subjected to momentary heating at 180 F; desorption of copper from the cream phase occurred at higher flash heat temperatures. System B, in which the milk dialyzate dispersion of mi- cellar casein was replaced by the supernatant of centrifugally separated casein, failed to demonstrate the cOpper adsorption peak in the cream fraction in this system, the copper content of the cream attained a maximum and constant value following flash heating from 180 to 200 F. 56 MODEL SYSTEM: Milk Dialyzaie, Washed Cream, Purified Casein, A-SOOpq ADDED Cu/I. B- CONTROL, (I30)... RESIDUAL Cu/I.) _ Cu IN CREAM FRACTION 2'0 __ Cu IN SKIM FRACTION (by difference) I\ FAT IN MILK, I2-53°/.-, SKIMMILK,0-07°/e ‘570 9 :2 I90_ A 550 Z =1 U) 2 _ 5 LI. '70-— 530 g o k -n -; I5C3a. 5:2 \\ O 5l0 o a — j 2 I30_. 490% 9 L ~ I— ‘c o IIOI— 470m: I'IIININI>IS NI no >I"IIW :IO 59 Egperiment No. 16. Figure 16 illustrates the effects of flash heat treatments on the distribution of copper in a model milk system of milk dialyzate, washed cream and 500/ug added Cu/l in the presence and absence of beta-lactoglobulin (0.2%). Heating the system containing this whey protein to temperatures of 170 to 180 F induced maximum migration of copper to the cream phase. As much as 48 per cent of the total copper in the system appeared in the cream fraction as a result of flash heating in this temperature range. (Curve A). The control system (B) containing no beta-lactoglobulin also demonstrated maximum copper adsorption by the cream fraction following flash heating from 170 to 180 F. It is interesting to note that approximately 80 per cent of'the total copper in the system was recovered from the cream fraction of samples heated to these temperatures. Furthermore, as much as 56 per cent of the added cOpper migrated immediately to the cream fraction upon addition of copper to the system. 60 MODEL SYSTEM: Milk Dialyzafe, Washed Cream, Micellar Casein, 500pg Added Cu/I., A- 200 ppm. N-thyl Maleimide Added Before Flash Healing B-Conlrol (TOTAL Cu, 6I5pg/ I.) x —Cu IN CREAM FRACTION g — Cu IN SKIMMILK FRACTION(by difference) LL FAT IN SYSTEM, 479%; SKIMMILK, 040% 2 o 280 4550 2 < A T m a - - ,X 3.240: \ 2‘5") E z. I: W _ g 200_. B v -470 I; I— - : 8 r - a E ISO: :430 g ,5- - 8 <1 _ .. Lu _. E 0: 80 #3502 0 j :— Z 40 13m % :> L g 0 0- I I I I I I I I “270 I; 80 I60 I70 I80 I90 200 TEMPERATURE, F FIGURE 17. The effect of NEM on heat-induced copper migration in a model milk system containing micellar casein. 61 §§periment No. 11. Flash heat treatment of a model system containing a milk dialyzate dispersion of‘micellar casein resulted in copper adsorption in the cream fraction.below 180 F with subsequent desorption of copper from the cream fraction at higher temp- eratures (Figure 17, Curves B). when this system was treated with 200 ppm NEM prior to flash heating, the extent of mi- gration of copper to the cream following flash heating to 180 F was reduced by 65 per cent (Curves A). Once again, the heat-induced copper adsorptionpdesorption phenomenon in the cream was evident. 62 OF MILK ‘<‘ cm I80 2‘ z.~ I60 MODEL SYSTEM- Milk Dialyzafe, Washed Cream, Supernatant from Spun Casein, 500 pg Added Cu/I., A- 200 ppm. N-thyl Maleimide Added Before Flash Healing B - Control I40 I20 I00 IN CREAM FRACTIO CU m o 0 L FAT IN -MILK,4-79°/e -, SKIMMILK, O-I0% 4603 :. TOTAL Cu IN SYSTEM, 540pg /I. Z I; _Cu IN CREAM FRACTION 4400) P _Cu IN SKIMMILKIby difference) g ‘ L 2 — O _ B If 420 r- x I. 400 t L A IQ T A 380 - \ g _ v B 360 :. s _ I- _ 340x 1 320 — I I I I I L I I 300 80 I60 I70 I80 I90 200 TEMPERATURE, F FIGURE 18. The effect of NEM on the heat-induced migration of copper in a model milk system containing the supernatatn from spun casein. 63 §§periment No. 18. Figure 18 illustrates the effects of NEH on heat-induced migration of copper to the cream fraction of a model system containing the supernatant from the spun casein used in the preceding experiment. The control model system (B) contain- ing no sulfhydryl group blocking agent failed to exhibit a copper adsorption peak in the cream at a flash heat treatment of 180 F. In this system the copper content of the cream increased with the temperature of flash heating. When 200 ppm HEM was added to the system prior to flash heating, the extent of heat-induced copper migration to the cream was notably reduced at all flash heat temperatures. 64 MODEL SYSTEM: Milk Dialyzafe, Washed Cream, Purified Casein(2-5°/e), 500% Added Cu / l, A-200p.p.m. Iodoacetamide Added Before 0 x Flash Heating. C g B-ConIrOII TOTAL Cu IN SYSTEM, 800IIg/I.) z u. — Cu IN CREAM FRACTION 3’3 0 \— Cu IN SKIMMILK FRACT|0N(by difference) A 3 _~ 280K A 1720 - \ L I- 3240* — —- . o _680'" .. — :D Z : ' > 9200.. ' M64033 5 I— e - 6 I.— v __ 2 < P .. .. 0: I60__ .00 0 IL _ " ‘c ' Ill 3 IZOE - . . 560\ DJ 2. B '— a: L o 80/ A 420% 2 K '3 3 - 40 FAT IN SYSTEM, 47 8%;SKIMMILK, 009% :480I- : f - x 0 I'— '—I Ch I I I I I I I I ‘440 80 I60 I70 I80 I90 200 TEMPERATURE, F FIGURE 19. The effect of Iodoacetamide on heat-induced migration of copper in a model system containing 2.5% purified casein. 55 Experiment No. 12. This experiment was performed to determine whether the addition of iodoacetamide would partially inhibit heat- induced copper migration in a model system containing sodium caseinate. In the control system (B) containing no sulfhydryl group blocking agent, the copper content of the cream fraction increased with temperature as previously reported (Experiment 15). However, when the system was treated with 200 ppm Iodo- acetamide prior to heat treatment (A), the extent of heat- induced copper migration to the cream diminished at all flash heat temperatures and the characteristic copper adsorption maximum appeared in the cream at 180 F.' The migration of cOpper to the cream was reduced by 43 per cent in the presence of iodoacetamide at this flash heating to 180 F. 1.!IIIIIIIII Ill-1| llll'.II llll‘ ' III III-I‘ll 66 MODEL SYSTEM: Milk Dialyzafe, Washed Cream, 500)“; Added Cu/l., A- 200p.p.m. N-thyl Maleimide Added Before Flash Heating. 8- Control.( TOTAL Cu, 550pg/I.) _Cu IN CREAM FRACTION _ Cu IN SKIMMILK FRACTIONIby difference) A\'\\ .