sLIBRARY Michigan State ”a Undermy "+95“D This is to certify that the thesis entitled CHEMICAL, PHYSICAL AND FUNCTIONAL PROPERTIES OF SELECTED MILK PROTEIN CO-PRECIPITATES presented by DAVID WARREN TOBELMANN has been accepted towards fulfillment of the requirements for M.S. Jegreein FOOD SCIENCE 0« Major professor Date August 15, 1979 07639 OVERDUE FINES: 25¢ per day per item RETURNING LIBRARY MATERIALS: Place in book return to remove charge from circulation records CHEMICAL, PHYSICAL AND FUNCTIONAL PROPERTIES OF SELECTED MILK PROTEIN CO-PRECIPITATES By David Warren Tobelmann A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Food Science and Human Nutrition 1979 ABSTRACT CHEMICAL, PHYSICAL AND FUNCTIONAL PROPERTIES OF SELECTED MILK PROTEIN CO-PRECIPITATES By David Warren Tobelmann Acid (ACP), pH 3.5 (P3C), hexametaphosphate (HMP) and calcium (CAC) co-precipitates were prepared from fresh skimmilk. Standard proximate, calcium and phosphorus analyses were performed on all samples. Functional properties examined were solubility, viscosity, water hydration capacity, emulsifying capacity and whippability. HMP exhibited greatest solubility, low viscosity and good water hydration. ACP and P3C displayed moderate solubilities, with ACP exhibiting higher viscosity and water hydration than P3C. CAC was extremely insoluble at neutral pH, and displayed low water hydration. None of the co- precipitates formed a stable foam. ACP, P3C. and CAC exhibited emulsi- fying capacities similar to sodium caseinate, whereas HMP was similar to nonfat dry milk. Co-precipitates were examined by electrophoresis in urea and sodium dodecyl sulfate (SDS). Urea had no effect on the insoluble fraction of the samples. 505 gels of P3C revealed B-lacto- globulin and arlactalbumin bands in addition to the casein bands. SDS alone failed to solubilize the insoluble fractions of ACP and CAC. To my mother and father-- who always encouraged me to strive for the best. ii ACKNOWLEDGEMENTS I wish to express my sincere gratitude and appreciation to Dr. J. R. Brunner for his patience, counsel and the many interesting discussions we had. I also appreciate his help in the preparation of this thesis. Thanks also go to Dr. C. M. Stine and Dr. R. C. Chandan of the Department of Food Science and Human Nutrition, and to Dr. H. A. Lillevik of the Department of Biochemistry for reviewing this manuscript and for serving on my graduate committee. I wish to acknowledge the fine technical expertise of Ursula Koch, and especially appreciate her help with the amino acid analyses. The interactions with the many graduate students who have passed through during the course of this study is also gratefully appreciated. I also thank Rose for her patience and understanding on those many occasions when I had to work. Finally, I gratefully acknowledge the financial assistance pro- vided by the Department of Food Science and Human Nutrition and the Charles Lukens Huston Fellowship Foundation. iii TABLE OF CONTENTS Page LIST OF TABLES .................................................. vii LIST OF FIGURES ................................................. viii INTRODUCTION .................................................... l REVIEW OF LITERATURE ............................................ 3 Co—precipitates ............................................. 7 Manufacture ............................................. 7 Compositional Properties ................................ l2 Nutritional Properties .................................. l4 Applications ............................................ l5 Functional Properties ....................................... l6 Solubility .............................................. l6 Whipping Properties ..................................... 20 Emulsifying Properties .................................. 22 Water Holding Capacity .................................. 26 Viscosity ............................................... 28 Effect of Heat on Milk Proteins ............................. 3l Casein .................................................. 31 Serum Proteins .......................................... 33 K-Casein and B-Lactoglobulin Interaction ................ 36 EXPERIMENTAL .................................................... 41 Chemicals and Materials ..................................... 4l Chemicals ............................................... 41 Commercial Products ..................................... 41 Milk .................................................... 4T Equipment ............................................... 42 iv Page Preparative Procedures ...................................... 43 Acid Co-precipitate ..................................... 43 pH 3.5 Co—precipitate ................................... 43 Hexametaphosphate Co-precipitate ........................ 43 Calcium Co-precipitate .................................. 43 Chemical Methods ............................................ 44 Nitrogen ................................................ 44 Fat ..................................................... 45 Moisture ................................................ 45 Lactose ................................................. 45 Ash ................................................ ..... 46 Calcium ................................................. 46 Phosphorus .............................................. 47 Amino Acids ............................................. 48 Methionine and Cystine .................................. 49 Tryptophan .............................................. 49 Sulfhydryl (SH) and Disulfide (SS) ...................... 50 Soluble Protein ......................................... 5l Functional Tests ............................................ 52 Solubility .............................................. 52 Water Hydration Capacity ................................ 53 Viscosity ............................................... 53 Emulsifying Capacity .................................... 54 Whipping ................................................ 54 Physical Methods ............................................ 55 Polyacrylamide Gel Electrophoresis (PAGE) ............... 55 SDS PAGE ................................................ 56 Staining ................................................ 57 Statistics .................................................. 57 RESULTS AND DISCUSSION .......................................... 58 Preparation of Co-precipitates .............................. 58 Composition ............................................. 60 Yields .................................................. 63 Functional Properties ....................................... 7O Solubility .............................................. 7O Viscosity ............................................... 73 Page Water Hydration Capacity ................................ 76 Emulsifying Capacity .................................... 79 Whipping ................................................ 80 Electrophoretic Analysis of Co-precipitates ................. 82 PAGE .................................................... 82 Urea PAGE ............................................... 84 SDS PAGE ................................................ 86 CONCLUSIONS ..................................................... 94 REFERENCES ...................................................... 95 APPENDIX ........................................................ llO vi LIST OF TABLES TABLE I. w \IO‘U'l-b II. Al. A2. A3. A4. Estimated production of milk protein products in the USA, 1977 ........................................................ . Approximate compositions of selected dry dairy products ..... . Optimum conditions for production of selected co- precipitates ................................................ Functional properties of proteins and applications .......... . Abbreviations fer products referred to in this study ........ . Proximate compositional analysis of co-precipitates ......... . Amino acid composition of co-precipitates and selected commercial products ......................................... Comparison of essential amino acids and chemical scores of selected milk protein products .............................. . Yield and whey protein content of co-precipitate samples.... 10. Foam stabilities (l/2 times) fer co-precipitates compared to other milk protein products .............................. Levels of SH and SS groups in unheated and heated milks at normal pH and at pH 3.5 ..................................... List of chemicals used in this study ........................ Preparation of buffers used in various experiments .......... Method fer determination of fat ........ . .................... Numerical values, standard deviations and levels of signifi- cance for various functional tests ...................... .... vii Page II I7 59 60 62 64 65 81 91 110 112 114 ITS FIGURE I. IO. LIST OF FIGURES Electrophoretic patterns of proteose-peptone specimen from skimmilk (A), and co-precipitate aqueous phases of ACP (B), P3C C), HMP (D) and CAC (E). Total acrylamide con- centration was 5% and 10% in the spacer and running gels, respectively .............................................. . SqubiIity profiIe at pH 3.0, 4.5 and 7.0. SWC(O), BWC (D), NAC (A), NDM (O) .................................. . Solubility profile of co-precipitates at pH 3.0, 4.5 and 7.0. ACP (O), P3C (0), MP (A), CAC (O) .............. . Effect of concentration on absolute viscosity of different milk protein samples. NAC (O), SWC (Cl), BWC (A), MON (0) .................................................. . Effect of concentration on absolute viscosity of differ- ent co-precipitates. ACP (CD), P3C ([3), HMP (AS), CAC (CI) .................................................. . Water hydration capacities of co-precipitates compared to other protein samples ..................................... . Emulsifying capacities of co-precipitates compared to other protein samples .................. . .................. . Specific volumes for co-precipitates compared to other protein products .......................................... . Electrophoretic patterns of whole casein (A), acid whey (B), ACP (C), P3C (D), HMP (E) and CAC (F). Total gel concentration was 5% and 10% in the spacer and running gels, respectively ........................................ Electrophoretic patterns of whole casein (A), acid whey (B), ACP (C), P3C (D), HMP (E) and CAC (F) in 7 M urea. Total gel concentration was 5% and l0% for the spacer and running gels, respectively ................................ viii Page 68 72 72 74 74 77 77 BI 85 FIGURE Page 11. Electrophoretic patterns of whole casein (A), acid whey (B), ACP (C), P3C (D), HMP (E) and CAC (F) in SDS. Total gel concentration was 10% ................................. 87 12. Electrophoretic patterns of whole casein (A), acid whey (B), ACP (C), P3C (0), HMP (E) and CAC (F) in $05 with 2-mercaptoethanol. Total gel concentration was 10% ....... 89 13. Standard curve for molecular weight determination. Total acrylamide concentration was 10% .......................... 90 ix INTRODUCTION Milk protein products represent an important source of ingredients to the food industry. These products offer nutritional enrichment, desirable functional properties and are highly accepted by the consumer. Typical products used extensively include nonfat dry milk, sodium caseinate and whey solids. Co-precipitates are unique in that they contain both casein and whey proteins in a concentrated form, making them of high nutritional quality. Because of their low lactose content, co-precipitates can pro- vide total milk protein in applications where lactose constitutes a problem, e.g., sandiness in ice cream, caking in dry mixes and excessive browning in baked goods. Furthermore, the biological oxygen demand of whey from co-precipitate manufacture is lower than that for most casein and cheese wheys, thereby providing less of a waste disposal problem. The demand for functional protein by the food industry continues to increase. While such milk protein products as nonfat dry milk, sodium caseinate and whey protein concentrate offer many important func- tional properties, little work has been conducted on the functionality of co—precipitates. Assessment of these properties could provide insight into new applications for these products. In this study several types of co-precipitates were prepared in order to assess their functional properties and compare these to nonfat dry milk, sodium caseinate and whey protein concentrate. Materials were blended in liquid model systems, and solubility, viscosity, water hydration capacity, emulsifying capacity and whipping properties were examined. In addition, the nature of the protein complement contained in co-precipitates was examined in polyacrylamide gels in several dissoci- ating systems. REVIEW OF LITERATURE Milk protein products represent an important and valuable source of protein ingredients to the food industry. The success of these products is due to their recognized nutritional, organoleptic and func- tional properties (Morr, 1979). As seen from Table 1, nonfat dry milk and dried whey are the major sources of milk protein ingredients cur- rently used in the food industry, although many new products are now available which can contribute special advantages for certain food applications. Table 2 reviews the composition of some selected dry milk protein products. Nonfat dry milk (NFDM) is the most popular source of milk protein (Hugunin and Ewing, 1977). In addition to imparting excellent flavor, functional properties and nutritional quality, NFDM is available in a convenient, storage-stable fbrm. NFDM can be produced by spray drying or roller drying, although the former is the most satisfactory method. NFDM is manufactured by heating centrifugally separated skimmilk to meet the desired heat treatment, pasteurizing, concentrating to 45-50% solids and drying. Dry whole milk and dry buttermilk are similar to NFDM except that their fat contents are higher. Muller (1971) reviewed the manufacture of caseins and caseinates, an industry heavily concentrated in New Zealand and Australia. Casein curd is produced from pasteurized skimmilk by treating with rennet, Table 1. Estimated production of milk protein products in the USA, 1977. Amount, Protein Content, Product 1000 metric tons % dry basis Nonfat dry milk 421 36 Casein/caseinates 66 95 Co-precipitates 5 90 Partly delactosed whey 14.5 20 Partly demineralized whey 11.8 13 Whey protein concentrates 3.6 50 Whey solids in dry whey 210 13 Source: Morr (1979) Table 2. Approximate compositions