. . r 1 ft . A . .3 I , . i~ V...fl..1.fi.. .. . y . , pr? ‘ . ‘ .‘ y . ‘ . . 1.90.... . .. . . . ‘ . . . . . .1 A y . , . ‘ V . h‘l...h .E E. {.12 71.3). v: ,1), JitthI... ' )VI )1 | 'pct- v lazily. 1V... . ti.) 1!. 51.35%} ’9; i pit... (3:51- 31.11323! ( gégn)‘ '4 >16? " I 7. . é.) ,; . aria}. 5 Add... . . ‘1’po {a . I.) {5.} :9, :5; ta .3 z x u 1. at. Radix: 1.3.15‘. 5? (.13‘..IIIL v C v\ 1.1.51.2; €0.74 3' I'll .. 5 (OR, ivlr...‘ E... :P. K :...K.&fl1...h...l..’!f§fnlnfl.b ‘ 13" .522: 9.»: I! .x. (1'. .1.) . .) £3 . 1. . . hilkfi. . 5.. 1.... . . ..ht.:(¢.:... It 7 .‘ r .a , Yr):<.0..v . . , . . . 4‘: , 4. . . .54. - I )1. 4.x . . .. {.1 9w”... u?” "River. 91.3" . g u _ . c.......!.l.. . . ...‘ ..‘ ‘. . L . ..... , . . .12... ..v V .. AI v.9..5 ‘ ck .) .0. . . L . : . \Hnyunfishuldnhviflubxuhufiwwu quwnmvfilbfl , . A . .. . ‘ ‘ . ”H4J1J‘fllflh.‘ a; , I249! 00:: ‘ I . I .‘....5» I . ‘ H. .4 L .. . o“ . ‘ . u. .. Will. ‘ y x ...I. .n . . I w .. . .l E [ DLIX‘IOJ’ ~ h li..nr 51;. . 1: .Mn N. 4‘ , . . , - , . u v v.5. . ; .T .. . . LEP. 5.9!: afffinwLLmWwywfiho- 1.....4 :5}. .2.“ . 9..-: r mu... ,.,. n . ; 2.5.4.1? -th o .yln)‘ ‘ Au: 5.7.4; .\ «1 $34K? :13 .JJ? .3 , . . . I... n . 1 AU “3.. ... ...!n.(Lu. .122: «Efflurh; nitl.ilfi.l)..1\asf..ol . by... 6. Li . 31%.)..Q2 vr‘ .. : KI. I04! I: Ill»: (‘1‘ ‘ o, n .1. ... .. A . . ‘ , . g .5 rflvfifiu l, t . | III! II’ " ‘ carrifrdlli'F r 'w. ‘ I» ‘ME‘JD‘H " I MICHIGAN STATE UNNERSITY LIBR UHIHIN nu II III m "limit 3 1293 00609 6675 l l This is to certify that the dissertation entitled Effect of Lipid Oxidation on Functional and Nutritional Properties of Chicken Myofibrillar Proteins Stored at Different Water Activities presented by Hajar (Sheila) Noormarji has been accepted towards fulfillment of the requirements for Ph.D. degree in Food Science [WA/{yin ‘ Major professor Date January 12, 1990 MS U i: an Affirmative Action/Equal Opportunity limitation 0-1277 1 PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE MSU Is An Affirmative Action/Equal Opportunity lnetttmion EFFECT OF LIPID OXIDATION ON FUNCTIONAL AND NUTRITIONAL PROPERTIES OF CHICKEN MYOFIBRILLAR PROTEINS STORED AT DIFFERENT WATER ACTIVITIES BY HAJAR (SHEILA) NOORHARJI A DISSERTATION Submitted to Michigan State University In partial fulfillment of the requirements for the degree DOCTOR OF PHILOSOPHY Department of Food Science and Human Nutrition 1989 7 It1k KOCH} ABSTRACT EFFECT OF LIPID OXIDATION ON FUNCTIONAL AND NUTRITIONAL PROPERTIES OF CHICKEN MYOFIBRILLAR PROTEINS STORED AT DIFFERENT WATER ACTIVITIES BY HAJAR (SHEILA) NOORHARJI Effects of lipid oxidation, lipid protein interaction, water activity and storage on functional and nutritional properties of freeze-dried chicken myofibrillar proteins were investigated using a model system. The effects of freezing and freeze drying on. protein solubility' and lipid oxidation were studied using chicken myofibrillar proteins as a model system. ‘ Freezing and freeze drying decreased (P<305) protein solu- bility and increased (P<,05) lipid oxidation. These effects were more «drastic when Inethyl linoleate was added to myofibrillar protein at 15% of the dry protein weight. The effect of several different water activities on lipid oxidation and protein func- tionality in a freeze-dried chicken myofibrils (water activity environment of 0.43) was studied. It was demonstrated that the aw 0.43 had a protective effect, while the model system, aw 0.11 and aw 0.85 (below BET monolayer value and far above BET value) showed_ prooxidant effects. Moreover, control myofibrillar protein had a lower monolayer value than the lipid treated coun- terparts. Lipid oxidation increased (P(.05) and percent soluble pro- teins decreased (P(.0S) with storage time. Proteins stored at a water activity close to their monolayer value were less suscep- Hajar (Sheila) Noormarji tible to lipid oxidation. Water-holding capacity and gel strength were reduced (P{.05) by addition of lipid, storage and increase in water activity. On the other hand, neither in vitro digestibility nor the ability of myofibrillar proteins to support growth of Tetrahymena pyriformis were affected by addition of lipid, storage for 3 weeks or water activity. Thus, lipid oxida— tion may affect the functional properties rather than the nutri- tional quality of chicken myofibrillar proteins. To my husband, MEHDI, for his love, support, hard work and patience during my doctoral studies. Without him I could never reach my goal. To my children, MARYAM and HOOMAN, who have been the light of my eyes and the joy of spirit. To my parents, who taught me the real values of life and encouraged me to pursue my education, words are totally inadequate. ii ACKNOWLEDGEMENTS The greatest appreciation, which cannot be expressed in words alone, is given to Professor James F. Price, Chairman of my Doctoral committee for his guidance, patience and support. His kind consideration and understanding have been an incentive for the completion of this dissertation. Sincere gratitude is extended to Drs. Denise M. Smith, my co-advisor, for her valuable time, generous assistance and encouragement; Maurice R. Bennink, a member of my com- mittee, for his advise, guidance and constructive comments. Thank you very much. I am proud to have had as a guidance committee, Dr. Pericles Markakis and Dr. Cal Flegal, who freely gave me their time and interest. Appreciation is also expressed to my excellent family, especially Nahid and Nasrin, who gave me tremendous amount of love and support. I also wish to extend my appreciation to my friends, Carlos Lever, Abdalla Babiker, Abdulwahab Salih, Valente Alvarez and Debi Beeuwsaert, and my dear friend, Mr. Jack English, for their help. iii TABLE OF CONTENTS Page LIST OF TABLES........... .................... . ......... vi LIST OF FIGURES........................................ vii LIST OF APPENDICESOOOOO0.0...OOOOOOOOOOOOOOOOOOOOOOOIO. ix INTRODUCTION........................................... 1 LITERATURE REVIEW...................................... 4 Role of Proteins in Food Systems.................. 4 Muscle Proteins................................... 7 Myofibrillar Proteins............................. 8 Solubility and Myofibrillar Proteins.............. 9 Emulsification and Myofibrillar Proteins.......... 10 Gelation and Myofibrillar Proteins................ 13 Water-Holding Capacity and Myofibrillar Proteins.. 17 Lipid Oxidation................................... l9 Mechanism of Lipid Oxidation...................... l9 Lipid Oxidation in Muscle Foods................... 26 Measurement of Lipid Oxidation.................... 29 Antioxidants...................................... 34 Lipid-Protein Interaction......................... 36 Reaction of Amino Acids with Oxidized Lipids and Subsequent Effect on Nutritional Quality of Proteins .................................... 38 Methods to Measure Nutritional Quality of Proteins.. 39 Effect of Freezing and Frozen Storage on Protein Functionality................................ 41 Freeze-drying of Meat and Meat Products........... 43 Water Activity and Its Influence on Food Product Quality and Stability........................ 44 Moisture Sorption Isotherm........................ 46 EXPERIMENTAL PROCEDUREOOO..0.0.0...OOOOOOOOOOOOOOOOOOOO 51 Experimental Design............................... 51 Materials......................................... 52 Methods........................................... 54 Meat Sample Preparation...................... 54 Isolation of Myofibrillar Proteins........... 54 Model System................................. 55 Protein Solubility........................... 59 Gelation Preparation of Myofibril Gels....... 60 Water-holding Capacity....................... 60 Gel Strength................................. 61 Thiobarbituric Acid Test (TBA)............... 61 iv EXPERIMENTAL PROCEDURE (continued) Methods (Continued) In Vitro Protein Digestibility................ 6l Tetrahymena Bioassay of Protein Quality....... 62 Determination of Moisture Sorption Isotherms.. 63 Statistical Analysis.......................... 64 RESULTS AND DISCUSSION..0..0...OOOOOOOOOOOOOOOOIOOOOOO Effect of Freezing and Freeze Drying on Chicken MYOfibrj-llar Proteins..OOOOOOOOOOOOOOOOOOOOO Moisture Sorption Isotherms............ Protein Solubility..................... Lipid Oxidation........................ Water-Holding Capacity................. Heat-Induced Gelation...0....OOOOOOOOOOOOOOOOOOOO Effect of Lipid Oxidation on Nutritional Quality Of PrOtEin......... SUMMARY AND CONCLUSIONS...... APPENDICESOOOOOOOOOOOOO0.0... BIBLIOGRAPHY...0.00.00.00.00. .. 66 .. 72 .. 78 .. 83 .. 89 O. 99 ..106 O .168 O .122 Table Table Table Table Table Table Table Table Table Table Table 10. 11. LIST OF TABLES Page Typical functional properties performed by proteins in food systems.................... 5 Proximate composition of chicken meat.......... 23 Comparison of the saturated, mono-, di- and polyunsaturated fatty acid content in beef and in chicken white and dark meat triglycerides and phospholipid fractions....... 25 Unsaturated fatty acid content of lipids in some muscle fOOdSOOOOOOOOOOOIOOOOOOOOCOOOOO. 28 Relative humidities of saturated salt solutions at given temperature................. 58 Percentage solubility of chicken breast myofibrillar proteins during different stages of freeze drying........................ 67 Lipid oxidation of chicken breast proteins during different stages of freeze drying....... 69 Monolayer values (mo) of freeze-dried chicken breast myofibrillar proteins calculated using the BET equation.............. 77 Effect of methyl linoleate and water activity on water—holding capacity of rehydrated freeze-dried myofibrillar proteins....................................... 90 The main effect of sstorage period on the water-holding capacity of rehydrated freeze-dried myofibrillar proteins............. 92 Effect of methyl linoleate, water activity and storage time on gel strength of rehydrated freeze-dried chicken myofib- rillar proteins................................ 95 vi Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure 10. ll. 12. LIST OF FIGURES Page Mechanisms of lipid oxidation.......... ....... 21 Moisture isotherm curve....................... 48 Flow diagram of myofibrils extraction......... 56 Model system flow chart....................... 