\ 450 I — 4:. CD CD [D .b N L” TRIITTTIIITIVITIVIWT (N \I C” (N C” CD -IIIIIIIIIIIII CU IN CREAM FRACTION ,pg/I. 0F MILK IIII 8 >I‘IIW :10 MW ‘NOIIOVHH >i'IIININDIS NI no 225 200 I75 5: o 325 I25 1‘ 300 A j I 00 FAT IN SYSTEM, 469%; SKIMMILK, 009% : 275 I l I I I l I I '475 80 I60 I70 I80 I90 200 TEMPERATURE, F FIGURE 20. The effect of NEM on the heat-induced migration of copper in a model system devoid of skimmilk proteins. 5? gngpiment No. 20. Flash heat treatment of a model system comprised only of milk dialyzate, washed cream and added copper, (Figure 20, Curves B) induced maximum migration of cOpper to the cream in the range of 175 to 180 F as reported in an earlier ex- periment (Experiment 16). The presence of the sulfhydryl group blocking agent (A), NEM, not only reduced the adsorption of cOpper by the cream in the unheated system but also reduced the extent of cOpper migration to the cream by 75 per cent at the flash heat temp- erature of 175 F. This system demonstrates maximum copper adsorption by the cream at 185 F. In the control system (B) devoid of skimmilk proteins and sulfhydryl group blocking agent, 45 perlcent of the added copper migrated to the cream prior to heat treatment and 75 per cent migrated to the cream as a result of flash heating at 175 F. DISCUSSION Raw milk sampled from individual Holstein, Jersey and Brown Swiss cows in their third and fourth month of lactation contained residual cOpper ranging from 20 to 80,pg Cu/l (Figures 1, 2, 3, 4, 5, 7, 9, 10, 11, 12, and 13). The resid- ual copper content of fresh milk was much higher (186 and 220 ‘pg Cu/l) during the month following parturition (Figures 6 and 8).) This observation is in agreement with the findings of earlier investigators (43). Capper added to fresh raw milk was preferentially bound to the skimmilk proteins, as reported by previous workers (30. 40. 54. 55)- The observation that as little as 2-3 per cent of the added cOpper migrates to the cream fraction (Figures 1 and 2) agrees with data published by King et a1. (40). The affinity of the skimmilk proteins for added cOpper was substantiated by data obtained from experiments with model milk systems. In the absence of the skimmilk proteins, as much as 75 per cent of the added copper migrated to cream fraction of the unheated model milk system (Figures 16 and 20). No definite statement can be made regarding the distribution of natural cOpper in raw milk. The data obtained did suggest that natural cOpper was primarily adsorbed at the fat globule interface; however, the tremendous copper-binding surface 68 69 area presented by the small fat globules in the skimmilk (0.1 - 0.2% fat) may have distorted this Observation. The results of eXperiments involving fluid milk indi- cated that residual and added copper migrated from the skim- milk to the cream fraction as a result of heat treatment. Milk having a total copper content in the normal range of 40 to 240’pg Cu/l exhibited maximum and constant migration of copper to the cream following flashrnating (Figures 1, 2, 3, 4, and 13) or 10 minute holding periods at temperatures of 180 to 206 F (Figure 7). The initial heat-induced mi- gration of copperin milk having a normal copper content was detected at flash heat temperatures in excess of 140 F. An increase of the total copper content of the milk to that level known to catalyze lipid autoxidation, i.e. 500 - 700‘pg Cu/l, resulted in maximum migration of cOpper to the cream in milk flash heated to approximately 180 F (Figures 5, 6, 8, 9, 10, 11, 12, and 13) or following a 10 minute holding at 170 F (Figures 5 and 7). The author suggests that the difference in occurrence of maximum copper adsorption in the cream fraction may be explained by different rates of reaction at the two temperatures, and that the same adsorption phenomenon is occurring in both heat treatments. On the other hand, the flash heat temperature producing maximum migration of copper to the cream may be characteristic of milk produced by an individual cow. Heating fluid milk systems having a 70 total copper content in excess of 500 pg/l induced desorption of copper from the cream fraction at flash heat temperatures‘ greater than 180 F (Figures 6, 8, 9, 10, 11, 12 and 13) or 10 minute holding at temperatures greater than 170 F (Figures 5, 6 and 7). This phenomenon may be the result of heat de- naturation of the fat globule membrane protein followed by partial desorption and migration of this copperbbearing fraction into the aqueous phase. Data from the washed cream experiments indicated that heat-induced adsorbed copper is most tenaciously bound to the cream phase in milk heated to 180 F (Figures 10 and 11). In addition, the amount of adsorbed copper removed by washing the cream fractions was dependent upon.the temperature of the wash water; more copper was lost in wash water at 80 F than in water at 36 F. The fact that a greater portion of the adsorbed copper was removed by washing cream fractions of milk flash heated to temperatures greater than 180 F, agrees with the postulation that the tenacity with which the copper- bearing membrane protein adsorbed to the fat globule interface was altered by heat treatment. The extent of heat-induced copper migration to the cream appeared to be related to the degree of cOpper contamination (Figure 8) and the fat content of the milk. The level of total copper in the cream fraction was increased by storage at 34 F following heat treatment and prior to separation 71 (Figures 5 and 6). Slight migration of copper to the cream did take place when heated skimmilk was mixed with unheated cream. Moreover, the total copper content of the cream fraction significantly increased when copper was added to heated whole milk (Figure 7). The latter observations indi- cate that the greatest extent of cOpper migration occurred during heating, but time is an element in the attainment of copper equilibrium in the milk. Data obtained from heat treatment of model milk systems containing milk dialyzate, washed cream and centrifugally separated micellar casein or whey protein supernatant from micellar casein, indicate that the latter two milk protein fractions bound approximately the same amount of added copper (Figure 14). Flash heat treatment of the milk system con- taining micellar casein resulted in the characteristic copper adsorption maximum in the cream at 180 F (Figure 14 and 17); however, the cOpper content of