of selected dry dairy products. Moisture Protein Fat Lactose Ash Product % % % % % Dried Whole Milk 2.0 26.4 27.5 38.2 5.9 Nonfat Dry Milk 3.0‘ 35.9 0.8 52.3 8.0 Dried Buttermilk 2.8 34.4 5.3 50.0 7.6 Casein 7.0 88.0 1.0 -- 4.0 Caseinates 4.0 92.0 0.8 -- 1.5 Dried Whey 4.5 12.9 1.1 73.5 8.0 Whey Protein Concentrates 2.0 20-60 2-9 18-60 3-18 Co—precipitates 4.0 83.0 1.5 1.0 10.0 Source: Hugunin and Ewing (1977) direct acidification with acid (usually hydrochloric, sulphuric, or lactic), or by culturing with lactic—producing micro-organisms. Generally, a pH of 4.3-4.5 is achieved to precipitate the casein. At this pH a product with a lower ash content is produced than if higher pH values are used. The curd is easily separated from the whey frac- tion, washed and dried. Isoelectric caseins are insoluble and are use- ful only in applications where solubility is not needed. Preparation of Na, K, and Ca caseinates, which are simply salts of casein, is routinely achieved by neutralizing casein curd with the corresponding base to pH 6.8-7.5, pasteurizing, and spray drying the solubilized protein sol. Caseinates are excellent food ingredients, possessing good storage stability and imparting many unique functional properties. In 1976, over 34 billion pounds of whey (both sweet and acid), equivalent to 2.2 billion pounds of whey solids, were produced in the United States (Clark, 1979). However, only 56.3% of these solids were further processed. As seen in Table 1, dry whey solids represent the second most popular source of milk protein. This product can be used as a source of crude lactose, milk solids, milk proteins or total solids. Processing of whey involves centrifugal separation and clarification to remove fine casein particles and fat, pasteurization, concentration to 45-50% solids, a holding period to permit a portion of the lactose to crystallize and spray drying. To produce free-flowing powders of low hygroscopicity, a two-stage drying process is used to ensure that the lactose is in the alpha-hydrate form. Conventional lactalbumin represents a concentrated source of whey protein. It is produced by adjusting the pH of whey to 4.5-5.2 and heating to denature and precipitate 70-80% of the whey proteins (Robinson gt 21:9 1976). However, this results in a yellowish-brown, completely insoluble, gritty powder which greatly limits its use in the manufacture of foods. Recently, Modler and Emmons (1977) developed a novel approach for preparing a heat precipitated whey protein product. This involves heating whey at pH 2.5-3.5 at 90° C, then adjusting the pH to 4.5 to effect precipitation. The resulting product exhibits high solubility, but yields are only 35-53% of total protein in the whey. Over the last 20 years, there has been considerable interest in implementing processes for recovering whey protein concentrates which retain their native and functional state (Morr, 1979). Whey protein concentrates (WPC) contain higher than normal concentrations of whey proteins. These products can be manufactured by many different pro- cesses. Cold precipitation techniques involve the use of polyelectro- lytes, including carboxymethylcellulose (Hansen gt 31., 1971), ferri- polyphOSphate (Jones gt 31,, 1972), polyacrylic acid (Sternberg gt_§1,. 1976), and sodium hexametaphosphate (Richert, 1972). Other processes fbr the manufacture of WPC include electrodialysis (D'Souza gt_§l,, 1973), ultrafiltration/reverse osmosis (Fenton-May £5 21:: 1971), and gel filtration (Morr §t_§l,, 1967). These WPC have excellent solubility and diverse functional properties, although their compositions vary considerably depending upon the process used in their manufacture. Co-precipitates Procedures fer the precipitation of whey protein along with casein were developed in the 1950's and 1960's. Co-precipitates contain combinations of both caseins and whey protein, although the exact nature of this combination is not well understood (Beeby gt 11., 1971). According to Muller (1971) development of the commercial manufacture of co-precipitates was motivated by several considerations: (1) to increase the yield of protein recovery from milk; (2) to enhance the range of functional properties in foods; and (3) to increase the nutritional role of milk protein due to the contribution of the casein-whey protein combination. In addition, co-precipitates contain lower lactose levels when compared to NFDM. The initial developments in the production and uses of co-precipi- tates were carried out in the United States and Russia. However, the Australian Commonwealth Scientific and Industrial Research Organization (C.S.I.R.0.) was the first group to fully exploit the commercialization of co—precipitates (Southward and Goldman, 1975). Manufacture Much of the early information concerning co—precipitates is found in patent literature. Muller (1971) notes that in the 1950's, it was observed that protein precipitated by acid or CaCl2 from heated milk contained both casein and whey protein. However, Rowland (1937) had observed this years earlier when he found that 76% of the total soluble nitrogen was coagulated in heated milk when the denatured whey proteins were co-precipitated with casein at pH 4.7. Heat-treated co-precipitates are produced by heating skimmilk sufficiently to denature the whey proteins, which appear to complex with the caseins (Southward and Goldman, 1975), followed by the addition of some precipitating agent. Scott (1952) described a batch process for making a co-precipitate. Alkali was added to skimmilk to reduce the acidity so as to inhibit initial precipitation of the protein when heat- ing the milk to 90° C. Acid was added slowly to effect precipitation and the resulting curd was separated from the whey, washed and dried. Howard gt a1. (1954) used a similar batch method but without first adjusting the acidity of the milk. They also dispersed the product with alkali at pH 6.6-7.2 and dried it in that form. Scott (1958) dispersed the co-precipitate described in his earlier process at pH 6.6-6.7 with alkali, then add NH OH to pH 8.0-8.5 before heating at 77-85° C and 4 drying. During drying ammonia was evaporated and the pH of the powder was reduced to 6.6-7.3. Loewenstein (1961a) produced a soluble co- precipitate from Scott's product by mixing it with a diglyceride and lecithin and homogenizing the mixture in skimmilk. This powder had a protein content of 40-85% and improved solubility and dispersibility. In another patent, Loewenstein (1961b) described the manufacture of a co-precipitate with high water binding capacity, involving the addition of hydrocolloids such as alginates before drying. To obtain a product with low water binding capacity for use in baked goods, Loewenstein (1965) neutralized Scott's co-precipitate with 1% of lime-in-water to a pH of 7.0-7.4. Following this neutralization and the addition of a calcium salt, the dispersion was spray dried. Engel and Singleton (1968) produced a co-precipitate by heating skimmilk for 10 min at 82-99° C, concentrating to about 30% solids, applying a heat treatment of 138° C for 15 sec, cooling to 101° C and acidifying to produce a curd. In the early American methods fer co-precipitate manufacture, no provision was made for controlling the calcium content. In the U.S.S.R. D'yachenko gt 31. (1953) described the preparation of a co-precipitate containing 95% of the total skimmilk protein. This was accomplished by adding CaCl2 to milk heated to 85-95° C, yielding a co-precipitate with calcium and phosphorus contents of 2.65% and 1.40%, respectively. D'yachenko (1957) described a continuous process for the manufacture of this product. Calcium chloride was added to milk which was then heated to precipitate the protein. The curd was separated from the whey in a continuous, horizontal solid-bowl centrifuge. Arbatskaya gt a1. (1962) developed a two-stage heating system for the manufacture of co- precipitates. Skimmilk was first warmed in a heat exchanger by regener- ation with hot whey. Calcium chloride was injected and the mixture of milk and CaCl2 was subsequently heated to 95-97° C by steam injection. Rostrosa and D'yachenko (1968), while studying conditions for co-precipi- tation of milk proteins, concluded that the whey proteins and the casein-calcium phosphate complex coagulated jointly at 80-95° C at a CaCl2 concentration above 0.83 g/l. In another study, Rostrosa gt 31. (1968) demonstrated that when milk containing 1.25 g CaC12/l was heated to 80-95° C up to 50% of the added calcium was bound to the curd in low acid milk and about 25% in high acid milk. Australian work on co-precipitates began with Buchanan gt 31. (1965) when they modified the procedure of Arbatskaya et 21, (1962) to IO produce a calcium co-precipitate while reducing losses of curd particles during processing. Manufacture was carried out on commercial acid casein equipment. A combination of heat (90° C) and CaCl2 (0.24%) was used to precipitate the proteins. Following washing, the curd was dis- persed in 2% sodium tripolyphosphate and spray dried. Muller gt al. (1967) introduced the concept of producing a range of co-precipitates with different functional properties by varying the calcium content. Co-precipitates with a high (2.5-3.0%), medium (1.5%), and low (0.5-0.8%) calcium contents were made using a continuous process. These workers found that by varying the time of heating milk at 90° C, CaCl2 concentra- tion and pH of precipitation, different ranges of calcium content could be obtained. More recently, Southward and Aird (1978) conducted a thorough investigation into the optimum conditions necessary to produce a variety of co-precipitates. They conducted experiments to determine the effect of pH of skimmilk before heating, heat treatment of milk, quantity of acid or CaCl2 added and coagulation temperature on the characteristics and yield of curd. The results of this study are SURF marized in Table 3 on the following page. Polish workers have developed a new form of co-precipitate which was summarized by Mann (1976). Raw milk with added CaCl2 was separated, the skimmilk pasteurized at 90-92% C for a few seconds and the proteins precipitated at 40° C by the addition of HCl to pH 4.5. The curd was washed twice and dissolved with NaOH at 70-75° C to a pH to 6.6. Finally, after the addition of NH40H, the suspension was spray dried, yielding a product ranging in composition from 3.9-4.6% moisture, 0.20- 0.73% fat, 74.4-84.4% protein, 5.2-13.2% lactose, 4.6-5.9% ash, and II Table 3. Optimum conditions for production of selected co-precipitates. Condition High Ca Med. Ca Acid Initial pH of milk 6.4 6.4 6.4 Heat treatment 85 C/6 min 85 C/6min 85 C/6 min Coagulant CaCl2 sulphuric acid sulphuric acid Coagulation temperature 77° C 65° C 58-60° C pH of whey 5.8-5.9 5.1-5.5 4.8-5.1 Source: Southward and Aird (1978) 0.10-0.53% calcium. Production of various salts of this co-precipitate (sodium, calcium, ammonium and sodium-calcium) has been described by Chojnowski gt 31. (1975). Aird (1971) discussed the preparation of co-precipitates from combinations of whey with skimmilk or casein products. A patent was issued to Wakodo Company Ltd. (1969) in which whey was heated and mixed with sodium caseinate or skimmilk. After further heating, the mixture was acidified to pH 4.5-4.6, and the curd was dissolved in alkali and dried. A co-precipitate for use in baked foods (Netherlands, Bedrijven van het Nederlands Institut voor Zwivelonderzock, 1973) was prepared by mixing fOur parts whey with one part buttermilk or skimmilk. The pH was adjusted to 6.5-7.1, and the mixture was subsequently washed, mixed with polyphosphate and dried. Raeuber gt a1. (1977) produced co-precipitates using acid whey and skimmilk, or skimmilk, acid whey and blood serum. Co-precipitates can also be prepared from mixtures of milk and other proteins. Pokrovskii and Levyant (1968) produced a co-precipitate of milk and blood proteins using heat and calcium lactate. Sato and Ito 12 (1970) found that egg-white protein could be precipitated with casein from heated mixtures of egg-white and skimmilk. Using defatted soya flour and cottage cheese whey, Loewenstein and Paulray (1972) prepared a co-precipitate. The acidic (pH 4.7) mixture was heated for 30 min at 90° C and the precipitated curd was dried. Thompson (1977) co-precipi- tated rapeseed and cottage cheese whey proteins by heating at 95° C for 15 min at pH 4.6. Thompson (1978) prepared co-precipitates using cottage cheese whey and either soybean or cottonseed proteins, again using acid and heat treatment. The use of other precipitating agents to make co-precipitates has been cited by Beeby et a1. (1971) as an area requiring further research. Hansen (1966) found that some casein is precipitated from milk by carrageenin in the concentration range 0.01-0.10%. Several workers have studied the interactions of carboxymethylcellulose and milk proteins (Asano, 1969; Hansen gt 21,, 1971; Cluskey gt 11., 1969). Also, sodium hexametaphosphate has been used to precipitate whey protein at acid pH (Gordon, 1943; Hartman and Swanson, 1966; Hidalgo gt 31., 1972; Richert, 1972). These precipitating agents may also be feasible for the produc- tion of co-precipitates. Compositional Properties The composition of co-precipitates is affected by several factors. The most obvious of these, of course, is the method of production. Muller gt_a1, (1967) noted that the calcium content of a co-precipitate is determined mainly by pH of precipitation. As the pH is decreased the calcium content is reduced. The use of other precipitating agents will 13 also affect the composition of a co-precipitate (e.g., use of polyphos- phates will increase the phosphorus content). D'yanchenko gt 11. (1953) noted that co-precipitates with higher calcium contents also had corre- spondingly higher phosphorus levels. Southward and Aird (1978) found that the fat content increases as coagulating temperature and pH of precipitation are decreased. The extent of washing has also been shown to affect the composi- tion of co-precipitates (Buchanan gt_gl,, 1965). As the number of washes increased, fat, calcium, lactose and ash levels were shown to decrease. Southward and Aird (1978) noted a similar decrease in the lactose, calcium and ash levels with repeated washes, but found the fat content was unaffected. Neutralization and dispersion procedures will also affect composi- tional properties. Buchanan gt_gl, (1965) and Muller gt_al, (1967) dispersed co-precipitates in sodium tripolyphosphate at levels from 2-6% (w/w). This practice will increase the phosphorus content of the dried protein concentrate. Neutralization with lime will increase the calcium content, whereas use of other alkali (e.g., NaOH) will increase the levels of ash and the particular salt used. The exact nature of the proteins in co-precipitates has not been thoroughly studied. Southward and Aird (1978) conducted an electro- phoretic analysis of a high-calcium co-precipitate. Most of the caseins were accounted for, while only half the B-lactoglobulin, most of the bovine serum albumin, and none of the a-lactalbumin or inmunoglobulins were accounted for. The authors noted that 25% of the co-precipitated whey proteins could not be detected in the analysis. This discrepancy 14 was due to the insoluble protein which could not penetrate the gel. It should be pointed out that a greater proportion of whey protein is precipitated in a high-calcium co-precipitate than either the low- or medium-calcium forms. Nutritional Properties Casein has long been recognized as being a high quality protein, although the nutritional superiority of whey protein has been estab- lished (Mitchell and Block, 1946; Riggs gt_al,, 1955). Lohrey and Humphries (1976) measured PERs for several milk products and found that casein, co-precipitate, skimmilk powder, WPC and lactalbumin had values of 2.50, 2.78, 2.67, 3.03 and 2.92, respectively. The nutritional superiority of co-precipitates over casein is probably due to the higher concentration of sulfur-rich amino acids from the whey proteins. Southward and Goldman (1978) measured the PERs of high-, medium- and low- calcium co-precipitates at 2.91, 2.78 and 2.72, respectively. Examination of amino acid composition may also be useful in assess- ing nutritional value. Amino acid analyses of co-precipitates have been determined by Resmini gt_§l, (1971) and Lohrey and Humphries (1976). Resmini gt_al, claimed the biggest change in amino acid patterns from casein to co-precipitate was in the percentages of cysteine/cystine, alanine, proline and aspartic acid. Southward and Goldman (1975) though, point out that the analysis of the amino acids cysteine/cystine, which are low in both casein and co-precipitate, is subject to some error in determination. 