57 Moisture adsorption isotherm of the control and treated freeze-dried chicken myofibrillar proteins at 25°C................. 73 The BET monolayer plot of the freeze- dried chicken breast myofibrillar proteins at 250C.OOOOOOOIOOOOOOOOOOOOOO0...... 75 Effect of storage time and water activity on protein solubility of control and treated freeze-dried chicken myofibrillar proteins................................. ..... 79 Effect of methyl linoleate on protein solubility of control and treated freeze-dried chicken myofibrillar proteins...................................... 82 Effect of water activity and storage time on lipid oxidation of control freeze-dried chicken myofibrillar proteins...................................... 85 Effect of water activity and storage time on lipid oxidation of treated freeze-dried chicken myofibrillar proteins...................................... 86 Lipid oxidation of control and methyl linoleate-treated freeze—dried chicken myofibrillar proteins......................... 88 Effect of methyl linoleate and storage time on water-holding capacity of rehydrated freeze-dried chicken myo- fibrillar proteins............................ 93 vii Figure Figure Figure Figure Figure l3. 14. 15. l6. 17. LIST OF FIGURES (continued) Page Effect of methyl linoleate and storage time on gel strength of rehydrated freeze-dried chicken myofibrillar proteins.... 97 Digestibility of protein from casein, navy bean and adulterated casein... ..... ...... 100 Percent protein utilization by Tetra- hymena Pyroformis as affected by different protein sourceSIOOOOOOOOOOOOOOOOOOOOOOOOOO...O 102 Digestibility of protein as affected by treatment and water activity.................. 193 Percent protein utilization by Tetra- hymena Pyroformis as affected by treat- ment and water activities..................... 105 viii APPENDICES Appendix A. Table Table Table Table Table Table Table Table LIST OF APPENDICES Statistical Procedures A01. A02. Analysis of variances for protein solubility of freeze-dried chicken mYOfibrillar proteinSOOOOOOOOOOOOOCOOI... Analysis of variance for lipid oxidation of freeze-dried chicken mYOfibrillar proteiDSQQoooooooooooooooooo Analysis of variance for water- holding capacity of freeze-dried chicken myofibrillar proteins............ Analysis of variances for gelation of freeze-dried chicken myofibrillar proteins.-o.o.0000.00.00...0.000.000.0000 Summary statistic for Tetrahymena BioassaYOOOOOOOOOOOOOOOOOOIOOOOOOOOOOOOOO One way analysis of variance for Tetrahymena BioassaYOOOOOOOOO0.0.0.000... Statistical analysis of the linear regression for casein, casein + glucose, casein + NaOH and navy bean protein............................. T-value for differences in regres- Sion coeffiCientOOOOOOOOOOOOOOOOOOOOOOOOO ix Page 108 109 110 111 112 113 114 115 Appendix B. Tetrahymena Bioassay Procedure, Assay Medium and Figures. Tetrahymena Bioassay Procedure........... Maintenance of Tetrahymena Pyriformis W.. Figure 1 and 2. Figure 1. Figure 2. The pH vs time curves obtained by incubating casein, navy bean protein and adultrated casein with the multienzyme system........ The pH vs time curves obtained by incubating control and treated sample at different water activities with multi- enzyme system.................. Page 116 119 120 121 INTRODUCTION Proteins are one of the most important components determin- ing the functionality of food systems. The importance of pro- teins as functional components of food derives to a large extent from the following: 1. Proteins, along with polysaccharides, determine rheolog- ical properties of food which in turn determine food texture. 2. Proteins, along with polysaccharides, determine the water-holding properties of foods. 3. Proteins, often in combination with lipids, are the most important stabilizers of dispersed systems in foods. 4. Enzymes catalyze reactions that significantly affect food utility. The properties of proteins in food systems depend mostly on the interactions of proteins with other components, such as water and lipids. Most proteins in foods are closely associated with lipids, and lipids are important to physical and chemical charac- teristic of muscle foods. Lipids are prone to autoxidation. Damage to proteins due to reaction with oxidized lipid is an important deteriorative mechanism in the processing and storage of foods (Melton, 1983; Matoba et al., 1984; Sikorski, 1978). Exposure of proteins to peroxidizing lipids or their secondary products may lead to undesirable changes in nutritional, biochem- ical and functional properties. Oxidizing lipids cause protein 1 insolubization, polymerization, scission, poor digestibility, amino-acid and vitamin destruction (Funes et al., 1982; Funes and Karel, 1981; Pokorny, 1977; Nakhost and Karel, 1984). Function- ality is the most important property for proteins in food items. Some proteins cannot be used because they lack such functional properties as gelation, solubility, water absorbability, foam- ability or emulsifiability (Kinsella, 1982, 1976). There is an increasing interest in understanding the nature of lipid-protein interactions in biological and nonbiological systems. Since proteins play an important role in food systems, it is important to understand the effect of oxidized lipid-pro- tein interactions on protein functionality. It is important to adjust the conditions of processing and storage of meat products in order to minimize, or at least to regulate, the changes due to lipid—protein interaction. The purpose of this study was to examine the effect of lipid oxidation on functional and nutritional properties of myofibril- lar protein. The objectives of this research were: 1. To investigate the effect of lipid oxidation on protein functionality. 