the cream fraction separated from the milk system containing the whey protein fraction increased with temperature of flash heating (Figure 14 and 18). The irreversible adsorption of copper by the cream fraction of the latter system may be due to preferential adsorption of surface-active, heat-denatured, copper-bearing whey proteins, thus maskingthe characteristic copper adsorption peak in the cream at 180 F. The fact that whey proteins are more easily denatured by heat than micellar casein, and that maximum 72 copper adsorption was not evident in the cream phase of the system comprised of whey proteins, suggests that heat denatur- ed proteins are more strongly adsorbed at the fat globule interface than are more heat resistant proteins. Conversely, when purified betaalactoglobulin (BLG) was present as the only protein source in the aqueous fraction, maximum migration of copper to the cream was detected in the system flash heated to 175 F (Figure 16). Comparison of graphical data from this experiment with that for a system possessing all whey protein fractions (Figure 14) suggests that some copper bearing whey protein other than BLG exhibits more pronounced migration to the fat gldbule interface with increasing heat treatment. BLG is a heat sensitive protein and readily heat denatured. Possible heat-induced migration of this denatured protein fraction to the cream phase with subsequent competition for cOpper-binding sites may explain not only the less pronounced adsorption peak in the cream at 175 F, but also the marked increase in adsorbed copper in the cream of the control system which contained no skimmilk protein. Presumably, the added copper of the control system remained in the ionic state. Preparation of sodium caseinate results in a more dis- persed micellar structure. When this purified casein was present as the only protein source in the aqueous fraction of a model milk system, the level of copper in the cream 73 fraction once again increased with temperature of flash heat treatment (Figures 15 and 19). The postulation that protein denaturation is a prerequisite to heat-induced COpper migra- tion to the cream phase may not be applicable to these data. The sodium caseinate used in this milk system was undoubtedly more stable to heat than either the whey proteins or the mi- cellar calcium caseinpphosphate complex of normal cow's milk. The loss of the copper adsorption peak in the cream phase at 180 F may be explained by an increased in the pH of the milk system (above 7.0) resulting in a more polydisperse, surface-active caseinate system which migrates to and is strongly adsorbed at the fat globule interface during heat treatment of the solution. The different heat-induced cOpper migration patterns for supernatant whey protein, micellar casein, sodium casein- ate and BLG indicate that their individual copper-protein complexes may well have differenent reactivities and stability and the overall pattern exhibited in a given milk system would therefore be a function of not only temperature, time of heating and fat globule interfacial area, but also prOtein composition and quantity. Furthermore, the relative amount of copper adsorbed at the fat globule interface would be a function of the avidity of a particular protein for added copper and the molceular size of the copper-proteinate so formed. 74 The fact that the heat-induced cOpper adsorption-desorption pattern was observed in the cream fraction of a model milk system lacking skimmilk proteins (Figure 16 and 20) indicates that this phenomenon is characteristic of the fat fraction alone. The washed cream separated from this model system contained as much as 80 per cent of the toufl.copper content of the system following flash heating at approximately 180 F, or three to four times the copper content of the cream fraction of Whole fluid milk subjected to the same experimental con- ditions. This observation suggeststhat the fat globule mem- brane has a greater affinity for copper in the absence of milk protein and that copper-protein complexes, whatever they may be, compete for the copper binding sites at the fat globule interface as manifested through heat treatment. Preheating fluid milk to temperatures well above those employed in pasteurization of milk has been common practice for many years in the dry milk industry. Holm et a1. (33a) in 1926 showed that such heat treatment of fluid milk prior to drying improved the keeping quality of the dry whole milk. This observation was substantiated by many other research‘ workers in this field. Somewhat later, both Gould and Sommer (24) and Josephson and Doan (37) observed that heat treatments high enough to induce pronounced cooked flavors and liberate volatile sulfur compounds enhanced the resistance of such fluid milk to oxidation. 75 Oxidized flavor in cold milk does not appear until cooked flavor and sulfur odor have disappeared (or been masked). Concurrent with the production of these volatile "cooked" odors in milk is the "activation" or "liberation" of free sulfhydryl groups in milk which are capable of reacting with non-specific reagents such as sodium nitrOprusside and also with the more specific reagent thiamine disulfide. From the observation that activated sulfhydryl groups in heated milk slowly disappear on standing, as indicated by a diminishing nitrOprusside test and a decrease in titer of thiamine di- sulfide reducing substances (TDRS) and orthoiodosobenzoic acid reducing substances (IBRS), Stine (70) theorized that the heat liberated sulfhydryl groups presumably function as antioxidant in the liquid system through oxidation to some other end product. On the other hand, Harland et a1. (28a) observed that TDRS and IBRS are unchanged following storage of dry whole milk in air at 37 C. These data suggested that the reducing substances per as are not functioning as antioxidants. It might well be that the production of such activated groups Inerely accompanies some more significant change in the milk as it is heated. The relative ease of determining the sulf- hydryl group content may result in unwarranted importance being attached to this group as a potential antioxidant in dry whole milk. 76 Thomas (75a) attempted to elucidate the mechanism of sulfhydryl action in milk through the addition of