15 Applications The use of co-precipitates in food products is well-documented in the literature. Co-precipitates combine a wide range of physical and functional properties with superior nutritive properties which allow fer tremendous potential as an ingredient in a variety of fOOd products. In baked cereal foods, co-precipitates can supplement lysine- deficient cereal proteins. A high-calcium co-precipitate was selected as the milk protein source for the Australian milk biscuit (Townsend and Buchanan, 1967) and for a milk biscuit "pre-mix" (Henderson and Buchanan, 1970). Some workers (Marston, 1971; Craig and Colmey, 1971) have indi- cated that co-precipitates have potential for use in the bread baking industry either for protein fortification or as a replacement for nonfat dry milk. Goldman (1973) found that insoluble and dispersible co- precipitates with relatively low farinograph water absorption produced doughs with better consistency than did soluble co-precipitates. Presently, sodium caseinate and NFDM are the main milk protein products added to meat systems. Thomas 35.31, (1973) evaluated the use of co-precipitates in sausages. Replacement of 40% of meat protein with high-calcium co-precipitate produced sausages that were not significantly different from controls in general acceptibility. Thomas §t_gl, (1976) produced a meat loaf type canned product using a low-calcium co-precipi- tate. The product resembled meat in chewiness, sliceability, appearance and flavor. Thomas gt_al, (1978) found that beef patties extended with 5% medium-calcium co-precipitate and 2.5% wheat flour had better appear- ance, flavor, texture and acceptibility than all-meat controls. 16 Co-precipitates have also found some applications in other dairy products. Modler (1974) notes that co-precipitates of casein and whey can be used to increase cheese yields. Thomas (1969, 1970) reported that the incorporation of co-precipitates in processed cheese prevented browning reactions, and improved the body and texture. Stavrova and Mochalva (1978) produced a spread-type dairy product using high-calcium insoluble co-precipitate. Functional Properties Functional properties are physico-chemical properties which give information on how a protein will behave in a food system (Hermansson, 1979). For protein ingredients, these properties can be useful in deter- mining fields of application. Table 4 summarizes some specific func- tional properties and applications for proteins. Functional properties are influenced by protein source, methods of isolation, precipitation and drying, concentration, modification, and environmental conditions like temperature, pH and ionic strength. Chou and Morr (1979) have noted that most functional properties are dependent upon the protein's conformation, solubility, and water-binding properties. Solubility Many protein functional properties such as solubility, viscosity, gelation, foaming and emulsification are directly related to the manner in which the protein interacts with water. According to Chou and Morr (1979), the most critical step in imparting a protein's functional prop- erty to a food system is its interaction with water to rehydrate, swell 17 Table 4. Functional properties of proteins and applications. Functional Property Food Application Emulsification coffee whitener, soups, meat products Stabilization soups, meats, desserts Fat absorption meat products Water absorption/retention meat products, bread, cakes Viscosity soups, gravies ' Gelation simulated meats, cheese Fiber/texture simulated meats Dough formation baked foods Adhesion/cohesion meat products, baked foods Aeration whipped toppings, chiffons, desserts Source: Morr (1979) and solubilize the protein. Mattil (1971) noted that studies of the functional properties of proteins can be made more efficient if one first investigates the solubility properties of the protein in different ionic environments. The solubilization of a protein molecule is a simultaneous process involving wetting, swelling, salvation and dissolu- tion, and is dependent upon conformation, pH, ionic strength, concentra- tion, temperature, mechanical disruptive fbrces and a number of other factors. A solubility profile over a range of pH values under standardized conditions can be used as a good guide to protein functionality. Although there are many terms used to describe solubility of food pro- teins (e.g., water-soluble protein, nitrogen solubility index) there are several basic steps in determining this property. Generally, the 18 protein is dispersed in water, followed by adjustment of the pH, centri- fugation, and determination of the nitrogen or protein content of the supernatant. Many workers have studied the solubility behavior of sodium caseinate and WPC. Hermansson and Akesson (1975a) examined the solubil- ity of WPC and sodium caseinate in water at room temperature and after heat treatments ranging from 70-100° C. Unheated samples of sodium caseinate and WPC had solubilities of 80.8 and 78.3%, respectively. Heat treatment had little effect on the solubility of either sample. Hermansson and Akesson (1975b) also studied the effects of NaCl on the solubilities of these two proteins. Concentrations of up to 1.0 M NaCl had little or no effect on the solubility of either sodium caseinate or WPC. However, when WPC was heated to 80° C in the presence of salt, solubility decreased. This was thought to occur from increased protein- protein interactions favored by the presence of salt. Morr gt 31, (1973) studied the solubility behavior of WPCs pre- pared by metaphosphate complex, electrodialysis, ultrafiltration, gel filtration, dialysis, CMC complex and iron complex. The solubilities for iron, CMC and metaphosphate complex WPC were highly dependent on pH. McDonough £5.31, (1974) studied the properties of WPC produced by ultra- filtration. They found that the solubility of this WPC remained constant over a variety of treatments, e.g., pasteurization, evaporation and spray drying. Sternberg gt_al, (1976) reported that the solubility of whey proteins isolated with polyacrylic acid was independent of pH over the range of 3.0 to 9.0. The solubility of this product at pH 6.5 decreased at concentrations greater than 10%. Modler and Emmons (1977) 19 prepared a WPC from sweet whey by adjusting the pH to 2.5-3.5, heating at 90° C for 15 min and adjusting the pH to 4.5. This product had a minimum solubility of 78%. Jost and Monti (1977) used a partial enzymatic hydrolysis by trypsin to increase the solubility of ultra- filtered WPC. Monti and Jost (1978) compared the efficiency of pancrea- tic trypsin, papain and bacterial protease for solubilization of heat- denatured whey, and found that trypsin worked best. Buchanan gt_al, (1965) noted that a high-calcium co-precipitate, unlike casein, would not dissolve or disperse at neutral pH without the addition of a calcium sequestering agent. Consequently they dispersed this product in 4% sodium tripolyphosphate (STPP). Smith and Snow (1968) examined the factors affecting aqueous dispersions of co-precipi- tates. They found that a low-calcium co-precipitate could be dispersed by pH adjustment alone. In order to dissolve a medium-calcium co-pre- cipitate at neutral pH, STPP was added at a 2-4% level to the mixture of product and alkali. High-calcium co-precipitate was dissolved at neutral pH by addition of 6% STPP. None of the co-precipitates were as soluble as sodium caseinate at pH 7, with the insoluble fraction con- sisting mainly of whey proteins. Southward and Aird (1978) examined the solubilities of several co-precipitates at pH 7 by stirring 3.2% solu- tions for 60 min at 60° C. Acid co-precipitate was found to be 95-99% soluble. The medium-calcium product was 70-90% soluble, but increased to 95-98% in the presence of 2% STPP. High-calcium co-precipitate was only 10-20% soluble, but increased to 90-95% with the addition of 6% STPP. 20 Whipping Properties Foaming or whipping pertains to the capacity to form stable foams with air. Food foams generally consist of air droplets dispersed in and enveloped by a liquid containing a soluble surfactant (Kinsella, 1976). In protein foams, the surfactant protein reduces surface tension and forms structural, cohesive filns around the air droplets. Generally whipping properties are measured as foam expansion, foam capacity or overrun. These terms relate to the volume increase of a protein disper- sion following the incorporation of air. Foam stability refers to the ability of a formed foam to retain its maximum volume over time. Webb (1941) made one of the first attempts to utilize milk protein foams by studying overrun and stability of whipped, reconstituted NFDM. High total solids, low temperature and addition of acid all increased foam stability. Also, high heat-treated powders produced superior foam capacity on reconstitution. El-Rafey and Richardson (1944) studied the foaming properties of the casein, lactoglobulin and lactalbumin frac- tions of milk protein. For the casein fraction, best foam stability was achieved at temperatures between 0 and 55° C and at fat levels below 0.15%. Foam stability increased with concentration to a maximum value, then decreased. The lactoglobulin fraction exhibited limited stability. Foam stability for the lactalbumin fraction increased with concentration to a maximum value, then leveled off. Optimum temperature was between 5 and 25° C. The whipping ability of whey proteins has been studied prolifically over the last ten years. Hansen and Black (1972) examined the whipping properties of whey protein-CMC complexes. Best foams were achieved at a 21 4% level of the complex in water. Heat treatment (75° C/15 min) proved detrimental to foam stability, but sugar addition was found to stabilize feams when added immediately after whipping. Jelen (1973) found that delactosed cheese whey produced poor overruns and stability. Heat treat- ment and higher total solids improved stability. Morr gt 91. (1973) investigated the functional properties of WPC (metaphosphate complex, electrodialysis, ultrafiltration, gel filtration, dialysis, CMC complex and iron complex). None of these products produced as high an overrun as sodium caseinate. The CMC complex WPC produced the most stable foam. Richert gt 31, (1974) studied the effect of heat treatment, pH, calcium concentration, redox potential and sodium lauryl sulphate upon the foaming properties of WPC. Heating temperature and redox potential affected overrun, while foam stability was affected by pH. Heat treat- ment of 70° C at pH 7.0 under oxidizing conditions gave the best overrun and most stable foam. Kuehler and Stine (1974) hydrolyzed WPC with pronase, prolase or pepsin. A limited amount of hydrolysis increased the foaming but stability was greatly decreased. This was thought to be due to an increased polypeptide contentwhich allowed air to be incor- porated. The polypeptides, though, did not have the strength to give a stable foam. McDonough gt_al, (1974) found that the total solids concentration was critical in ultrafiltered WPC foams, with 25% solids providing improved whippability. DeVilbiss gt 31. (1974) attempted to bake angel fbod cakes using WPC in place of egg white. The WPC did not function well, as the foam structure was not strong enough to retain loaf volume during baking. Morr (1979) related this failure in angel 22 food cake to lower sulfhydryl and disulfide content when compared to egg white protein. Little work has been reported on the whipping properties of co- precipitates. Goldman and Toepfer (1978) compared the overrun and foam stability of egg albumin, soluble high-calcium co-precipitate, sodium caseinate and WPC. Sodium caseinate and co-precipitate whipped alone and with sugar gave lower overruns than egg albumin, whereas WPC gave higher overruns. The most stable unsugared whips were fresh egg albumin and co-precipitate. In the presence of sugar all milk protein products exhibited maximum stability. Southward and Goldman (1978) found that soluble co-precipitates had whipping properties slightly better than sodium caseinate but not as good as those of egg albumin. Emulsifying‘Properties In food processing it is often necessary to blend edible fats and oils with various hydrophilic materials in an emulsion system (e.g., ice cream, salad dressings, gravies, meat emulsions). The use of many emulsifying agents is restricted due to food hygiene and food legisla- tion requirements. Consequently the ability of