2. To determine how different conditions (water acti- vities, storage time) affect protein-lipid interaction. To investigate the effect of lipid-protein interactions on protein solubility, water—holding capacity and gelling properties. To determine the effect of lipid-protein interaction on nutritional quality of myofibrillar proteins. L ITERATURE REVI EW Role of Proteins in Food Systems The importance of proteins, peptides and amino acids in all living tissue should be evident. Proteins function as structural components, enzymes, regulatory agents and transport factors. Proteins are the principal functional and structural components of processed meats and determine the characteristic handling, texture and (appearance> of these products (Hermansson et al., 1986; Hermansson, 1985). Proteins are used tx> fabricate and facilitate the engineering of new foods. To facilitate their use in foods and their conversion to desirable ingredients, they must have appropriate functional properties. Functional properties can be defined as physiochemical properties of a protein which determine their utility in foods, including its nutritional value, organoleptic properties, safety and response to processing and storage (Kinsella, 1982). Physicochemical properties are derived from a protein's amino acid composition, amino acid sequence, secondary structure, tertiary structure, and quaternary structure (Pour-El, 1981). Physiochemical properties include surface charge, sulfhydryl content, hydrophobicity, molecular weight and conformational stability (Wilding et al., 1984; Kin- sella, 1982). Several typical classes of functional properties are listed in Table l. Table 1: Typical functional properties performed by proteins in food systems (Kinsella, 1982). Functional Mode of Action Food System Property Example Solubility Protein solvation Beverages Water absorption and binding Viscosity Gelation Cohesion-adhesion Elasticity Emulsification Fat absorption Flavor-binding Foaming Hydrogen-bonding of water, Entrapment of water Thickening, Water binding Protein matrix forma- tion and setting Protein acts as adhesive material Hydrophobic bonding in gluten, Disul— fide links in gels Formation and stabili- zation of fat emulsions Binding of free fat Adsorption, entrapment, release Form stable films to entrap gas Meats, Sausages, Breads, Cakes Soups, Gravies Meats, Curds, Cheese Meats, Sausages, Baked goods, Pasta products Meats, Bakery Sausages, Bologna, Soup, Cakes Meats, Sausages Donuts Simulated meats Bakery, etc. Whipped top- pings, Chif- fon desserts, Angel cakes Different food applications require different characteris- tics, e.g.; in beverages, protein should be soluble, and in com— minuted meats they should have emulsion stabilizing and gelling properties. In processed meat products, water binding, solubili- ty, emulsifying capacity, viscosity and gelation are typical properties of proteins that determine their impact on the final quality (Hermansson, 1985; Acton et al., 1983; Kinsella, 1982). As discussed earlier, the functional properties of proteins depend on such intrinsic physiochemical characteristics as amino acid composition and sequence, molecular weight, conformation and charge distribution on the molecules. The amount of hydrophobic amino acids affect conformation, hydration, solubility, gelation and denaturation (Kinsella, 1982). In general, the higher the polarity and the lower the hydrophobicity the higher the solubil- ity (Nakai, 1983). Polar amino acids contribute 1x) the water binding ability of a protein (Phillips and Beuchat, 1981). Myosin contains 38% polar amino acids with a large content of aspartic acid and glutamic acid residues which may bind 6-7 molecules of water each (Harrington, 1979). The gelation. and emulsifying ability of proteins are affected by their structure. Voutsinas et a1. (1983) showed a positive relation between gelation and hydrophobicity of the unfolded protein and sulfydryl content. Shimada and Matsumoto (1980) reported that proteins, e.g.; soy- bean, albumin, bovine serum containing 26-31% hydrophobic amino acids, formed gels upon heating, whereas proteins with 31% hydrophobic amino acids coagulatad. Kato and Nakai (1983) re- ported a close relationship between the hydrophobicity of pro- teins and their emulsifying capacity. Extrinsic factors which influence protein functionality include pH, ionic strength (Kinsella, 1982), specific ions, cooking temperature and frozen storage (Smith, 1988b; Whiting, 1988). Muscle Proteins Muscle proteins can be divided into three categories based on their solubility characteristics: 1) water soluble proteins, 2) salt soluble proteins and 