the sulf- hydryl group blocking reagent p-chloromercuribenzoic acid (pCMB) to fluid milk stored at 40 C. He found that the addi- tion of pCMB to fluid milk accelerated the rate of lipid oxi- dation. The observation that the IBRS titer remained high in milk treated with this chemical lead to the investigator's suggestion that activated sulfhydryl groups function as anti- oxidants in fluid milk systems. Contrary to these findings, Stine (70) reported that the addition of pCMB, iodoacetamide (IODO) or N-ethylmalemide (NEM) often retarded the oxidation of lipid material in the dry whole milk to a variable, and occasionally dramatic extent. In addition, this investigator observed that the effect of the blocking agent in retarding the oxidative process was more pronounced in powders prepared from milks of low and inter- mediate preheating treatments than in those dry milks manu- factured from fluid milk which had been held at 190 F for ten minutes. Examination of the graphical representation of data ob- tained from experiments with whole fluid milk indicates that the greatest change in extent of heat-induced copper migration to the cream occurs during heat treatments from 165 to 175 F. Many investigators have reported sulfhydryl group activation in the non-casein.protein fraction as a result of heat treat- 77 ment in this temperature range (19, 24, 25, 37). In order to investigate the possibility that these activated sulfhydryl groups may be involved in the copper adsorption-desorption phenomenon experienced in the cream phase, milk was treated with sufficient HEM to react with sulfhydryl groups liberated during the course of heat treatment. The observation that the presence of NEM partially inhibited heat-induced copper migration to the cream suggeststhat the heat-activated sulf- hydryl groups were probably involved in the adsorption» desorption phenomenon (Figures 12 and 13). The addition of NEM prior to heat treatment of model milk systems containing micellar casein (Figure 17) or super- natant from spun micellar casein (Figure 18) markedly reduced the extent of cOpper adsorption by the cream as a result of flash heating to temperatures from 160 to 200 F. This obser- vation agrees with the theory postulated for whole milk systems, that heat activated sulfhydryl groups or heat induced volatile sulfides do play some role in heat-induced copper adsorption by the cream phase. An interesting observation was made in a model system containing 2.5 per cent sodium caseinate and iodoacetamide as the sulfhydryl groups blocking reagent. In this instance, the addition of iodoacetamide to the system prior to heating not only reduced the extent of migration of cOpper to the cream, but also resulted in the cOpper adsorption-desorption 78 peak in the cream which failed to appear in the control model system containing no sulfhydryl group blocking agent. Possi- bly the presence of this chemical inhibited the previously postulated heat activated migration of some cOpper-bearing fraction of the caseinate from the aqueous phase to the fat interface. (Figure 19). The addition of NEM to a model milk system free of skim- milk proteins reduced the extent of copper migration to the cream phase at all flash temperatures (Figure 20). moreover, it may be noted that the presence of NEM resulted in the greatest inhibition of copper migration to the cream when the milk system was flash heated at temperatures below 185 F3 Graphical data from this experiment suggest that NEH reacts at the level of the fat globule membrane through blockage of heat activated sulfhydryl groups at this location. In a milk system devoid of skimmilk proteins, no protein is available in the aqueous phase to complex with cOpper and possibly the majority of the added copper retains its ionic state, although no data are available to substantiate this. Assuming that heat liberated sulfhydryl groups in the membrane protein are chelation or adsorption sites for copper in this form, NEM may react with these loci rendering them unavailable for copper binding. The reaction of NEH with sulfhydryl groups in the membrane protein appears to be more pronounced in systems flash heated below 180 F (Figure 20). Above this temperature 79 possible heat induced desorption of denatured copper-bearing membrane material may be reflected in reduced inhibition of copper adsorption at the fat globule interface. The reaction of NEH with heat liberated sulfhydryl group, in the fat globule membrane may be manifested in still another way in systems containing skimmilk proteins. Available sulf- hydryl groups may be necessary as binding sites for copper proteinates that migrate to the fat globule interface as a result of heat treatment. This would attempt to explain the reduced level of copper adsorption by the cream during flash heat treatment of model systems containing skimmilk protein fractions in the presence of NEH (Figures 17 and 18). The sulfhydryl group blocking reagent iodoacetate appears to be effective in inhibiting adsorption of some unidentified cOpper - sodium caseinate complex by the cream when the model system was subjected to flash heating at temperatures greater than 180 F (Figure 19). One cannot disregard the possiblity that NEH causes partial inhibition of heat-induced copper migration to the cream of whole fluid milk by reacting with heat activated sulfhydryl groups in the fat globule membrane, thus making these sites unavailable for binding of complexed, surface- active cOpper proteinates as well as possible ionic copper freed from its protein complex as a result of heat treatment (Figure 13). 