ingredient proteins to form and stabilize oil/water emulsions in food systems is extremely important (Cante gt al,, 1979). The formation of an emulsion proceeds as follows. Oil is split into minute droplets (e.g., by a mixer, homogenizer, etc.) in the bulk phase (usually water). Protein molecules then diffuse from the bulk phase to the oil/water interface where adsorption occurs. This leads to a lowering of the interfacial tension and hence lowers the mechanical 23 energy required to produce a given emulsion particle size. The particle size of the oil droplets depends on the splitting mechanism or process, as shown by Tornberg and Hermansson (1977). After emulsion formation, the protein film at the interface serves to stabilize the emulsion by retarding coalescence of the oil droplets. Emulsion capacity and emulsion stability measurements are often used in functional characterizations of proteins. Emulsifying capacity denotes the maximum amount of oil that is emulsified under specified conditions by a standard amount of protein. Emulsion stability relates the ability of an emulsion with a certain composition to remain unchanged. A brief description of factors involved in the measurement of emulsion capacity (EC) follows. The usual procedure for determining EC involves addition of oil to a protein dispersion until the emulsion inverts to a water/oil emulsion. The registering of the inversion point, however, varies from visual appearance of the viscosity decrease (Swift gt 31., 1961) to change in the amperage required to drive the mixer (Crenwelge gt 21., 1970) to change in electrical resistance of the emulsion (Webb gt 31., 1970). It is not surprising to find variability in EC data among different laboratories. Saffle (1968) has noted that factors such as equipment design, container slope, rpm of blending, rate of oil addition, temperature, pH, protein source, type of oil used and ionic strength all affect EC determinations. It is also worthwhile to note that the relationship between EC and the amount of protein required to produce a satisfactory emulsion is unclear since most emulsions 24 contain an amount of oil less than that required fer phase inversion (Pearce and Kinsella, 1978). Richert (1972) found that a WPC provided better emulsion stability than either sodium caseinate or egg albumin when 2.5% dispersions were mixed with corn oil. The WPC emulsion was stable for more than two weeks. Morr et 31. (1973) studied the EC of seven different WPC. Those prepared by metaphosphate complex, electrodialysis, ultrafiltra- tion, gel filtration, dialysis and iron complex had similar EC values. A CMC complex WPC had an EC value about double the others. Kuehler and Stine (1974) found that the EC of WPC decreased upon enzymatic hydroly- sis. Tornberg (1978b) measured the creaming stability of several pro- tein stabilized emulsions and found WPC to be a better stabilizer than sodium caseinate. Pearce and Kinsella (1978) examined the emulsifying activity index for whey protein powder and several isolated milk pro- teins using a turbidimetric technique. This index has units of area of interface stabilized per unit weight of protein. Bovine serum albumin demonstrated the best emulsifying activity, followed by sodium caseinate, B-lactoglobulin and whey protein powder. Cante and Moreno (1975) patented an edible whippable emulsion which contained the proteose- peptone fraction of cows' milk as the proteinaceous emulsifier. The authors felt that the proteose-peptone moieties were linear with hydro- philic and hydrophobic groups distributed along the length of the mole- cules. Thus, the protein could fit itself to the curve of the oil droplets in the emulsion. Pearson gt a1. (1965) studied the EC of potassium caseinate and NFDM. Potassium caseinate was most effective as an emulsified at pH10.5 25 and low ionic strength. The EC of NFDM was greatest at pH 5.6 regard- less of ionic strength. Protein solubility was found to be closely related to EC. NFDM emulsions were most stable at pH 5.4 and ionic strength of 0.3 while potassium caseinate emulsions were quite stable at all pH values at ionic strength 0.3. Sabharwal and Vakaleris (1972) studied the emulsifying properties of sodium caseinate in oil/water systems containing 4% coconut fat both with and without other emulsi- fiers. In the absence of emulsifiers, emulsion stability increased up to 0.4% sodium caseinate, then decreased. With 0.5% emulsifier (HLB 11) added, stability increased up to 0.5% sodium caseinate, then decreased. Smith and Dairiki (1974) studied emulsion stability with sodium caseinate and NFDM. In 25% milk fat emulsions, sodium caseinate alone acted as an emulsifier with optimum activity at 0.5%, while the NFDM gave best emulsion stability at a 6% level. Emulsifying properties of co-precipitates have been studied to some extent. Thomas gt 21, (1974) examined the emulsifying properties of calcium co-precipitate in a model meat system. Three types of cal- cium co-precipitate improved the emulsifying capacity of the system when replacing 20% of the meat at pH 5.6. Emulsion stability was improved when 20% of the meat was replaced with a soluble high-calcium co-precipi- tate at pH 5.6. Chojnowski gt_al, (1975) feund that calcium co-precipi- tate and milk protein concentrate (co-precipitate not neutralized) showed good emulsion stabilizing properties. The emulsifying capacity of sodium caseinate, and high-, medium- and low-calcium co-precipitates were investigated by Gronostaiskaya and Kholvdova (1978). Sodium caseinate possessed a greater EC than the co-precipitates. When used as 26 emulsifiers with sunflower seed oil, emulsions were produced with fat phase content at protein concentration of 80% at 0.2%, 80% at 0.25-0.30%, 79% at 0.4% and 74% at 0.94% for sodium caseinate, low-, medium- and high-calcium co-precipitates, respectively. Southward and Goldman (1978) noted that high-, medium- and low-calcium co-precipitates had emulsion stabilizing properties similar to sodium caseinate. Water HoldingTCapacity The ability of proteins to bind and immobilize water is one of the most important functional properties in food applications (Chou and Morr, 1979). Water holding capacity is a quantitative measure of the amount of water retained within a protein matrix under certain defined conditions, and usually includes entrapped water. Other terms fre- quently seen in the literature are water adsorption, which refers to the water adsorbed by a dry protein powder after equilibration against water vapor of a known relative humidity, and swelling, which refers to the spontaneous uptake of water. The degree of water binding in a food system is much more complex, and is influenced by pH, ionicity, tempera- ture and protein concentration. The mechanism of water binding has been reviewed by Kinsella (1976) as it applies to pure proteins. Pure proteins sorb water by bind- ing at specific hydrophilic sites (0H, NHZ’ COOH, C=O) at low water activities, followed by multimolecular adsorption and clustering at higher activities. These polar sites bind an average of four to eight water molecules. Other factors can affect the water binding in food systems. For instance, salts can have a dramatic effect. Bull and 27 Breese (1970) found that monovalent cations decreased the amount of water bound to egg albumin in the order Li+> Na+> K+> Rb+>Cs+. The water adsorption of several pure milk proteins has been measured. Bull (1944) measured the amount of water sorbed by 1yophil- ized B-lactoglobulin and bovine serum albumin to be 28.25 g water/100 g dry protein and 28.70 g water/100 g dry protein, respectively. Ruegg gt_al, (1974) found the water binding of casein to be 0.435 g water/g dry mass. Berlin gt_al, (1970) used differential scanning calorimetry to estimate the extent of water binding of casein, bovine serum albumin and B-lactoglobulin at 0.553, 0.498 and 0.553 9 water bound/g protein. Hagenmaier (1972) measured the water adsorption of casein, and deter- mined this to be 21.6 g water/l6 g N. Water adsorption was independent of pH. Pure proteins (those free of salts, fat, carbohydrates) are generally not used in food systems, and hence the water bound by protein concentrates is more useful to know. Berlin £3 31, (1973), while study- ing the water adsorption properties of WPC, noticed that products con- taining more low molecular weight materials (e.g., lactose, salts) bind more water. The authors also noted that thermal denaturation did not significantly change water adsorption. McDonough st 31, (1974) studied the water affinity of ultrafiltered WPC. Degree of entrapped water was measured by adding the protein to skimmilk and heating at 85° C for 5 min. WPC gave better entrapment of water than egg albumin via a classi- cal gel structure. Delaney (1976) measured the water absorptive capaci- ties of casein, sodium caseinate, NFDM and ultrafiltered WPC to be 68%, 250%, 70% and 50%, respectively, by a Farinograph method. The water 28 uptake of ultrafiltered WPC was measured by de Wit and de Boer (1975) and was found to increase after heat treatment to about twice the powder weight. Sternberg gt_al, (1976) compared the water hydration capacities of coagulated WPC (isolated with polyacrylic acid) and egg white, and feund that the WPC gave slightly lower values. Quinn and Paton (1979) examined the water hydration capacity of a number of protein concen- trates. Sodium caseinate and WPC had values of 2.33 and 0.97 ml water/g sample, respectively. Hermansson and Akesson (1975a, 1975b) investi- gated the effects of heat and salt on swelling properties of sodium caseinate and WPC. In this study, swelling was defined as the ability to absorb water in amounts smaller than those producing low viscosity dispersions. The swelling ability of casein decreased slightly with increasing temperature, whereas the swelling ability of WPC increased. The WPC swelled to a lesser extent than sodium caseinate, though. Swelling was also measured at ionic strengths ranging from 0 to 1.0 at 25° and 80° C. Sodium caseinate showed a decrease in swelling ability with increasing ionic strength at the lower temperature, while swelling increased slightly with ionic strength at 80° C. WPC showed no change in swelling with ionic strength at 25° C, but showed a slight decrease at 80° C. Viscosity Viscosity is a flow property and often reflects changes in hydro- dynamic properties due to absorption of water. Knowledge of the viscos- ity of protein dispersions is of practical importance in relation to processing, new product development, designing of quality control tests, 29 and mouth-feel and physical appearance. Flow properties are governed by the molecular size, shape, charge, solubility and swelling capacity of proteins. These are affected by temperature, concentration, pH, ionic strength and previous processing history (Hermansson, 1975). A knowledge of the solubility and swelling properties of a protein enables one to predict viscosity. Highly soluble, nonswelling proteins (e.g., albumins, globulins) have low viscosity. Soluble proteins with high initial swelling (e.g., caseinate) show a concentration dependent viscosity, reflecting the amount of swelled, but not fully solvated particles. Proteins with high limited swelling capacity show a rela- tively high viscosity at low concentrations (Hermansson and Akesson, 1975a). Dolby (1961) studied the viscosity of lactic casein and found that the pH of coagulation was important. The viscosity index increased as the precipitation pH increased due to increase retention of calcium in the casein. Washing temperature and drying time also affected viscosity slightly. Hayes and Muller (1961) investigated viscosity-pH relation- ships of acid casein in various alkalis. Confirmation of the logarithmic relationship between concentration and viscosity was made. Heated casein showed increased viscosity. The authors also confirmed the role of cal- cium in determining viscosity. In casein solutions near neutral pH, increased calcium had only a minor effect on viscosity, but at values greater than 8, a sharp rise in viscosity occurred with increases in calcium. Towler and Dolby (1970) studied the effects of precipitation pH, temperature of precipitation, acidulation time and wash conditions on viscosity of casein. High precipitation pH and temperature, short 3O acidulation time and washing in CaCl2 solution favored a high viscosity casein product. These conditions also favored high calcium retention in the curd. Richert (1972) produced WPC by precipitation with sodium hexameta- phosphate followed by neutralization with Ca(0H)2 to remove the phos- phate. Viscosity was extremely low for 2.5% solutions. However, a 15% WPC solution had a higher viscosity than a 15% egg albumin solution. The viscosity of 10% protein solutions of ultrafiltered WPC at different pH values and different heat treatments was studied by de Wit and de Boer (1975). At pH 6.7, the viscosity was constant up to 70° C, but showed a sharp increase at higher temperatures. At high and low pH values, the viscosity measurements showed sharp increases at lower temperatures. Hermansson (1975) examined the flow properties of casein- ate and WPC solutions as affected by concentration, ionic strength and pH. At concentrations below 12%, caseinate exhibited Newtonian flow characteristics, whereas higher concentrations yielded pseudoplastic behavior. Addition of salt caused an increase in viscosity, although no changes in swelling or solubility occurred. Caseinate had a very complex, pH-dependent viscosity which was maximum at pH 9.8-l0.0. In contrast, pH and ionic strength had little effect on the viscosity of WPC. WPC exhibited Newtonian flow behavior at concentrations below 12%. Viscosities of co-precipitates were studied by Hayes gt_al, (1969). A low-calcium co-precipitate exhibited non-Newtonian flow, and viscosi- ties were somewhat higher than casein. With added calcium, Viscosities for the co-precipitates increased at pH values above 7.0. This response to added calcium was different than that observed with casein, and was 31 attributed to the influence of protein-protein linkages formed during the manufacture of co-precipitates. Chojnowski gt_al, (1975) found that viscosity varied for milk co-precipitates neutralized by different salts. Southward and Goldman (1978) measured Viscosities of high-, medium- and low-calcium co-precipitates. Apparent Viscosities decreased with increasing shear rate, indicating pseudoplastic flow behavior at concentrations of 150 g/kg and 100 g/kg. At each concentration the viscosity of the co-precipitates decreased in the