3) insoluble proteins (Forrest et al., 1975). The water soluble proteins or sarcoplasmic proteins are also soluble in salt solutions of low ionic strength ( 0.1). These proteins compose about 30% of the total muscle protein or about 5% of the weight of muscle (Lawrie, 1979) and include oxidative, glycolytic and lysosomal enzymes, myoglobin and other water soluble proteins (Kramlich, 1978). About 100 different proteins are known to be present in the sarcoplasmic fraction (Scopes, 1970). The sarcoplasmic proteins influence functionali- ty, but differently than the myofibrillar proteins (Gillett, 1987a, 1987b; Sikorski et al., 1984). Myofibrillar or structural proteins which are soluble in concentrated salt solutions (ionic strength of 0.5 to 0.6) include myosin, actin, tropomyosin, troponin, actinins and others. Connective tissue proteins or insoluble proteins constitute 10-15% of the total muscle proteins ami are composed mostly of collagen (40-60%) and elastin (10- 20%). Myofibrillar Proteins The salt soluble proteins which compose the myofibrils within the muscle fibers are collectively defined as the myofi- brillar proteins. The major proteins in this category are myo- sin, actin and regulatory proteins, which make up about 55%, 25% and 20% of the fraction, respectively. The myofibrillar pro- teins, myosin in pre-rigor and actomyosin in post-rigor muscle are generally considered to contribute the most functionality to processed meat products (Smith, 1988a). Myosin is the major constituent of the thick filaments in the sarcomere. The myosin molecule is a large molecule consist- ing of two heavy chains with a molecular weight of about 205,000 daltons, and four light chains with an average molecular weight of 20,000 daltons. About 35% of the muscle proteins are myosin (Hanson and Lowey, 1964). Myosin has a high content of basic, acidic polar (70%) and sulfhydryl (3%) amino acids (Whiting, 1988). Myosin possesses adenosine triphosphatase (ATPase) enzy- matic activity, which transfers the chemical energy of ATP into the contractions of the muscle. Actin is the second most abundant protein in the contractile units. Actin forms the backbone of the thin filaments and ac- counts for 22% of the myofibrillar protein (Yates and Greaser, 1983). Actin can exist in two forms, as a monomer termed G-actin (globular), or as a polymer, termed F-actin (fibrous), depending Ir- on environmental conditions (Steiner et al., 1952). At physio- logical concentrations of salt, globular G-actin polymerizes to form F-actin which can interact with myosin filaments to produce mechanical energy for muscle contraction (Bandman, 1987; Pollard et al., 1981). Actin monomers (G—actins) are relatively small, having a molecular weight close to 46,000 dalton. Solubility and Myofibrillar Proteins Myofibrillar proteins are the most important functional components in meat, and their structure, conformation and inter- action affect the yield, juiciness and tenderness of food products (Kinsella, 1982). Their presence is necessary for fat binding, water binding and gel formation in meat products (Acton et al., 1983). The myofibrillar proteins must be solubilized in order to be functional (Kinsella, 1976). Salt serves many functions in meat systems, one of which is to solubilize myofibrillar proteins. The addition of salt (NaCl) to minced and communited meats causes solubilization and extraction of myofibrillar proteins. Salts also contribute to flavor, influence shelf life and affect the functional properties of muscle proteins (Olsen, 1982). Salt facilitates protein extraction from the meat tissue through a salting-in or solubilizing effect. The way in which salt ions bind to proteins is primarily electrostatic due to the attraction of the salt ions by the positively or negatively charged groups of the proteins (Schellman, 1953). Factors which influence the extractability of meat proteins are post-mortem age, pH, ionic 10 strength, temperature, and freezing (Bard, 1965). The extract- ability of muscle proteins is generally reduced during the early post—mortem period when pH is low and muscle temperature is still high (Borchert and Briskey, 1965). Comminution physically dis- rupts muscle tissue by damaging the sarcolema (Hamm, 1975), endomyosin (Wilding et al., 1984) and the integrity of muscle fibers. Comminution of muscle at ionic strengths above 0.6 causes swelling of muscle fibers (Wilding et al., 1984), depoly- merization of myosin, solubilization of myosin and extraction of myofibrils from the muscle fibers (Hamm, 1986). Hamm (1973) reported that the effect of pH on the protein solubility of a meat system is dependent on the presence of other factors in the system. One of the most important factors is the presence of salt. The presence of salt results in a lowering of the isoelectric point of the proteins, thus signifi— cantly increases the solubilization of protein. Meink et a1. (1972) studied the relationship between protein solubility, salt concentration and pH. Their data indicated that myofibrillar protein-solubility increased with increasing salt concentration at approximately pH 6. Protein solubility has been linked to functionality by many researchers