80 The author suggeststhat heat liberated sulfhydryl groups in fluid milk may serve in part as antioxidants, but more significantly as binding sites at the fat globule interface for copper in one form or another. Time appears to be a factor in the extent of reaction between heat activated sulfb hydryl groups and copper or copper proteinates as suggested by continued migration.of cOpper to the cream on storage of the heated milk accompanied by gradual reduction of available -SH groups (70). Possibly a copper proteinate adsorbs at the fat interface in a manner that not only provides a physical barrier against metal catalyzed lipid autoxidation but also chelates or associates with copper in a manner that destroys or retards its oxidative catalytic activity. SUMMARY AND CONCLUSIONS Data from experiments employing fresh milk from individual cows indicated that added copper is preferentially adsorbed by skimmilk proteins. The cream fraction of raw whole milk adsorbed as little as 2-3 per cent of added copper following storage at 34 F for 18 hours. When this milk was heated to temperatures greater than 140 F, the copper content of the cream phase increased by as much as 600 per cent. The extent of’migration of residual and added copper from the skimmilk to the cream as a result of heat treatment of the milk appears to be a function of the fat content of the milk, the total copper content and the temperature history of the milk. Milk having a total copper content in the normal range of 40 to 240‘ug Cu/l exhibited maximum and constant migration of copper to the cream phase following flash or 10 minute holding at 180-200 F. When the total copper content of the milk was increased to that level known to induce lipid autoxi- dation, i.e., SOC-700 pg Cu/l, a copper adsorption peak was observed in the cream phase following a flash heat treatment to approximately 180 F or following 10 minute holding at 170 F. The time-temperature treatment inducing maximum adsorption of copper by the cream appears to be related to rate of re- action and, or properties characteristic of milk from an individual cow. Heating fluid milk having a copper content greater than 81 82 SOO‘ug Cu/l to temperatures greater than 170 F for 10 minutes or momentary heating at 180 F, resulted in desorption of copper from the cream phase. Data from washed cream experiments indi- cate that copper is most tenaciously bound to the cream fraction in milk subjected to that heat treatment inducing maximum ad- sorption of copper by the cream. The author suggests that these observations may be explained by desorption of heat altered fat globule membrane material at elevated temperatures. The copper adsorption-desorption phenomenon exhibited by the cream may also occur in milk having a total cepper content of less than 220 pg Cu/l; however, the sensitivity of’the analytical proced- ure employed may be limited in detecting this characteristic. The fact that the amount of copper adsorbed by the cream fraction was increased by storage at 34 F following heat treat- ment, and that some migration of copper took place when copper was added to heated milk, suggests the following conclusions. Firstly, the greatest extent of migration of COpper to the cream takes place during the course of heat treatment, and secondly, time is a factor in the attainment of copper equili- brium in the milk. Possibly, the polymorphic crystallization of milk fat is closely related to the latter factor. Data observed from experiments with model milk systems indicate that added copper is bound primarily to the skimmilk proteins rather than the fat globule membrane. The heat- induced copper adsorptionpdesorption phenomenon experienced 83 in the cream appears to be characteristic of the cream fraction alone. Heat activated sulfhydryl groups seem to play some role in copper adsorption by the cream phase. The fact that the presence of a sulfhydryl group blocking agent partially inhibited copper migration to the cream in a system devoid of skimmilk proteins, suggests that copper is complexed in some fashion by the heat activated sulfhydryl groups in the fat globule membrane as a result of heat treatment. The data presented do not offer a direct explanation as to why the heat treatment of milk to temperatures greater than 165-170 F confers antioxygenic preperties to the milk. Indirect evidence does indicate that the copper-protein com- plexes, whatever they may be, are altered in some manner as indicated by the adsorptionpdesorption phenomenon described in this thesis. Possibly some cOpper-protein complex(es) adsorbs at the fat globule interface in a manner that presents not only a physical barrier against metal catalyzed autoxida- tion, but also sequesters or associates with copper in a mode that inhibits its ability to breakdown hydroperoxides to free radicals. 10. BIBLIOGRAPHY Allen, J. E. 1950. The oxidation of ascorbic acid in dairy products. J. Dairy Res., 21:54. Anson, M. L. 1939. The reactions of denatured egg albumin with ferricyanide. J. Gen. Physiol., 22:247. Beck, G. H., C. H. Whitnah and W. H. Martin. 1939. Relation of Vitamin C, Lecithin and Carotene of milk to thg development of oxidized flavor. J. Dairy Sci., 22:16 . Bernhardt, F. W. and Elizabeth Linden. 1950. The effect of heat treatment on pro-oxidant activity of copper in milk. J. Dairy Sci., 22:166. Brown, W. C., L. M. Thurston and R. B. Dustman. 1936. Oxidized flavor in milk. III. The time of copper contamination during production and processing, and aeration vs. no aeration as related to oxidized flavor development. J. Dairy Sci., 12:753. Brown, W. C., L. M. Thurston and R. B. Dustman. 1937. IV. Studies of the relation of the feed of the cow to oxidized flavor. J. Dairy Sci., 29:133. Brown, W. C., A. H. Vanlandingham and C. E. Weakley. 1939. Oxidized flavor in milk. VII. Studies of the effect of Carotene and Ascorbic acid in the feed of the cow on the susceptibility of the milk to metal-induced oxidized flavor. J. Dairy Sci., 22:349. Brown, W. C. and L. M. Thurston. 1940. A review of oxidation in milk and milk products as related to flavor. J. Dairy Sci., 22:629. Brown, W. C. and F. C. Olson. 1942. Oxidized flavor in milk. XII. Further studies of ascorbic acid mechan- isms in the production of oxidized flavor in milk. J. Dairy Sci., 22:1041. .Carter, H. E. 1949. Biochemical Preparations. Vol. 1, P. 32. John Wiley and Sons, Inc., New York. Conn, L. W., A. H. Johnson, H. A. Trebler and V. Karpenko. 1935. Determination of minute ammmflnsof copper in milk. Ind. Eng. Chem., Anal. Ed., 1:15. Day, E. A. 1960. Autoxidation of milk lipids. J. Dairy Sci., 52:1360. 84 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. Dahlberg, A. C. and D. C. Carpenter. 