order of high-, medium- and low-calcium preparations. Effect of Heat on Milk Proteins 9.15.213. The caseins exist as random coils in comparison with other protein secondary structure (Payens and Schmidt, 1965). Hence, in the chemical sense, caseins are considered to be "denatured", having little, if any, a-helix or B-structure. However, caseins do exhibit a native quaternary structure due to self-association with other caseins. Because the casein micelles are stabilized by hydrophobic, electrostatic and to a limited extent hydrogen bonding, temperature can drastically affect micelle alterations. Casein is not regarded as a heat-coagulable protein. In normal fluid milk, it is very stable to heat and may resist coagulation fbr as long as 14 hr at boiling temperatures and 1 hr at 130° C. Heat coagula- tion of casein in normal milk occurs as a result of increased acidity, the conversion of soluble calcium and phosphates to colloidal forms 32 and interactions (e.g., denaturation, hydrolysis) between protein components. Prolonged heating or heating above 100° C causes decomposition of caseins with the release of inorganic phosphorus and nonprotein nitrogen (Pyne, 1962). This release of phosphorus is related to acidity develop- ment in heated milk. Belec and Jenness (1961a, 1961b) studied dephos- phorylation of sodium caseinate and skimmilk casein heated at 110-140° C. Dephosphorylation conformed to first-order kinetics with an energy of activation of 25-29 kcal/mole. The release of phosphate was independent of pH in the range 6.0-7.0 and was greater fer a- than fbr B-casein. At a given temperature, dephosphorylation was slower in skimmilk than in sodium caseinate sols. Alais gt El: (1967) studied the effect of heating at 120° C fer 20 min on aS-, B-, and recasein solutions. All three casein fractions were affected, with B-casein being the most resistant to heat action and as-casein the most labile. For|<-casein, the action of rennin and heat were somewhat similar, indicating the existence of labile linkages which are split by different processes. Zittle (1969a) showed that K-casein heated at 96-100° C for 5 min in 0.05 M NaCl lost its ability to stabilize as-casein against precipita- tion by calcium ions, possibly by aggregation through disulfide bonds. The presence of aS-casein in a 1:1 weight ratio when the K-CBSEIH is heated prevents the heat lability of kappa (Zittle, 1969b). This fact plus the stability of'K-casein heated in 0.15 M CaCl indicate that the 2 aggregation of'K-casein in heated milk is not significant. Fox et_al, (1967) examined the nonsedimenting nitrogen of milk heated for 30 min at various temperatures. They fbund that heat 33 treatments greater than 110° C resulted in greater amounts of nonsedi- menting nitrogen, and that this fraction consisted primarily of casein. This effect was attributed to disaggregation of the casein micelles. Josephson gt_gl, (1967) noted that heating milk at 80° C fbr 30 min produced no noticeable alterations in the size, shape or charge density of casein micelles. Morr (1969) supported the finding of Fox gt._1, (1967) by the observation that small amounts of nonsedimenting casein components were observed with UHT milk. Serum Proteins Serum proteins are quite soluble, even at their isoelectric pH values. These proteins are not precipitated nor rendered nondispersible by routine heat treatments at normal pH. Whey proteins are sensitive to heat at temperatures above 65° C. Harland gt_al, (1952) studied the effects of heating time and temperature on serum protein denaturation in heated skimmilk. In the temperature range of 62-80° C the relationship of temperature to time for a constant level of serum protein denatura- tion is semilogarithmic. The heat treatments required for pasteuriza- tion of milk were below serum protein denaturing conditions. Larson and Rolleri (1955) examined the heat denaturation of specific serum proteins in milk by a quantitative electrophoretic analysis of the serum from heated skimmilk. Their classical denaturation curves place the immune globulins, bovine serum albumin, B-lactoglobulin and u-lactalbumin in order of increasing resistance to heat denaturation. Complete denatura- tion of the serum proteins was achieved by heating at 77.5° C for 1 hr or 90° C fbr 30 min. Morr and Josephson (1969) note that there are 34 three stages in whey protein aggregation. The first is denaturation, in which the proteins unfold from a compact, globular confbrmation to a random configuration. Following this, the proteins aggregate to fbrm intermediate sized complexes which are not sedimented at 1000 x 9. Finally, aggregation in the presence of calcium ions forms precipitate particles that are sedimentable at 1000 x g. B-Lactoglobulin is the major whey protein of cows' milk, and many heat-induced changes in milk have been correlated to the reactions of this protein. At neutral pH, B-lactoglobulin (B-lg) is thought to exist as a 36,000 molecular weight dimer containing two sulfhydryl groups and four disulfide linkages. The SH-groups are buried in the dimer complex and exhibit low reactivity. Sawyer (1968) has proposed a probable path- way for the heat denaturation of B-lg near pH 7. The initial effect of heat treatment is an increase in dimer dissociation to 18,000 molecular weight monomers. This is followed by molecular unfblding, in which the SH-group is fully exposed. The primary aggregation reaction occurs as a result of the unfolded monomer units associating via disulfide inter- change or sulfhydryl oxidation. The final stage is a secondary reaction involving a nonspecific aggregation (not resulting from the fermation of intermolecular disulfide bonds) and a heavy, 295 component is fermed. This heavy component undergoes further aggregation and opalescence develops in the solution. The susceptibility of the three genetic variants to denaturation was found to be in the order C>B>A. Watanbe and Klostermeyer (1976) examined heat-induced changes in sulfhydryl and disulfide levels of B-lg. They feund that SH levels decreased and SS levels increased in B-lg solutions heated at 95° C, and 35 concluded that SH-initiated-SS exchange reactions were important in the formation of high molecular weight polymers in the presence of air. Hillier and Lyster (1979) studied the denaturation of whey protein in heated milk and determined that the denaturation of B-lg followed second-order kinetics. The aggregation and precipitation of heat-denatured 3-19 is dependent on pH and the presence of calcium ion. little and DellaMonica (1956) observed a progressive decrease in the calcium ion concentration required to precipitate heated B-lg as the pH was lowered. The calcium bound at negatively charged carboxyl sites on the protein, reducing the net charge to zero, and bringing about isoelectric precipitation. The second major whey protein of bovine milk is a-lactalbumin (14,400 daltons) and it contains no SH groups and four disulfide link- ages. This protein is the most heat stable of all the serum proteins. The denaturation of a-lactalbumin appears to be first-order, but is felt to be a second-order reaction displaying pseudo first-order kinetics (Hillier and Lyster, 1979). There is a lack of information available on the effects of heat on a-lactalbumin. There have been reports of an interaction between a-lactalbumin and 8-19 in heated systems. Hunziker and Tarassuk (1965) obtained chromatographic evidence for this heat-induced interaction in phosphate buffer at pH 6.7. Lyster (1970) implicated the SH group of 3-19 as being important in the denaturation of a-lactalbumin. Elfagm and Wheelock (1977a) reported that a-lactalbumin interacts with an aggregated form of 3-19 when heated in synthetic milk ultrafiltrate. Help and 36 Hansen (1978) obtained electrophoretic evidence fer this complex in UHT milk. K-Casein and B-Lactoglobulin Interaction The complex formation between B-lg and K-casein during heat treat- ment of milk is believed to be of importance in the manufacture of co- precipitates consisting of whey proteins and casein (Southward and Goldman, 1975). Therefore, examination of the existing literature on this subject is in order. The first information regarding the interaction of casein and B-lg was provided by Tobias gt_al, (1952) and Slatter and van Winkle (1952), who detected a possible interaction by moving-boundary electrophoresis of heated skimmilk. DellaMonica gt_al, (1958) studied the effects of heat and CaClzcnisolutions of B-lg and casein. Heated B-lg solutions were very sensitive to precipitation by calcium, while a heated mixture of 1% B-lg and 2% casein would not precipitate until the calcium concen- tration was close to that required to precipitate casein alone. It was concluded that no complex was fbrmed and that B-lg precipitated inde- pendently of casein. Trautman and Swanson (1958) observed that in skim- milks heated to 180° F for 30 min, two-thirds of the original B-lg dis- appeared from the electrophoretic patterns. However, skimmilks contain- ing sulfhydryl-blocking agents gave electrophoretic patterns similar to unheated skimmilk. Trautman and Swanson (1959) later related this com- plex fermation to heat stability in evaporated milks. Kannan and Jenness (1961) suggested that formation of a complex between B-lg and casein in heated milk interferes with the primary action of rennet, 37 causing an increased coagulation time. Zittle gt_al, (1962) heated a mixture of x-casein and B-lg at 90° C fbr 15 min and noted a component of intermediate electrophoretic nobility and a $20 value three times greater than that of K-casein. They also observed that mixing heated 8-19 and unheated K-casein at room temperature caused complex formation. Evidence implicating the possibil- ity of intermolecular disulfide bond formation between B-lg and K-casein was presented by Sawyer gt a1, (1963). Kresheck gt_a1, (1964) calcu- lated a weight average molecular weight fer the complex fbrmed from heated mixtures of B-lg and recasein to be 6.54 x 106. Hartman and Swanson (1966) used polyacrylamide gel electrophoresis (9% T) to investi- gate this complex. Complex formation was complete when equal parts by weight of 3-19 and K-casein were heated at 85° C for 30 min. The com- plex was able to enter the spacer gel, but not the running gel. No complex was formed when K-casein was heated with either bovine serum albumin or a-lactalbumin. Grindrod and Nickerson (1967) reported that the heat-induced com- plex of B-lg and recasein did not enter polyacrylamide gels (7% 1). However, in urea and 2-mercaptoethanol, the proteins migrated the same as unheated controls, indicating possible complex dissociation. It was also noted that hydrogen peroxide would not induce complex formation. Purkayastha §t_a1, (1967) did not observe complex fermation when B-lg was heated with alkylated K-casein, or when B-lg was heated with K-casein in the presence of N-ethylmaleimide. These workers also observed complex dissociation in polyacrylamide gels containing urea and 2-mercaptoethanol, and concluded that thiol-disulfide interchange was 38 important in complex formation. In further model system work, El-Negoumy (1974) studied the B-lg/K-casein interaction in various salt solutions using a heat treatment of 110° C for 30 min. A minimum interaction of 13.8% B-lg and 17.1% K-casein occurred in de-ionized water, while a maximum of 76.6% B-lg with 83.3% K-casein occurred in milk dialysate. The effect of pH on protein aggregation in heated milk was invest- gated by Creamer gt El- (1978). Milk samples at pH 6.50, 6.65 and 6.80 were heated at 100° C for 30 min. Electron microscopy of the heated sample of pH 6.65 milk revealed small protuberances attached to the micelle surface. The pH 6.50 milk had small irregular attachments on the micelles, whereas the pH 6.80 milk had fewer attachments on the micelles and many interspersed particles that were irregularly shaped. The heated pH 6.65 sample was chromatographed over Sepharose 4B and the void volume was found to contain K-casein and whey proteins, but no other proteins. Mercaptoethanol and 6 M urea were necessary to resolve the void volume fraction by gel electrophoresis. Amino acid analysis of this fraction was not markedly different from a 4:1 mixture of whey protein and K-casein. In a study on gel fermation in yoghurt, Davies gt_al, (1978) also noted the appearance of appendages on the micelles of milk heated to 95° C for 10 min. These appendages appeared to give rise to a firmer curd with lower tendency to syneresis. The exact nature of the B-Ig/KPCGSEIU complex is still unknown. Many of the earlier studies were performed in buffers employing isolated milk proteins at protein concentrations different from that found in milk. Tessier et_al, (1969) were unable to demonstrate the existence of such a complex in heated milk. Thus, the involvement of sulfhydryl 39 groups in complex formation in heated milk remains unanswered. Moreover, if a complex is formed between B-lg and K-casein in heated milk, it probably is not identical to that found in heated model systems. El-Negoumy (1974) noted that various salts or salt mixtures could en- hance or suppress thiol group reactivity, which would lead to a variable degree of interaction and complex formation. Beeby gt_a1, (1971) pointed out that sulfhydryl groups may not be necessary fer the formation of the B-lg/K-casein complex, although they may be involved in the primary reaction of 3-19 aggregation. Lyster (1970) noted that three forms of denatured B-lg existed: one involving disulfide interchange and two others involving hydrophobic, hydrogen and ionic bonding. Long gt_al, (1963) observed that heated B-lg formed a complex with heated or unheated K-casein, suggesting the possible in- volvement of noncovalent interaction. McKenzie gtngl, (1971) has sug- gested that disulfide bonds may be a factor in the stability of the complex, but some less specific associations (e.g., physical entangle- ment of polypeptide chains) may also occur. Morr and Josephson (1968) proposed that denatured whey proteins are stabilized against heat- induced gross aggregation in skimmilk by complexing with casein micelles through calcium-dependent linkages. Recently, Shalabi and Wheelock (1976) demonstrated that a-lactal- bumin inhibited the primary phase of chymosin action on casein micelles, thus implicating this protein as part of a heat-induced comples. Baer gt_al, (1976) were able to show a heat-induced interaction between a-lactalbumin and B-lg in a model system. Elfagm and Wheelock (1977a) also demonstrated this, and felt that the a-lactalbumin interacted with 40 an