and is used as an index of functionality (Kinsella, 1976). Emulsification and Myofibrillar Proteins When frankfurters, hot dogs or bologna are manufactured, the meats are extensively chopped (comminuted) to produce small 11 particles. One can observe a heterogeneous mass of lean and fat transformed into a meat batter of homogeneous appearance. Meat batters are fluid and primarily composed of water, fat and pro- tein. One key element in manufacturing these products is stabi- lizing the fat and moisture to prevent excessive losses or product failure. The meat proteins stabilize the fat, therefore, meat batters have historically been defined as meat emulsions (Foegeding, 1988). Meat batters are considered an oil-in-water emulsion and stability of this emulsion depends on the meat protein behavior. In all of the emulsion products, the type of meat protein used is important. Salt soluble myofibrillar proteins, actin, myosin and actomyosin are primarily responsible for emulsion formation and are recognized as the important emul- sifiers and stabilizers in meat systems (Gaska and Regenstein, 1982). The physiochemical properties of myosin which may allow it to function as an emulsifier include: 1) a hydrophobic region which orients toward the fat globule, 2) a hydrophilic region which orients toward the continuous matrix and 3) molecular flexibility for unfolding at the interface to lower surface tension (Jones, 1984). Nakai et a1. (1986) reported a good correlation between emulsification properties of salt extracted meat proteins and physiochemical protein properties of surface hydrophobicity, sulfhydryl group content, and solubility. Li- Chan et a1. (1985) stated that solubility by itself is not a good predictor of emulsifying properties. They reported that both protein hydrophobicity and solubility were important parameters 12 affecting the' emulsifying properties of :meat proteins. The sarcoplasmic proteins play a small role and are in fact poor participants in meat batter systems, possibly because they are not able to form a gel structure and/or they interfere with the gel (matrix) structure of myofibrillar proteins (Gaska and Regen- stein, 1982). Hegarty et a1. (1963) reported that functionability of water soluble proteins, in the presence of salt, were enhanced, and the protein was capable of considerable emulsion stabilization at the pH of fresh meat, 5.6 - 5.8. However, water soluble and salt soluble proteins reacted differently. The emulsification capaci- ty of sarcoplasmic proteins increased with lower pH, but de- creases with lower pH in the salt soluble fraction. This sup- ports the work of Swift and Sulzbacher (1963) which showed that peak emulsification capacity was dependent upon both pH and salt concentration for both protein types. As the salt concentration increases, the emulsifying capacity also increases, due 1x) the fact that salt serves to solubilize the myofibrillar proteins (Cunningham and Froning, 1972). A great deal of work has been done concerning the effect of pH on the properties of muscle proteins (Froning and Janky, 1971; Froning and Neelakantan, 1971; Hwang and Carpenter, 1975; Swift and Sulzbacher, 1963). These researchers have all reported that as the pH of the system is moved away from the isoelectric point of muscle proteins, the emulsification properties of the protein are increased. This is probably due to improved solubility of the proteins at these pH levels. 13 The emulsifying capacity can vary among different types of poultry meat. May and Hudspetch (1966) reported that in various classes of poultry the amount of total protein that was salt soluble was greatest in hen white meat (40.67%), followed by broiler dark meat, turkey dark meat and hen dark meat (16.67%). However, they stated that in all classes of poultry, the dark meat proteins display greater emulsification capacity than did the white muscle proteins. McCready and Cunningham (1971) indi- cated that although dark meat was lower in total protein and salt soluble protein, its ability to emulsify oil was greater than that of broiler light meat, which was higher in total and salt soluble protein content. They stated that the pH of dark meat was higher than that of light meat and suggested that pH was more important to emulsification capacity than was the percentage of salt soluble protein in the meat tissue. Gelation and Myofibrillar Proteins In manufacturing of processed meat products, gelation, water holding and fat binding are the most important functional proper- ties that influence product quality. In meat products that are chopped when raw, the heat processing transforms the highly viscous raw material into a solid, gel-like product. Muscle proteins denature and aggregate to form the gel matrix, hence the gelation properties of muscle proteins, particularly myosin and actomyosin, are important in determining their functional role in processed meats. 