1936. The influence of method of sterilizing equipment upon the development of oxidized flavor in milk. J. Dairy Sci., 22:541. Davis, G. K. 1950. Unpublished results cited in a "Symposium on COpper Metabolism" P. 215. John HOpkins Press, Baltimore. Davies, W. L. 1932-1933. The mode of combination and distribution of heavy metals in dairy products. J. Dairy Res., 4:255. Diehl, H. 1937. The chelate rings. Chem. Rev., 22:39. Digeser, A. 1939. A study of the occurrence and preven- tion of Oily-metallic off-flavor in milk. Molkerei-Z (Hildesheim), 22:407. (Cited from Chem. Abstr., 22:5077. Dills, W. L. and J. M. Nelson. 1942. Isolation of a COpper bearing protein from cow's milk. J. Am. Chem. SOC. , git:16160 Drabkin, D. L. 1939. Report on copper. J. Assoc. Offic. Agr. Chem., 22:320. Elvehjem, C. A., H. Steenbock and E. B. Hart. 1929. The effect of diet on the c0pper content of milk. J. Biol. Chem., _82:27. Forster, T. L. and H. H. Sommer. 1951. Manganese, trypsin, milk proteins and the susceptibility of milk to oxidized flavor development. J. Dairy Sci., 22:992. Garret, 0. F. 1941. Some factors affecting the stability of certain milk prOperties. V. Interrelation of certain metals and metallic ions and the deve10pment of oxidized flavor in milk. J. Dairy Sci., 2fl:103. Gebhardt, H. T. and H. H. Sommer. 1931. I. The solubil- 12y406 cOpper under various conditions. J. Dairy Sci., 1 : l . Gehrke, C. W., C. V. Runyon and E. E. Pickett. 1954. A quantitative spectrographic method for the determina- tion 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., 21:1401. 23. 24. 25. 26. 27. 28. 28a. 29s 30. 31. 32. 33. 86 Gjessing, E. C. and Ga M. Trout. 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. Jo Dillry Sci., _2-2:373e Gould, I. A. and H. H. Sommer. 1939. Effect of heat on milk with especial reference to the cooked flavor. Hich. Agr. Expt. Sta. Tech. Bull., 223:48. Gould, I. A. 1941. Cooked flavor in milk, a study of its cause and prevention. Int. Assoc. Milk Dealers, Assoc. MIC. Q:553. Gould, I. A. and F. C. Ewbank. 1943. Oxidation of butter oil as influenced by previous heat treatment of the oil, butter, or cream. J. Dairy Sci., 22:409. Greenbank, G. R. 1948. The oxidized flavor in milk and dairy products. J. Dairy Sci., 22:913. Guthrie, E. S. and V. N. Krukovsky. 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., 22:786. Harland, H. A., S. T. Coulter and R. Jenness. 1952. The interrelationships of processing treatments and oxida- tionpreduction systems as factors affecting the keeping quality of dry whole milk. J. Dairy Sci., 22:643. Hartman, G. H. and 0. F. Garrett. 1943. Some factors affecting the stability of certain milk properties. Je D‘iry 801e, 22:337e Hartman, S. 1944. De oxydative virkningen i smor og maelk. Denmark Statens Forcgsmerjeri Beretning. (Cited from J. Dairy Sci., 32:780. 1959) Herder, P. C. den and B. M. Krol. 1950. Routine deter- minations of cOpper, iron and manganese in butter. Neth. Milk and Dairy J. ‘fi:42. Hotrick, J. H. and P. H. Tracy. 1945. Determination of copper in milk powder by a direct carbamate method. J. Milk Technol.,'§:5. Hollander, H. A. and P. H. Tracy. 1942. The relation of the use of certain antioxidants and methods of process- ing to the keeping quality of powdered whole milk. 333- 34. 35- 36. 37. 38. 39- 40. 41. 42. 43. 44. 45. 87 Holm, G. E., G. R. Greenbank and E. F. Deysher. 1926. Results of preliminary experiments upon the effect of separating, or clarifying and pasteurizing of a milk on the keeping quality of its milk powder. J. Dairy Sci., 2:512. Holman, R. T. 1954. Autoxidation of fats and related substances. Progress in the chemistry of fats and other lipids. Vol. 2. Academic Press, Inc., New York. Hubbard, D. M. and E. C. Spettel. 1953. Microdetermination of cOpper in biological material. Improved dithizone- polarographic method. Analyt. Chem., 22:1245. Jenness, R. and S. Patton. 1959. Principles of Dairy Chemistry. Ed. 1, P. 168. John Wiley and Sons, Inc., New York. , Josephson, D. V. and F. J. Dean. 1939. Observations on cooked flavor in milk - its source and significance. Milk Dealer, 22(2):35. King, R. L. 1958. Variation and distribution of copper in milk in relation to oxidized flavor. Ph.D. thesis, University of California, Davis. .King, R. L. and W. L. Dunkley. 1959. Relation of natural copper in milk to incidence of spontaneous oxidized flavor. J. Dairy Sci., 22:420. King, R. L., J. R. Luick, I. I. Litman, W. G. Jennings and W. L} Dunkley. 1959. Distribution of natural and added cOpper in iron in milk. J. Dairy Sci., 22:780. King, R. L. and W. F. Williams. 1963. Cepper distribu- Zéon in milk during early lactation. J. Dairy Sci., :11. King, R. L. 1963. Oxidation of milk fat globule membrane material. II. Relation of ascorbic acid and membrane concentrations. J. Dairy Sci., 22:267. KOppejan, C. A. and H. Mulder. 1953. The copper contents of milk. Proc. 13th. Intern. Dairy Congress, 2:1400. Krol, B. M. and P. C. den Herder. 1955. A routine method for the determination of copper in milk and dried milk. Neth. Milk and Dairy J. 2:56. Kruisheer, C. I. 1952. Defects in flavor of liquid milk gauged by oxidative processes. Neth. Milk and Dairy J. :2 2. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 88 Krukovsky, V. N. and E. S. Guthrie. 1945. Ascorbic acid oxidation, a key factor in the inhibition or promotion of the tallowy flavor in milk. J. Dairy Sci., 22:565. Krukovsky, V. N. and E. S. Guthrie. 1946. Vitamin 0, hydrogen peroxide, cOpper and the tallowy flavor in milk. J. Dairy Sci., 22:293. Krukovsky, V. N., J. K. Loosli and F. Whiting. 1949. The influence of tocopherols and cod liver oil on the stability of milk. J. Dairy Sci., 22:196. Krukovsky, V. N. 1952. The origin of oxidized flavors and factors responsible for their develOpment in milk and milk products. J. Dairy Sci., 22:21. Landers, J. W. and B. Zak. 1948. Determination of serum cOpper and iron in a single small sample. Am. J. Clin. Path., 22(6):590. Larson, P. B., I. A. Gould and G. M. Trout. 1941. Oxi- dation-reduction potentials and the oxidized flavor in homogenized milk. J. Dairy Sci., 22:789. Lea, C. H. 1953. Recent developments in the study of oxidative deterioration of lipids. Chem. Ind., 1952: 1303. Lusas, E. W., E. W. Bird and W. S. Rosenberger. 1956. The possibility of c0pper-induced oxidation of milk in szginless steel-white metal systems. J. Dairy Sci., 22:1 7. Monger, J. W. and H. Mulder. 1956. Natural and added copper in milk. Zuivelbereid en Handel, 22:528 (Cited from J. Dairy Sci., 22:146. 