aggregated form of B-lg. Elfagm and Wheelock (1977b) fbund that in heated milk, casein facilitates the interaction of a-lactalbumin and B-lg, supporting their view that a complex between B-lg, a-lactalbumin and K-casein occurs. They proposed a mechanism of action which involves the initial fermation of a complex between a-lactobumin and B-lg, followed by an interaction of this complex with K-casein. EXPERIMENTAL Chemicals and Materials Chemicals The principal chemicals used in this study and their sources are listed in the Appendix, Table A1. All were reagent grade unless other- wise indicated. Urea solutions were passed over a mixed bed ion exchange column (bed dimensions, 19*nnlx.35 cm) consisting of 50% Dowex 50x8 and 50% Dowex 2x8 to remove residual cyanate. Column bed was regenerated with 1.0 M NaCl. Distilled, deionized water was used in all experiments unless otherwise indicated. Corrmerc ial Products Grade A, low heat NFDM was obtained from Michigan Dairy Producers Co. Sodium caseinate and Eanro 50, a whey protein concentrate, were obtained from Stauffer Chemical Co., Westport, Connecticut. Sodium Protolac, a metaphosphate precipitated whey protein concentrate, was obtained from Borden, Inc., Columbus, Ohio. EDIE; The milk used in this study was obtained from the Michigan State University dairy herd which consisted of Holstein cows. All milk was obtained immediately after milking and was separated as soon as possible at 37-40° C on a Westfalia separator. The skimmilk was held at 4° C overnight and used the next day. 41 42 Egu'pment All equipment used regularly during the course of this study will be discussed here. Apparatus specific for certain experiments will be referred to in the appropriate section. All experiments were performed using stainless steel, plastic or Pyrex containers. An Instrumentation Laboratory pH/mV Electrometer, model 245, was used fer all pH measurements. Protein samples were dried from the frozen state by a laboratory-constructed 1yophi1izer. For most laboratory weighings, a top-loading digital Sartorious type 3716 balance was used. For analytical weighings, a Sartorious type 2433 balance was used. Low-speed centrifugations were performed with International Clini- cal or International Model U centrifuges. Intermediate-speed centrifu- gations were performed on a Sorvall, type RC2-B centrifuge equipped with temperature control. A Bausch and Lamb Spectronic 21 was used for spectrophotometric determinations. Vertical acrylamide gel electrophoresis was perfbrmed with electro- phoretic apparatus manufactured by Buchler Instruments, and using a Bio-Rad model 400 power supply. Gels were destained in a Bio-Rad model 170 diffusion destainer mounted on a Cole-Farmer model 4815 magnetic stirrer. Pictures of the gels were taken using a Polaroid MP-3 Land Camera. 43 Preparative Procedures Acid Co-precipitate Three liters of'skinnfilk were placed in a stainless steel con- tainer and brought to a temperature of 95° C. After heating for 30 min, the milk was cooled to 30° C and the pH adjusted to 4.6-4.8 with 1 N HCl. The precipitate was collected by centrifugation at 2000 rpm fbr 15 min. The precipitate was washed once with a small amount of water, then dispersed at pH 7.0, lyophilized and stored at 0° C. pH 3.5 Co-precipitate Three liters of skimmilk were acidified to pH 3.5 with 1 N HCl. This milk was then heated in a stainless steel container at 95° C for 30 min. After cooling to 30° C, the pH was adjusted to 4.8 with 1 N NaOH. The precipitate was collected by centrifugation, washed once with water, dispersed at pH 7.0, lyophilized and stored at 0° C. Hexametaphosphate Co-precipitate Three liters of skimmilk were brought up to room temperature and the pH adjusted to 3.0 with l N HCl. Sodium hexametaphosphate was then added to give a final concentration of 0.5% (w/v). The precipitate was collected by centrifugation, washed once with water, dispersed at pH 7.0, lyophilized and stored at 0° C. Calcium Co-precjpitate Three liters of skimmilk were heated in a stainless steel container at 95° C for 30 min. Immediately after heating, 20.5 ml of a 30% CaCl2 44 solution were added to give a CaCl2 concentration of 0.20% (w/v). The precipitate was collected by centrifugation, washed once with water, dispersed at pH 7.0, lyophilized and stored at 0° C. Chemical Methods Nitrogen A semi-micro Kjeldahl method was used for the nitrogen analysis (Swaisgood, 1963). The digestion mixture consisted of 5.0 g SeO2 plus 5.0 g CuSO4-5H20 in 500 m1 of concentrated sulfuric acid. Approximately 10 mg of dried sample was digested in 4 m1 of the digestion mixture over a gas flame for about 1 hr. The flasks were then cooled fer 15 min followed by the addition of 1 ml of 30% hydrogen peroxide. Digestion was then continued for an additional hr. Each flask was cooled and rinsed with water. The flask was then placed on distillation apparatus and neutralized with 25 ml of 40% NaOH. The released ammonia was steam distilled into 15 ml of 4% boric acid containing 5 drops of indicator. The indicator contained 400 mg of bromocresol green plus 40 mg of methyl red in 100 m1 of 95% ethanol. Distillation was complete when a total volume of 80 ml was collected in the receiving vessel. The ammonia- borate complex was titrated with 0.02 N HCl. A reagent blank and a tryptophan standard were analyzed to determine the average percent recovery of nitrogen, which ranged from 95-99%. Nitrogen analyses were performed in triplicate. Nitrogen was calculated as follows: %N = (m1 HCl - m1 blank)_(Norma1ity of HCl) (meg, wt._N) g of sampTe x 100 45 m The method of Hood (1973) was used for the determination of fat. Specific details of the method can be found in the Appendix, Table A3. Moisture Moisture was determined by a standard AOAC (1970) procedure. Approximately 0.5 g of sample was weighed into a previously tared metal dish and covered. This was placed into a vacuum oven at 95° C and left for 5 hr. This dish was cooled in a dessicator and reweighed. All moisture analyses were done in duplicate. tastes. Lactose was determined by the phenol-sulfuric acid method of Dubois gt_al, (1956) as modified by Barnett and Tawab (1957). Fifty to 100 mg of dry sample were ground in a mortar with about 5 ml of water. This mixture was transferred to a 50 ml volumetric flask and made to mark with water. A 10.0 ml aliquot was placed in a 25 ml volumetric flask and made to mark with 20% tricholroacetic acid. This mixture was then centrifuged at 7700 x g for 20 min. One ml of the clear supernatant was pipetted into a clean test tube and mixed with 1.0 m1 of water. To this mixture was added 1.0 ad of 5% (w/v) phenol and 5.0 ml of con- centrated sulfuric acid with thorough mixing after each addition. The test tubes were allowed to stand at room temperature until cool. Absorbance was read at 490 nm. All lactose analyses were perfbrmed in triplicate. 46 Ten ml of a solution consisting of 0.5135 g of lactose in 1000 ml of water were diluted to 100 m1. A standard curve was then prepared covering the range of 0 to 72.0 pg of lactose. A511 Ash was determined by the standard AOAC (1970) method. Porcelain crucibles were heated to 550° C in a muffle furnace, cooled and weighed. Two hundred to 300 mg of sample were weighed into the crucibles and a small amount of a 1% MgCl2 solution was added to prevent volatilization of phosphate. The crucibles were then returned to the furnace and ignited at 550° C for approximately 12 hr. The crucibles were cooled in a dessicator and weighed. Ash was defined as the percent of the original sample weight remaining as the residue (after correction for added Mg). Ash analyses were performed in triplicate. Calcium Calcium was determined by the EDTA method of Jenness (1953). A 100 to 200 mg sample was accurately weighed into a clean mortar and ground in 2 m1 of water. This mixture was washed into a 25 ml volu- metric flask, fbllowed by the addition of 15.0 ml of 20% tricholroacetic acid. The flask was made to mark with water and the contents centri- fuged at 7700 x g for 20 min. A 10 or 20 m1 aliquot of the clear super- natant was passed over a Amberlite IR-4B ion exchange column (bed size, 15 mm x 7 cm). The column was washed twice with 10 ml portions of water. To this 30 ml was added a little less than the expected amount of 0.02 M or 0.005 M EDTA. Sodium hydroxide pellets were dissolved in the mixture to bring the pH to about 13. Approximately 0.2 g of 47 of murexide indicator was added. The indicator consisted of 0.2 g of murexide plus 100 g of NaCl ground together in a mortar. The mixture was titrated with EDTA to a purple color that did not change with fur- ther additions. A blank and calcium standard were periodically analyzed to check recovery. All analyses were performed in triplicate. Phosphorus Phosphorus was determined by the method of Sumner (1944) as modi- fied by Swope (1968). Ten to 20 mg of sample were digested with 2.2 m1 of 50% sulfuric acid in test tubes. Digestion was carried out on a sand bath heated by an electric heater at a temperature of 160-170° C until the sample was thoroughly charred. The tubes were cooled, and 8 drops of 30% H202 were added. Heating was then resumed until all the H202 was driven off. After cooling, the digestion mixture was trans- ferred to a 25 ml volumetric flask. Five ml of a 6.6% ammonium molybdate solution and enough water were added to give a volume of approximately 15 ml. Then 4.0 m1 of a freshly prepared ferrous sulfate solution was added. The ferrous sulfate solution was prepared by dissolving 0.5 g of FeSO4-7H20 in 50 ml of water and adding 1.0 m1 of 50% sulfuric acid. The volumetric flask was made up to volume with water and mixed. After standing 30 min to allow for color formation, the absorbance was read at 660 nm. All phosphorus analyses were performed in triplicate. A stock solution containing 0.2742 g KZHPO4 dissolved in 200 ml water was prepared, and 10.0 ml of this solution were diluted to 100 m1 (.024 mg P/ml). This solution was used to make a standard curve cover- ing the range of 0 to 0.15 mg of phosphorus. 48 Amino Acids Amino acid analyses were performed on HCl hydrolysates of protein using a Beckman Amino Acid Analyzer, Model 120 C, according to the procedures of Moore §t_pl, (1958). Samples consisting of approximately 4 mg of protein were weighed into 10 ml ampoules. Five m1 of 6 N HCl were added to the ampoules. The contents were frozen in a dry-ice- ethanol bath and evacuated with a high vacuum pump. As the contents slowly melted, the gases were removed. The contents were then refrozen and the ampoules were sealed using an air-propane flame. The sealed ampoules were placed in an oil bath in a fbrced draft, recirculating oven regulated at 110° C for 24 or 72 hr. After hydrolysis, the ampoules were opened and 1.0 m1 of norleucine solution (2.5 moles/m1) was added as an internal standard. The hydro- lysate was then quantitatively transferred from the ampoule to a 25 ml pear-shaped flask. The hydrolysate was evaporated to dryness on a rotary evaporator. The dried sample was washed with a small amount of water and again taken to dryness. In all, three washings were performed to remove residual HCl. The washed and dried hydrolysate was dissolved in 0.067 M ditrate buffer (pH 2.2) and diluted to a volume of 5 ml. The solution was then filtered with a 0.22 m Millipore Filter, and 0.2 m1 aliquots were used for analysis. The chromatograms were quantitated by peak integration using a Spectra Physics Autolad System AA. Standard amino acid mixtures were analyzed using the same ninhydrin solution within a four-day period. 49 Methionine and Cystine Since methionine and cystine undergo a variable amount of oxida- tion during acid hydrolysis, they must be analyzed separately. The methods of Schram gt_gl, (1964) and Lewis (1966) were used. These methods involve performic acid oxidation of methionine and cystine to methionine sulfhne and cysteic acid, respectively. Approximately 5-8 mg of protein was weighed into a 25 ml pear-shaped flask. The protein was oxidized for 24 hr with 10 m1 of performic acid at 4° C. After oxidation, 1.0 ml of norleucine (2.5 moles/m1) was added. The performic acid was removed on a rotary evaporator. The dried sample was quantita- tively transferred to a 10 ml ampoule with 5 m1 of 6 N HCl. Hydrolysis and amino acid analysis were performed as previously discussed. Since the yield of cysteic acid is only 94 i 2%, the amount of cysteic acid obtained was divided by 0.94 to give a corrected value. Tryptophan The method of Spies (1967) was used. Approximately 3 mg of protein were weighed into vials. To this was added 0.1 m1 of a pronase solution (10 mg pronase/m1 of 0.1 M phosphate buffer, pH 7.5), and vials were incubated for 24 hr at 40° C. Following incubation, samples were placed in ice. To each vial was added 0.9 ml of phosphate buffer (see Appendix, Table A2), and the vials were placed into flasks containing 9.0 m1 of 21.2 N sulfuric acid and 30 mg of p-dimethylaminobenzaldehyde. Flasks were placed in the dark for 6 hr to allow the production of a chromophore resulting from the action to free tryptophan with the p-dimethylaminobenzaldehyde. Following this 0.1 m1 of sodium nitrite 50 was added, and the absorbance was read after 30 min at 590 nm. A pro- nase blank was run to correct fer the tryptophan content of pronase. A standard curve was prepared with tryptophan. Analyses were performed in duplicate. Sulfhydryl (SH) and Disulfide (SS)_ The SH and 55 content of milk samples were determined by modifying the procedure of Beverage gt_pl, (1974). For total SH, milk was diluted 1:1 with 0.2 M sodium phosphate buffer, pH 8, containing 8 M urea (for buffer preparation, see Appendix, Table A2). To a test tube containing 2.5 ml of the above sodium phosphate buffer and 0.02 ml of Ellman's reagent (40 mg of 5,5'-dithiobis-2-nitrobenzoic acid in 10 m1 of 0.2 M sodium phosphate buffer, pH 8) was added 0.5 m1 of the diluted sample. Color was developed for 15 min, and the absorbance was read at 412 nm against a blank containing protein and buffer, but no Ellman's reagent. The SS groups were determined by adding 0.1 ml of the milk sample to a test tube containing 1.0 ml of 0.2 M phosphate buffer, pH 8, con- taining 10 M urea and 0.02 ml of 2-mercaptoethanol. The tube was incu- bated for 1 hr in a 25° C water bath. Ten m1 of 12% trichloroacetic acid solution were added, and the mixture was incubated an additional hour at 25° C. The mixture was centrifuged at 1000 x g fer 50 min. The precipitate was twice resuspended in 5 ml of 12% trichloroacetic acid and centrifuged to remove all traces of the 2-mercaptoethanol. The precipitate was dissolved in 3.0 ml of 0.2 M phosphate buffer, pH 8, containing 8 M urea, and 0.03 ml of Ellman's reagent was added. 