14 The presence of salt-soluble myofibrillar proteins has been shown to be necessary for binding in both emulsion and restruc- tured meat products (Miller et al., 1980). Among the myofibril- lar proteins, myosin is essential for gel formation. Myosin is one of the most important functional proteins in meat, and its states of aggregation have an impact on functional properties such as texture, fat and water holding of meat products (Acton et al., 1983). Actin does not form a gel, but coagulates on heating (Samejima et al., 1969). Nakayama and Sato (1971a, 1971b) stated that myosin and actomyosin were the proteins that produce the greatest gel strength, and therefore, were the most important in binding. Yasui et a1. (1980) by using a model gelation system reported that the addition of myosin to actomyosin produced a gel that was much stronger than either myosin or actomyosin used separately. Gelation, by the myofibrillar proteins, primarily' myosin (pre-rigor) and actomyosin (post-rigor), is a: heat-induced pro- tein to protein interaction that leads to the formation of a three dimensional, well-ordered protein structural matrix (Acton and Dick, 1986; Hermansson et al., 1986). It is a two—step process which involves an initial heat denaturation of pmotein followed by the formation of three dimensional network of fibrous protein (Ferry, 1948). Native protein.——————9 Denatured protein.——4,Aggregated protein (gel matrix) 15 Thermal energy is the most important driving force in pro- tein transition from the native state to the denatured state. Thermal transition temperatures represent points at which confor- mational changes (denaturation) occur in native structure of protein during heating. The sol to gel transformation results in formation of a three dimensional network produced from protein -- protein interaction (Samejima et al., 1969; Ishioroshi et al., 1979; Siegel and Schmidt, 1979). Meat and poultry muscle proteins undergo three thermal transitions at about 55 - 60°C, 65-67°C and 80 - 83°C, depending on the species and test conditions (Xiong et al., 1987). Ishioroshi et a1. (1979) reported that the transi- tion from sol to gel by myosin begins at 30°C and reaches a maximum of gel rigidity at 60°C. Quinn et a1. (1980), using differential scannimg calorimetry, showed that denaturation of meat (beef) proteins begins at about 50°C and continues with increasing temperature up to 9a°cu Grabowska and Sikorski (1976), using fish myofibrils, reported that the increase in gel strength started at 30°C and continued up to temperature of 80°C. The ability of a gel to exhibit viscosity, rigidity and elasticity seems to be a function of the types of protein, the temperature and time of heating, protein concentration, pH and ionic strength (Asghar et al., 1985; Acton and Dick, 1984; Hick- son et al., 1980; Smith, 1988a). Smith et a1. (1988) developed a generalized mathematical model to predict the combined effects of pH, protein concentration, processing time and endpoint cooking 16 temperature on the strength of chicken myofibril gels. Myosin formed an ordered, fine lacy strand network in 0.25M potassium chloride (KCL), pH 6.0, but a disordered, coarse aggregated or sponge-like gel structure was formed in 0.6M KCL, pH 6.0 (Her- mansson et al., 1986). Intact heavy chains of myosin are neces- sary to obtain maximum gel strength, because the light chains are solubilized during heating (Samejima et al., 1984). Maximum gel strength in 0.6M KCL, pH 6.0, has been reported to occur at a free myosin-:-F-actin mole ratio of 27:1 which is equal to a weight ratio of 15:1 (Yasui et al., 1980; 1982). At this ratio, about 20% of the protein was actomyosin and 80% was free myosin. Gel strength is at a maximum in myofibrillar protein preparation prepared at pH 5.5 to 6.0 in 0.5 to 0.8 M salt solution (Yasui et al., 1980; 1982). In her review (Smith, 1988a) points out the following: "Several researchers have observed differences in fat and water- binding ability and textural properties between light and dark meat when used in further processed products (Froning and Norman, 1966). Chicken white muscle myosin (Asghar et al., 1984), chick- en white muscle actomyosin (Brekke et al., 1987) and beef white muscle myosin (Fretheim et al., 1986) exhibited greater gel strength than the corresponding protein from red muscle. White muscle myosin from beef had higher water—holding capacity and solubility below pH 5.7 than red muscle myosin (Fretheim et al., 1986). Acton and Dick (1986) reported that broiler breast acto- myosin had a lower transition temperature than thigh actomyosin during heat-induced aggregation." 17 Water-holding Capacity and Myofibrillar Proteins The ability of meat, particularly processed meat, to retain water is called water-holding capacity (WHC), (Hamm, 1960). Water-holding capacity is an important factor in determining meat quality and acceptability because of its close relation to taste, color, tenderness and juiciness (Lawrie, 1979; Forrest et al., 1975). Water is held by protein in two forms: 1) free water and 2) bound water. The free form or biologically active form of water is that water which is held by surface forces (Fennema, 1976; Forrest et al., 1975). The bound water also known as the structural or protective form is tightly bound as water of hydra- tion by functional groups of the protein in the form of mono- and multimolecular layers (Wismer-Pedersen, 1978). About 4% of muscle water exists