1959). Miller, D. E. and P. H. Tracy. 1952. The cOpper content of butter made by a continuous method. J. Dairy Sci., 22:292. Miscall, J., G. W. Cavanaugh and P. P. Carodemos. 1929. Copper in dairy products and its solution under var- iousagonditions. 11. Conclusions. J. Dairy Sci., 12:3 . Morris, 8. G. 1954. Fat rancidity. J. Agr. and Food Chem., 2:126. Morrison, 8. L. and Harriet L. Paige. 1946. Modified all-dithizone method for determination of traces of copper. Anal. Chem., 22:211. .. T ill... I I.) I'll"... I‘ll. lllll lllll .. 59- 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 89 Olson, F. C. and W. C. Brown. 1942. Oxidized flavor in milk. XI. Ascorbic acid, glutathione, and hydrogen peroxide as mechanisms for the production of oxidized flavor. J. Dairy Sci., 22:1027. Olson, F. C. and W. C. Brown. 1944. Oxidized flavor in milk. XIII. Studies of cupric complexes of ascorbic acid and certain amino acids and their possible relationp ship to oxidized flavor development in milk. J. Dairy Sci., 21:197. Perrin, D. R., F. R. Lightfoot and G. M. Moir. 1951. COpper and iron in dairy produce. Improved methods of analysis. J. Dairy Res., 22:77. Peterson, R. W. and J. H. Walton. 1943. The autoxidation of ascorbic acid. J. Am. Chem. Soc., 22:1212. Pont, E. G. 1952. Studies on the origin of oxidized flavor in whole milk. J. Dairy Res., 22:316. Pont, E. G. 1960. The chemistry of butter and butterfat. J. Dairy Res., 21:121. Rice, F. E. and J. Miscall. 1923. Copper in dairy pro- ducts and its solution in milk under various conditions. J. Dairy Sci., 2:261. Riel, R. R. and H. H. Sommer. Oxidized flavor in milk. I. Review of literature. L'Institute d'Oka, LaTrappe, Quebec, Canada. Roland, C. T., C. M. Sorenson and R. Whitaker. 1937. A study of oxidized flavor in commercial pasteurized milk. J. Dairy Sci., 22:213. Roland, C. T. and H. A. Trebler. 1937. The effect of fat content on oxidized flavor in milk and cream. J. Dairy Sci., 22:345. Smith, G. J. and W. L. Dunkley. 1962. The mechanism of spontaneous peroxidation in milk. J. Dairy Sci., 22: 170. Stine, C. M. 1958. A study of lipid oxidation in dry whole milk. Diss. Abs. 22:1391. Stotz, E., C. J. Harrer and C. G. King. 1937. A study of ascorbic acid in relation to copper. J. Biol. Chem. 119:511. II’ I I i I 1 72. 73- 74. 75. 753- 76. 77- 78. 79- 81. 90 Stull, J. W. 1953. The effect of light on activated flavor development and on the constituents of milk and its products. A review. J. Dairy Sci., 22:1153. Swanson, A. M. and H. H. Sommer. 1940. Oxidized flavor in milk. II. The relation of oxidation-reduction potentials.to its development. J. Dairy Sci., 22:597. Swartling, P. and S. Mattson. 1954. Determination of cOpper and iron in butter,milk and cream. Meddelande 42, Alnarp Institute, Sweden. Tarassuk, N. P. and J. Koops. 1960. Inhibition of oxi- dized flavor in homogenized milk as related to the concentration of copper and phospholipids per unit of fat globule surface. Thomas, Alan. 1954. A study of the mechanism of sulf- hydryl action in milk. Ph. D. thesis, University of Minnesota, Minneapolis. Thurston, L. M., W. C. Brown and R. B. Dustman. 1936. Oxidized flavor in milk. 11. The effects of homo- genization, agitation and freezing of milk on its subsequent susceptibility to oxidized flavor develop- ment. J. Dairy Sci., 22:671. Thurston, L. M. 1938. Theoretical aspects of the cause of oxidized flavor particularly from the lecithin angle. J. Dairy Sci., 22:256. Tobias, J. and E. O. Herreid. 1959. Determination of iegzagation products of ascorbic acid. J. Dairy Sci. 2: 2 . Tracy, P. H., R. J. Ramsey and H. A. Ruehe. 1933. Certain biological factors related to tallowiness in milk and cream. Ill. Agr. Expt. Sta. Bull., 389. Trout, G. M. and E. C. Gjessing. 1939. Ascorbic acid and oxidized flavor in milk. I. Distribution of ascorbic acid and occurrence of oxidized flavor in commercial grade A raw, in pasteurized irradiated and gisgetlirized milk throughout the year. J. Dairy Sci., : 7 . Willard, H. S. and C. S. Gilbert. 1953. Influence of dairy equipment on copper content and intensity of oxidized flavor in milk, cream, skimmilk. Milk Dealer, 22(6):46. 82. 83. 84. 85. 86. 91 Williams, E. B. and L. H. Burgwald. 1941. The effects of the direct addition of Carotene and mixed Tocopherols on the develo ment of oxidized flavor in milk. J. Dairy Sci., 2_:539. Williams, W. 1931-1932. The determination of copper and iron in dairy products. J. Dairy Res., 2:93. Zak, B. and N. Ressler. 1956. .Simultaneous microdeter- mination of cogper and iron using mixed phenanthrolines. Anal. Chem., 2 :1158. Zak, B. and N. Ressler. 1958. Serum co per and iron on a single sample. Clinical Chem., 2(1 :43. Zak, B. 1958. Simple procedure for the single deter- mination of serum cOpper and iron. Clinica Chimica APPENDIX 1. Preparation of Glassware. To avoid extraneous cOpper contamination, all glassware used in sampling, heating, separation, freeze-drying, and cOpper determinations was prepared for use by complete immer- sion in concentrated nitric acid followed by a five fold rinsing in deionized, redistilled water having an electrical conductivity of less than 0.9/u mhos. II. Standard Copper Solution. 1 mg Cu/ml 0.5000 g electro- lytic sheet copper is dissolved in 20 m1 6 N nitric acid and evaporated almost to dryness. Several drOps of glacial acetic are added, and the solution is quantitatively trans- ferred to a 500 m1 volumetric flask and made to volume with redistilled water. (1) Working standard cOpper solution. 11mg Cu/ml. 10 m1 of the standard cOpper solution is diluted to 100 ml; after mixing thoroughly, 10 ml of this intermediate solution is diluted to exactly 1000 ml, yielding a solution containing 1,M8 Cu/ml. This solution should be prepared each time a standard curve is made or for controlled copper addition to milk systems, and should be used immediately after prepara- tion. Otherwise, losses may occur due to adsorption of copper the surfaces of glassware. III. Copper Determinations. (1) Preparation of Standard Curve. 92 93 100 m1 aliquots of thoroughly mixed fresh raw milk are treated with the working cOpper solution in amounts rang- ing from 2 to l2’ug Cu per 100 m1 milk. These prepared stan- dard milks are frozen, freeze-dried, ashed in platinum cruci- bles at 550 C for 10 hours,and the copper content of the re- sulting ash determined by the method that follows, using the residual copper in 100 ml of the original raw milk as an appropriate blank in the copper determinations. (ii) Reagents: a. Ammonium Citrate. 