51 Color was allowed to develop for 15 min, and the absorbance was read at 412 nm against an appropriate blank. Total solids were determined from a standard procedure (AOAC, 1970). Calculation of SH in umoles/g solids was done according to the fellowing equation: pmoles SH/g solids = 73'53 3 A x D where A = absorbance at 412 nm 0 = dilution factor (12.08 for SH, 30.0 for 55) C = concentration in mg solids/m1 (8.84 fbr milk, 8.53 for pH 3.5 milk) 73.53 is derived from 106/13,600, where 13,600 is the molar absorptivity and 106 is fbr conversion from the molar basis to the pmoles/ml and from mg solids to g solids. Soluble Protein Soluble protein was measured by the method of Lowry gt_§1, (1951). Solution A was prepared by dissolving 100 g Na2003 in 0.5 M NaOH and making to one liter. Solution 8 was prepared by dissolving 1.0 g CuSO4-7H20 in water and making to 100 ml. Solution C consisted of 2.0 g potassium tartrate dissolved in water and made to 100 m1. Solution 0 consisted of 15 m1 A, 0.75 m1 8 and 0.75 ml C. Solution E consisted of 50 ml water plus 5 ml of the Folin-Ciocalteau reagent. To each tube was added 1.0 ml of a protein solution, followed by 1.0 m1 of solution 0. This was mixed and allowed to incubate far 15 min. Solution E was prepared during this period. After the incubation, 3.0 52 ml of solution E were forcibly pipetted into the tube, and the tube was vortexed immediately. This was then incubated an additional 45 min, and the color read at 540 nm. A standard curve was also constructed with a protein range of 0-300 pg. Bovine serum albumin was the standard protein. The protein content of this protein was determined by Kjeldahl to be 85.3%. Functional Tests Solubility Solubility was determined by a centrifugation method similar to the one used by Mattil (1971). Enough sample to give a 1% protein solu- tion (based on Kjeldahl) in 25 ml was weighed out and ground in a mortar with about 5 ml water. The contents were washed into a small beaker and the volume brought up to 15-20 ml. The pH was then adjusted to either 7.0, 4.5 or 3.0 using 0.5 N HCl or 0.5 N NaOH. The total volume was then made to 25 ml in a volumetric flask. Following stirring for 5 min, the solution was centrifuged at 4300 x g for 20 min. The supernatant was filtered through Whatman #1 filter paper. A 10 m1 aliquot was taken and diluted to 500 ml. One ml of this solution was used to determine protein content by the Lowry method. Solubility was determined as follows: % solubility = 20° ' 9 nggg" 1" “1‘1"“ x 100 Duplicate trials were performed for each sample at pH 7.0, 4.5 and 3.0. 53 Water Hydration Capacity To determine water hydration capacity (WHC) the method of Quinn and Paton (1979) was modified. WHC, as determined in this manner, represents the amount of water held by a centrifuged pellet without solubilization of the sample. Two 9 of a sample were weighed into a tared, transparent centrifuge tube. Water was added in small increments and the mixture stirred vigorously until a paste-like consistency was obtained. The tube was centrifuged at 8700 x g for 15 min, and the slight supernatant was discarded. The tube was reweighed immediately, and the difference in weight represented the approximate WHC in ml water/g sample. To get a more exact number, a series of fbur tubes, each containing 2 g of sample, were mixed with a range of measured volumes of water that encompassed the estimated WHC. For example, if the estimated WHC is 3.0, then 5.0, 5.5, 6.0 and 6.5 m1 of water would be added to the 4 tubes. Each mixture was stirred vigorously fbr about 2 min and centrifuged at 8700 x g for 15 min. Immediately after centri- fugation, the tubes were examined for the presence of a supernatant. At least one tube should have a supernatent and at least one should not, providing the estimated WHC was a good approximation. WHC was reported as the midpoint of the range between the tubes with and without super- natant water layers. Viscosity For the measurement of viscosity, samples were dispersed in water at concentrations of 3.0, 5.0, 7.5, 10.0 and 15.0% (w/v). The pH was checked, and was between 6.8-7.2. One rd of sample was placed in the 54 chamber of a Brookfield Viscometer, Model LVT, and allowed to equili- brate fbr several minutes. Temperature was maintained at 25° C with a Colora Ultra Thermostat circulating water bath. Shear stress was measured as a function of shear rate over the range 0-240 sec-1. Emulsifying Capacity The procedure of Webb gt_al, (1970) was modified slightly for the determination of EC. To a 600 m1 beaker containing 100 ml of 1.0 M NaCl solution was added 5.0 m1 from a protein sol equivalent to 10 mg protein, This mixture was weighed, and covered with a rubber stopper fitted with holes for a propeller, two electrodes and an oil delivery tube. The propeller was driven by a Talboy T Line variable speed, laboratory stirrer. The electrodes were attached to a Triplett Model 630 volt- meter. Refined corn oil was added from a 500 m1 separatory funnel at a constant flow rate of 12 ml/min while stirring the mixture at an initial speed of 4400 rpm. Oil delivery continued until the inversion point was reached, at which time the voltmeter showed infinite resistance. The beaker was then reweighed to determine the amount of oil added. The EC was calculated as the g of oil required to reach an infinite resist- ance minus a blank (100 ml of the 1.0 M NaCl plus 5.0 ml water) divided by the amount of protein. Analyses were performed in triplicate. Whi in Whipping properties of the various proteins were determined by whipping 50 m1 of a 3.2% protein sol at pH 7.0 fer 6 min at speed 8 in a Kitchen Aide Mixer, Model 3-C, equipped with a wire whip. Specific volumes (ml/g) were determined by transferring the foam to a tared 55 beaker of known volume and reweighing the beaker. Foam stability was measured by inverting the beaker over a half inch stainless steel mesh screen and collecting the liquid draining from the foam. The time required fer collection of liquid equal to one-half of the weight of the original feaniwas designated as foam stability. Stability times of less than 5 min were reported as zero. All analyses were performed in triplicate. Physical Methods Polyacrylamide Gel Electrophoresis (PAGE) PAGE was run according to the method of Melachouris (1969) in a discontinuous buffer system. All electrophoresis was performed in 6 mm 1.0., 2 mm walled, and 75 mm length glass tubes. The tubes were deter- gent washed, immersed in chromic acid and treated with Photoflo after use. Running gels of 10% total acrylamide concentration (10% T) and 5% crosslinking (5% C) were prepared using a 0.380 M Tris-H01 buffer, pH 8.9 (see Appendix, Table A2, for buffer preparation) with the ratio of acrylamide to bisacrylamide being 19:1. Spacer gels of 5% T and 5% C were prepared in 0.062 M Tris-HCl, pH 6.7. Polymerization was initiated using a 5% ammonium persulfate solution and Temed (N,N,N',N'-tetramethyl- ethylenediamine). The electrode buffer was a pH 8.3 solution of Tris- glycine. Samples of 0.5% protein were prepared in the pH 6.7 buffer. and sucrose was added to increase density. Bromophenol blue was added as a marker dye. 56 PAGE was also run in gels containing 7 M urea. All buffers remained the same, except for the presence of 7 M urea. A constant current of 3 mA/tube was applied to the gels until the marker dye reached the bottom of the tube, usually 45-60 min. Gels were then removed from the tubes and stained. SDS PAGE SOS gels were run according to the method of Weber gt_pl. (1972). Gels of 10% T with a acrylamide to bisacrylamide ratio of 37:1 were pre- pared in 0.1 M phosphate buffer, pH 7.2, 0.1% 505 (see Appendix, Table A2, fbr buffer preparation). Polymerization was initiated using ammonium persulfate and Temed, as above. Samples were prepared in 0.01 M phos- phate buffer, pH 7.0, 1% SDS, 1% 2-mercaptoethanol, with a final protein concentration of 0.2%. This sample was placed in a boiling water bath for 5-10 min. Some samples were prepared without mercaptoethanol. After cooling, sucrose and Bromophenol blue were added, as above. The electrode buffer consisted of 0.1 M phosphate buffer, pH 7.2, .01% SOS. A constant current of 8 mA/tube was applied to the gels until the marker dye reached the bottom of the tubes, usually about 4-5 hr. Gels were then removed from the tubes and stained. Relative mobilities were calculated by measurement of the protein zone migration distance, dye migration distance, and the length of the gels from the following relationships: distance protein migrated distance dye migrated length of gel bpfbre destaining Tength of gel after destaining Relative mobility = X 57 A standard curve was constructed using an electrophoresis calibra- tion kit for molecular weight determination of low molecular weight proteins (Pharmacia). The kit contained the following proteins: phosphorylase b, 94,000; bovine serum albumin, 67,000; ovalbumin, 43,000; carbonic anhydrase, 30,000; soybean trypsin inhibitor, 20,100; a-lactalbumin, 14,400. Staining Gels were stained by the method of Reisner gt_gl, (1975) using Coomassie Brilliant Blue 6250 in 3.5% perchloric acid (see Appendix, Table A2, for stain preparation). Staining was usually carried out overnight, and gels were destained in a diffusion destainer in 7% acetic acid. Statistics Analysis of variance was completed on all functional test data in addition to the yield and sulfhydryl information. If significant differ- ences were found, the Tukey technique was used to pinpoint the difference (Hays, 1973). RESULTS AND DISCUSSION Preparation of Co-precipitates Four co-precipitate samples were prepared in order to vary their functional properties and to study some physical properties of the products. Abbreviations for the various products referred to can be found in Table 5. Products produced in this study were not made under the same conditions as commercial co-precipitates. For example, the heated co-precipitates (ACP, P3C, CAC) all had heat treatments of 95° C for 30 min, followed by precipitation. American-made co-precipitates are made from milk heated at 90-95° C fer 20-30 min, concentrated to 10-30% total solids and heated to 140° C fbr 20 5 prior to precipitation (Noyes, 1969). Australian and New Zealand processes typically involve heat treatments of 85° C for 6-8 min by steam injection followed by precipitation (Southward and Aird, 1978). Muller gt;gl, (1969) noted that the heat treatment required to precipitate most of the whey pro- teins by acid was much greater (92° C far 15 min) than that required when calcium chloride was used as the precipitant (92° C for 5 min). Thus, although the acid co-precipitate (ACP) and calcium co-precipitate (CAC) were intended to simulate commercially available products, the similarities may only be approximate. 58 59 Table 5. Abbreviations for products referred to in this study. ACP - Acid co-precipitate P3C - pH 3.5 co-precipitate HMP - Hexametaphosphate co-precipitate CAC - Calcium co-precipitate NAC Sodium caseinate NDM - Nonfat dry milk SWC Whey protein concentrate (gel filtration) BWC Whey protein concentrate (metaphosphate) The preparation of the pH 3.5 co-precipitate was based on a paper by Modler and Emmons (1977) in which sweet whey was heated at 90° C for 15 min at pH 2.5-3.5, cooled and precipitated at pH 4.5. Protein recovery was 20-30% higher when FeCl3 was added prior to heating. By heating skimmilk at pH 3.5 for co-precipitate preparation, it was hoped that unique functional properties would be achieved. The use of sodium hexametaphosphate (SHMP) to produce a co-precipi- tate represents a method fer producing a protein concentrate without resorting to high temperatures. This method was used fbr recovering whey proteins (Hidalgo gt_al,, 1973; Richert, 1972). Zittle (1966) showed that caseins were precipitated by divalent or polyvalent anions whereas B-lg was precipitated by polyanions only under acidic conditions. In acidic environments, the protein acts as a polycation and the SlMP acts as a polyanion. Insoluble complexes form as the result of poly- cation-polyanion interactions. Likewise, long-chain polyphosphates promote intermolecular crossbonding and aggregation. 60 Composition Compositional properties of the four co-precipitates studied are summarized in Table 6. The pH 3.5 co-precipitate (P3C) has an unusually high moisture content. This characteristic may be a ramification of a relatively low ash content which could allow for more available water binding sites on the proteins. Also, the high lactose content could account for a portion of this moisture. Specimens were dried by lyophil- ization, therefore one must consider the possibility of incomplete dehydration in the system, as a large number of samples were dried at one time. Table 6. Proximate compositional analysis of co-precipitates. ACP P3C HMP CAC Moisture (%) 7.3 15.2 5.2 5.6 Fat (%) 3.0 2.1 2.0 2.5 Ash (%) 3.8 2.1 13.0 9.2 Protein (N x 6.38) (%) 78.5 73.0 67.7 69.5 Protein (dry basis) (%) 84.7 86.1 71.4 73.6 Lactose (%) 5.8 3.9 8.7 8.1 Calcium (%) 0.2 0.2 0.7 2.3 Phosphorus (%) 0.7 0.6 3.3 1.6 pH of whey after separation 4.6 4.6 3.1 6.2 Washing has been shown to affect the composition of isoelectric casein and co-precipitate (Buchanan gt_gl,, 1965). On the commercial scale, washing is accomplished with pH and temperature controlled 61 (pH 4.6, 30-35° C) water and is repeated 2-4 times. In this study, only one wash was utilized, and there was no pH or temperature adjust- ment of the wash water. Consequently, some of the values for lactose and ash may be slightly higher than normally encountered. For example, commercial calcium co—precipitate offered fbr sale by New Zealand Milk Products, Inc. (undated) have lactose and ash levels of 0.6% and 8.0%, respectively. The CAC prepared in this study had levels of 8.1% and 9.2% for these components, respectively. Similarly, the other co- precipitates may have been affected by the washing procedure. The fat contents of all feur samples were higher than found in commercial products (0.6-1.0%). Buchanan gt_gl, (1965) attributed the higher fat content of co-precipitates when compared to casein to fat- protein complexes formed during the manufacturing process which, presum- ably, are retained in the curd. In this study, co-precipitates were produced from inefficiently separated skimmilk which could account far the relatively high fat content. The calcium and phosphorus contents are similar to those found in commercial samples. The hexametaphosphate co-precipitate (HMP) has a very high phosphorus level, resulting from the use of SHMP as an aggre- gation agent. The slightly higher calcium level may have resulted from SHMP sequestering. CAC had higher calcium and phosphorus levels than the other co-precipitates. D'yachenko gt_g1, (1953) also observed higher calcium and phosphorus levels in calcium co-precipitate. The amino acid composition of the protein in each of the co-precipi- tates is shown in Table 7. Data for the commercial co-precipitates and casein were obtained from technical brochures. Under the conditions of 62 .msep mwmxpocuaz oLmN op umumpoaecuxm mew: moron ocwso omega so» mmape>m .Aeempv meeem ee eeeeee wee ea eeeceeeeee we: eeeeeeezee u .zpm>_uomammc .uwue upmumau was m=o¥F=m mcwcowgume we up:FEcmuwv mew: mcwuon\mcvmum»o ecu mcmcowsymzu .Aumueucsv .FPH .ucosmmoz ..o:H .muuzvoca xFPz use—emN 2oz seem :meeu can weepwawumcaiou Esvupeo a .Avmumucav .PPH .ovcu ..ocH ..ou :Pmmmu owcm sage opmuwapumcniou Europeuigmwzc mN.F em._ em._ eo.