430 g dibasic ammonium citrate are dissolved in 300 ml redistilled water. 200 m1 concentrat- ed ammonium hydroxide are added to the citrate solution and the total volume made up to 1000 g. Copper contamination is removed by adding 10 mg sodium diethyl dithiocarbamate and extracting the solution with three 100 m1 portions of analyti- cal grade carbon tetrachloride. b. Redistilled ammonium hydroxide. c. 2 per cent aqueous solution of sodium diethyl- dithiocarbamate. This solution is prepared fresh and used immediately after preparation. d. Carbon tetrachloride, A. R. grade. e. l per cent ethanolic phenophthalein. f. 2-4 N Hydrochloric acid. (111) Procedure. a. The ashed freeze-dried milk or milk fractions 94 are removed quantitatively from the platinum crucibles with 10 m1 boiling 2.4 N HCl and transferred to 100 m1 extraction flasks fitted with glass stoppers. The remaining ashed material is rinsed from the platinum crucibles with two additional 5 m1 portions of boiling 2.4 N HCl. b. 15 ml of the ammonium citrate buffer is pipeted into the extraction flask, followed by one drop of 1% ethan- olic phenolphthalein solution. Redistilled ammonium hydrox- ide is added to the flask contents until a faint pink end- point is obtained (pH about 9). c. 2 ml of 2% aqueous sodium diethyldithiocarbamate solution is added to the flask. The flask is shaken vigor- ously for 2 minutes and allowed to stand for 15 minutes until the reaction has gone to completion. d. Exactly 10.0 ml of carbon tetrachloride is pipeted into the flask, the flask is shaken vigorously for 5 minutes and allowed to stand for 10 minutes for complete solvent separation. e. Approximately 4 m1 of the lower yellow solvent layer is removed by plunging a 4 m1 pipet through the upper aqueous layer. The optical density of the removed solvent is determined in a Beckman DU spectrophotometer at a wave length of 435 mu with a slit width of 0.06 mm. f. A blank is prepared for each group of samples using 20 m1 of 2.4 N HCl instead of the ash solution. 95 IV. Preparation of Purified Casein. 2(8a) (1) Five one-gallon portions of fresh skimmilk are treated with 0.5 N H01 until the pH of the milk is adjusted to and maintained at pH 4.6-4.8. The resulting precipitate is dissolved in 0.5 N NaOH and reprecipitated with 0.5 N HCl at a pH of 4.8. The supernatant liquid is decanted and the residue from the five individual batches combined into one volume. (ii) The residue is mechanically dispersed for 5 minutes in redistilled water and the dispersion filtered through several layers of cheese cloth. This washing procedure is repeated a total of four times. (iii) The fourth casein dispersion is filtered through Whatman No. 1 filter paper in a Buchner funnel at a suction pump. The collected residue is washed twice with 95% ethanol for 5 minutes. After each washing the ethanol is removed by vacuum filtration. (iv) The casein residue is redispersed in absolute ethanol and filtered under suction a total of three times. (v) The casein residue is washed under continuous agi- tation for a period of five minutes in petroleum ether and again filtered under auction. This procedure is repeated. (vi) The casein residue is finally washed four consecutive times in redistilled water followed by suction filtration. Following the fourth washing, the casein is put into colloidal dispersion with 0.5 N NaOH; the resulting sodium caseinate 96 sol has a final pH of 7.8. (vii) The sodium caseinate sol is transferred to a semi- permeable cellophane membrane and dialyzed for 24 hours against 10 gal of distilled water maintained at 34-36 F. The dialysis is repeated against two additional 10 gal. volumes of distilled water under the conditions described. (viii) The dialyzed material is carefully removed from the membranes, transferred to shallow procelain trays, frozen at ~10 F and freeze-dried. (ix) The freeze-dried sodium caseinate is stored in air- tight bottles under refrigeration. V. Preparation of Iodoacetamide. (la) (1) Materials: ' alpha-chloroacetamide (Eastman Kodak) Sodium iodide Acetone (ii) Procedure: . 50 g alpha-chloroacetamide and 80 g sodium iodide are dissolved in l l.acetone. The solution is allowed to stand in the dark at room temperature for five days. The precipitate (Sodium chloride) is removed by filtration and the acetone is removed by distillation under reduced pressure until the vapor temperature begins to rise rapidly. At this point, the flask and contents are quickly cooled in an ice water bath. The crystals of iodoacetamide are filtered off 97 in the cold, and are washed with ice cold acetone. The iodoacetamide is purified by dissolving it in acetone, removing most of the solvent by distillation and recrystalli- zation in the cold. The iodoacetamide is recrystallized a total of three times from acetone. The crysatls are dried in a current of air, and are dissolved in an equal weight of water by warming rapidly. The aqueous solution is cooled rapidly in ice water and filtered in the cold. The crystals are dried in a desiccat- or and the recrystallization from aqueous solution is repeated. Melting point of the crystals is approximately 95.5 C. Iodoacetamide is slowly converted to iodoacetic acid in neutral and alkaline solution and slowly liberates iodide in acid solution. The pure crystals are stored in an amber bottle to retard decomposition; if the product reddens with age, it is recrystallized before use. VI. Nitroprusside Test. The procedure described by Josephson and Doan (37) was employed. ' (i) Reagents: I Solid ammonium sulfate, A.R. Sodium nitroprusside solution, 4.5% Solution in distilled water, prepared with reagent grade 98 sodium nitroprusside. Concentrated (28%) ammonium hydroxide, A. R. (ii) Procedure: A 5 ml sample of milk is saturated by adding ex- cess solid ammonium sulfate in a test tube and shaking. Five drops of a 4.5% solution of sodium nitroprusside (freshly made) are introduced with agitation followed by 5 drops of concentrated ammonium hydroxide. After again shaking the tube contents, the relative color intensity is noted.