~ ee._ -.N mN.N eeeeeeeeaee mm.m em.m Ne.e um.m em.e mm.e me.e eeeee_eF»eeea me.m oe.m ee.m -.e we.e mm.m ee.e eeeemeese oe.e em.m oe.m mw.c m_.e Ne.m mm.m eeeeees Fm.m eo.m ee.e e_.e e_.m Ne.e em.e eeeeeepemH mm.~ _e.~ N¢.N pe.~ Ne.p ,PN.~ eo.m eeeeeeeeez mm.e mm.e om.m we.m ep.e Po.e om.e eeepe> Ne.o o_._ Ne.o mo._ em.o oe.o Fo.P eeeeemee.epez Np.m oe.m No.m oe.~ mm.~ mm.m ee.~ eeeeep< e_.~ mo.~ FF.~ mm.P me.F am._ me._ meceepe me.mp em.e. o~.e mm.m FF.“ em.m ew.e eeecpeea ee.- mm.e_ om.m_ me.mp Ne.o~ em.e_ me.ep eeee easeeece e,.e _o.m mm.e pe.e Nm.e em.e Pe.e emceeem e~.e em.m oo.e mm.e m~.e e_.e mm.e emceeeeeee up.“ me.e ee.e om.e mN.e me.e Np.e ecee e_eeeem< em.m __.e mm.e we. Fo.m e~.m AP. eececme< om.~ ee.~ cm.~ e..~ ee.~ ee.~ mm.~ eeeeeem_= mo.m mm.e me.“ .m.e m~.w ee.e mp.e eeesz eccemee eaoe ema mm.~ ii mm._ w~.F no.~ an._ -.~ mm. o.~ cezaouaxgh mm. em.e oo.e m~.e mm.e mm.e mp.e mm.e o.e mcwcomech Nc.h mm.m on.m mm.op eo.op mm.m om.op NF.P— o.o mammoczu \mcpcmpmpxcmza mm.m m~.m em.m m~.m ow.m Fe.~ Po.m mo. m.m muvue ocean campzm en.m Ne.m me.“ mo.m pm.m m~.w m~.~ mp.n m.m enema; oo.mp eo.op ow.m om.m mm.n mp.m mc.m mm.m o.“ wcwuzwg mm.m oo.o mo.e pm.m o—.e up.m ~m.¢ em.e o.e ocmuzmpoma cue: cram: ua1 3’ . n (J Electrophoretic patterns of proteose-peptone specimen from skinnfilk.(A), and co-precipitate aqueous phases of ACP (8), P30 (C), HMP (D) and CAC (E). Total acrylamide concentra- tion was 5% and 10% in the spacer and running gels, respectively. 69 This whey fraction is presented in Table 9. Casein and whey were reported to have values of 1.65 and 11.21, respectively. Ratios fbr ACP, P3C, HMP and CAC were 2.36, 2.95, 3.72 and 3.43, respectively. The amount of whey protein in a co-precipitate could be found from a graph of whey fraction vs. proportion of whey protein in a casein-whey mixture. However, the original publication could not be located, so estimation of the whey protein content was not possible using this procedure. These values do indicate a moderate amount of whey protein in the co-precipitates based on the values for casein and whey. The whey protein content of co-precipitates was also estimated by de Koning and van Rooijen (1971), using the cysteine plus cystine con- tents. Cystine was reduced to cysteine with dithiothreitol and total cysteine was determined. The content of whey protein in a co-precipi- tate was calculated by the fermula: x - 0 26 % whey protein = 3 ' x 100 where x equals the percentage of cystine plus cysteine in the co-pre- cipitate. Using the amino acid data from this study, the amount of whey protein in each co-precipitate was estimated and results presented in Table 9. The percentage of whey protein in ACP, P3C, HMP and CAC was 27.1, 19.3, 12.1 and 30.4, respectively. The low value of HMP is prob- ably due to incomplete precipitation of B-lg, which results in a reduced cysteine content, as reflected in the amino acid data and chemical score. The value for P3C of 19.1 seems reasonable when considering that skim- milk protein normally contains 15-22% whey protein. The higher values for ACP and CAC could reflect losses of fine casein particles when 70 separating the curd and whey. Also, some error in the amino acid data is possible. de Koning and van Rooijen obtained percentage whey protein values ranging from 10.4-16.0% for different co-precipitate samples. Southward and Aird (1978) reported casein:whey protein ratios of 6:1, 7:1 and 8:1 for high-calcium, medium-calcium and acid co-precipitates, respectively. Functional Properties Southward and Goldman (1975) noted that little work had been reported on the functional properties of co-precipitates. Recently, some literature has appeared regarding functional properties (Chojnowski gt_gl., 1975; Southward and Goldman, 1978). However, several factors complicate the comparison of data from different laboratories. Varia- tion in how functionality is assessed creates a problem, since currently there are no standardized functional tests available. In addition, the term "co-precipitate" has never been defined. Functional evaluations in this study were performed in simple model systems. It must be emphasized that model systems do not necessar- ily simulate more complex food systems. Therefore, it should not be unexpected to find that a protein preparation behaves differently in real fbod systems than when evaluated in a model system. SqubiIity Solubility is regarded as the principal property that precludes the overall functionality of proteinaceous materials and provides a good index for specific applications (Kinsella, 1976; Chou and Morr, 1979). 71 Operationally, solubility represents the percentage of total protein retained in a supernatant after centrifuging a protein solution at a standardized speed and time. The solubility profiles of four commercially available milk protein products and the four co—precipitates prepared in this study are shown in Figures 2 and 3. The three heat-treated co-precipitates (ACP, P3C, CAC) have profiles similar in shape to that of NAC, although CAC is partially insoluble at neutral pH. The solubility curves fbr HMP and BWC are essentially similar. At pH 7.0, the solubilities of SWC, BWC and HMP were similar, whereas NAC, NDM, HMP and CAC differed at the 1% significance level. ACP and P3C were not significantly different from each other, but were different from all the other products at the 1% level. At pH 4.5, NAC, ACP, P30 and CAC were completely insoluble, whereas the remaining products had solubilities that were significantly different at the 1% level. At pH 3.0, the solubilities of all the products were significantly different from each other at the 1% level, with the exception of CAC and NDM, which were not different, and ACP and CAC were different at the 5% level. ' Southward and Aird (1978) determined the solubilities of high- calcium, medium-calcium and acid co-precipitates at pH 7.5 and noted that their solubilities were not as complete as that of acid casein due to the inclusion of heat-denatured whey proteins. Smith and Snow (1968) examined the solubilities of high-, medium- and low-calcium co-precipi- tates over the pH range 4-8. Solubility profiles were all similar to that of sodium caseinate. 72 .on 85 .AAV as: .An: 02 .AAVV ae< .o. A eee m. e .o. m :e ee mace» reapomsaiou do mAAeocn AuAppaapom .m weaved .fin OK 0.0 O.“ O.‘ 0.0 O . \IJIJ e . 1 ON a L L 0V 4 . 0 1 00 m 1 On 4 OO— (%) AIITIGI‘lTOS .A. 2% A3 2: .An: 2.5 .3 03m .9 A eee m. e .9 m .3 a 2:95 3:238 \/ .N weaned \rlwo 10" m rat. .I .nu m . nu I. .A 100 Cl. 10o 02 73 The variation in solubility as the pH increased or decreased from pH 4.5 is due to increased net charge of the protein molecules, result- ing in greater repulsion between protein molecules and greater relative attraction for water molecules. For the products precipitated with SHMP, solubilities decreased with pH, with complete insolubility observed at pH 3.0. At this lower pH, the SHMP-protein complex precipitates, due to maximum polycation-polyanion interaction. CAC exhibits little solubility at neutral pH due to its high calcium content. El-Negoumy (1971) attributed the instability of caseinate sols at high calcium concentra- tion either to a lower zeta potential or to crosslinks established by the divalent cation. At lower pH, CAC has an increased solubility due to the dissociation of calcium ion from the proteins at this pH. High- calcium co-precipitates are often rendered more soluble by adding com- plex polyphosphates such as sodium tripolyphosphate to sequester the calcium (Hayes gt p1,, 1969). Viscosity The viscosity of protein dispersions can be used to assess the thickening power of proteins, a property of practical interest in fluid foods, soups, beverages, batters and other formulations. A knowledge of viscosity behavior is important to both processing and process design and in new product development. The effect of concentration on absolute viscosity for eight milk protein products are shown in Figures 4 and 5. A comparison of the profiles for the co-precipitates (ACP, P3C, HMP, CAC) with NAC shows similarities in shape between the curves. For the shear rates examined, 74 .on 05 .AS 2: .23 e2 .on a: .mmuouAAAumLAiou newcomepn Go Aapmoump> mu=FOmnm co cowumcucmucou we acumen .m mcamwu 3L 202.5.th0200 9.... Wu. 0.2 ms 0.... win no + 4 d A u _ \II C II.“ m . D\ o.\\\\ ‘\\\\d.\\\ nan a o \\\\\\ ca: 0 o .A a Ava. mm nu .3 U .A Anon \: 3 .d “P 0.3 Axon .on .5: .2: 2.2 .2”: use .on us. .mapasem cemaocn AAAE acogmewwu ea Anemou im_> mazpomne co cowpecucmucou eo uumemw .e wcampm :3 20.2523on 0.2.— nfi— 0.0— n N 06.. n.« 0 . a A u n O nil-IQ “M\M\O \\\\\ IAfin O 04. .. 0.0— .. od— 1 0.0N (m) msoasm .. Odu 4 0.00 75 all samples displayed Newtonian flow behavior in the concentration range up to 15%. Hermansson (1975) studied the viscosity of casein in dis- tilled water. An almost Newtonian flow behavior was a common feature of caseinate dispersions below 12% concentration. The whey protein products (SWC, BWC) and NDM exhibited the lowest Viscosities; HMP was also quite low. Whey proteins are highly soluble and exhibit low swelling, and thus display low Viscosities. Harte (1978) observed the lower viscosi- ties of whey products when compared to casein products. HMP had lower Viscosities than the other co-precipitates. This product exhibited high solubility and moderate water hydration capacity (discussed in next section), probably a ramification of a reduced water- protein interaction. This characteristic was especially apparent when compared to ACP, P3C or NAC. However, the protein content of HMP was slightly lower than the other co-precipitates. The greater viscosity of ACP over NAC, especially at the higher concentrations, may be partially due to its higher calcium content and heat treatment. Caseinates (excluding calcium salts) typically contain 6-7% ash, but negligible amounts of calcium (Hargrove and Alfbrd, 1974). Hayes and Muller (1961) demonstrated the effect of calcium on the viscosity of casein solutions. Casein solutions with 1% calcium exhibited higher Viscosities than those devoid of calcium. Hayes gt_gl, (1961) ascribed some of the viscosity difference between a low-calcium co-precipitate and casein to the formation of protein-protein linkages during the manufacture of co-precipitate. P3C shows lower Viscosities than ACP which may reflect differences in ash content. Hermansson (1975) reported that increasing ionic strength 76 increased the viscosity of caseinate solutions. Also, the protein con- tent of P30 is about 6% less than ACP. Differences in method of prepar- ing P30 and ACP may also be important. CAO was extremely insoluble, and viscosity was very difficult to measure. In the use of proteins as thickeners, CAC would have no application. Morr (1979) noted that calcium caseinate exhibited a much lower viscosity than sodium caseinate. Southward and Goldman (1978) reported that acid co-precipitate yielded higher Viscosities than sodium caseinate. They concluded that the occluded calcium content of co-precipitate and caseines is the prime factor causing variations in the viscosity of these products, whereas co-precipitated whey proteins had little, if any, effect on viscosity. Water Hygration Capacity The ability of a protein to bind and hold water is extremely important in many food systems such as meat products, bread and cakes. The term water hydration capacity (WHC) denotes the maximum amount of water a protein material can imbibe and retain under conditions of food formulation. The WHC for eight milk protein products are presented in Figure 6. There is no significant differences in the WHCs or SWC, BWC and NDM, and in NAC and P30. All others are significantly different at the 1% level. ACP exhibited the greatest WHC, followed in decreasing order by P30 and NAC, HMP, CAC and SWC, BWC and NDM. The presence of other components in a fend system, such as salt and lipid, are expected to change the hydra- tion characteristics of a protein product considerably (Quinn and Paton, 1979). Proximate analysis (Table 6) reveals compositional variations 77 .mmpgsmm cpmuoea cmzuo op umeaaeou mmpeuAaAumcq icu eo mmvupumaeo mcpzm_m—:Eu Asset use... a. £5.33 95:55.: as ox annual, 0.0 i1 0.». 0* A 0.» 0d 0.— 0 idIIII1IIII1IIIJ F1 .A deemed U umeomam .m weaned A532... 5239 0.23.; 81 n. m. «I. a o ump am on» an mucmow$_:mpm monocmo « .pm>wp RF on» an ucmsmwmwu xpucmu_mwcmwm ogm uamgumsmazm cossoo a azogpwz meanssc .36; some c_;uwz DAN_.VKm.N qu~.v~m.m DA~_.Vm_.m aA~_.V~o._*GAN_.VNm N nA~,.vmm.m - Gapp.vmm.¢ afieo.vom.PP ofipm.vwm.mpuA_o.v-.~ UANF.Vmo.m on~.vom.e Lfimo.v~m._ aAmo.vhm.~ efl~o.v~m.~ oA~o.vmm._ afimo.fioo.~ eAm.vk.¢e . eo. mae.mvo.m~ so aAN PVN we so mam.vm.m afim.~vm.~m *uflo.FV~.Fo no JA~.evo._m aA~.pv~.m_ no aAm.Vo.mF Acpsvaswp ~\_ no uA¢.vo.__ ufip.vk.m_ no mfim..o.m m\_s mcpaawgz afimo.voo.m uAm_.V~o.e DAN_.Vm_.m aA-.vom.m mfle~.Vem.ofius\mvm=w»wwwmmnw » umam JAN—.vmm.m afioo.vom. *aAN_.V~P.m afioo.voo._ mfiso.vmo. Am\_ev cowwaaesm - nAmo.vmo.~ - uAm_.vmp.m mamo.vmm.m Rm. uAm~.vmm.~m afiop.vmm.. uAmo.Vo¢.¢~ aamo.V~m.. afiovme.~ No. nA~_.V¢_.mF - aAwo.voe.m - - 3m.“ efiovem.m - ufimo.vsm.m afieo.vmo._ mfiwo.vem._ am so. we. - _o. _o._ - - Rm .ucao nA V N mA v Amaov >u_moumw> *efim.vm.~m uAm.evm.oe Jae..vo.~m nam.vm.m afic._vo.~m o.m 2a 00 uaovo.- so afiovo.m_ mam.vo.mo m.¢ za *ufiq._vo._o oae.~vm.~o 5A~._vm.mm aAm._V~._a afim.va.~m o.~ Ia. ARV »»__Pn=_om 38 as... 0mm x. 9 1oz 3E 03m 35 338.5 ..>mu .uumv a=_m> .mwmme pmcowaucz$ maowgm> Low museummwcmwm mo mpm>w_ use m:o_umw>mc cgmucmum .mw=_m> pmuvng=z .¢< mpamh