in a bound form, being bound to the hydro- philic groups on proteins (Lawrie, 1979; Forrest et al., 1975). The bound water of hydration in meat is not readily released except under severe conditions such as protein denaturation, rigor mortis and change in muscle pH. Myofibrillar proteins play an important role in water— holding capacity of meat products. Generally as the amount of soluble protein increases, water-holding capacity also increases. Honiked et a1. (1981) demonstrated that. a close relationship between the change in solubility of myofibrillar proteins induced by postmortem metabolism and WHC of salted beef tissue homoge- nates. Cook (1967) using a filter paper moisture» absorption 18 technique found a highly significant correlation (r = —0.51) when comparing salt-soluble nitrogen and moisture released by press- ing. .All of these studies indicate that myofibrillar proteins are required for binding and water-holding capacity. Water-holding capacity can be affected by many factors such as pH, species of animal, sex, age and salt concentration (Hamm, 1960). The effect of pH on WHC is well known (Hamm, 1986; Hamm, 1960). As pH values are increased away from the isoelectric point of proteins, water-holding capacity and protein solubility increases (Honikel et al., 1981). Kaufman et a1. (1986) stated that the water—holding capacity is increased at pH levels consid- erably above or below the isoelectric point of a muscle. With the reference to animal species, pork meat has the highest water- holding capacity followed by beef and poultry. A number of different methods are used to measure water- holding capacity. INK) major types of water-holding capacity methods, water binding potential (WBP) and expressible moisture (EM) respond to different properties of flesh. Water binding potential (WBP) refers to the ability of a protein system to hold water present in the system and under the influence of an exter- nal force. It, therefore, represents potential maximum water retention of a protein system under the measurement condition (Regenstein et al., 1979). Expressible moisture refers to the amount of liquid squeezed from a protein system by the application of force, and measures the amount of loose water released under the measurement condi- tions (Regenstein, 1984). 19 LIPID OXIDATION Lipid oxidation is one of the major causes of deterioration in the quality of meat and meat products (Asghar et al., 1988). It can result in reduction of quality, nutritional value and safety of foods (Frankel, 1984; Pearson et al., 1983). The important fatty acids involved in oxidation are unsatu- rated oleic, linoleic and linolenic acid (Pearson et al., 1983). Younathan and Watts (1960) and Pearson et a1. (1977) reported that chicken fat is high in oleic and linoleic fatty acids; and because of this unsaturated nature, poultry meat is more suscep- tible to rancidity than the red meats. Labuza (1971) reported that the rate of oxidation increases geometrically with the degree of unsaturation. Generally speaking, the higher the proportion and degree of unsaturation of fatty acids, the more labile the lipid system is to oxidation (Dawson and Gartner, 1983). Mechanism pf Lipid Oxidation The oxidation of fatty acids proceeds by a free radical chain reaction mechanism which involves three stages, as shown in the following scheme (Frankel, 1984; Khayat and Schwall, 1983; Dugan, 1976; Labuza, 1971; Lundberg, 1962): 20 Initiation: RH initiator(s) ’ R. + H. RH + 0 ,R00. + H. Propagation: R. + O ’ R00. R00. + RH ) ROOH + R. Termination: ROO.R + R. >ROOR R. + R. ’ R-R ROO. + ROO. A ROOR + o I Where RH is an unsaturated fatty acid, R. is an alkylradical, R00. is a lipid peroxyradical and ROOH is a hydroperoxide. Initiation takes place when a labile hydrogen is removed from a carbon atom adjacent to a double bomfl in an unsaturated fatty acid with the formation of a free radical. Once initiated, the reduction is propagated by the level of hydroperoxides pro- duced due to their ability to decompose to free radicals. Termi- nation begins when the concentration of free radical is suffi- ciently high to begin interacting to form non-free radical products. Free radical inhibitors include antioxidants that may form inert end products as a termination step. Fig. 1 shows the overall mechanism of lipid oxidation. Lipid 'hydroperoxides are the first relatively stable in- termediates that are colorless, odorless and do not contribute to the off-flavor associated with lipid oxidation (Watts, 1954; Sato et al., 1973). Hydroperoxides are quite stable at low tempera- tures, however, they can be thermally or catalytically decomposed 21 Fig. 1 - Mechanisms of Lipid Oxidation Unsaturated Fatty Acids -—) Hydroperoxides 1 R00' + R' l ROOH + R’ ~V Breakdown Products off-flavor compounds hydrocarbons alcohols, ketones aldehydes, acids and Triglycerides Free Radicals RH R- + H- ——-—> RH + O : ROO + H -——fi-oxidation of pigments, \ I flavors and vitamins »——? Insolubilization of proteins v Polymerization dark color may be toxic H20§—~ epoxides, OH-glycerides Di-OH-glycerides I ! l 1 I O keto-glycerides‘;~_~__~‘§‘~J 22 leading to the formation of short chain aldehydes, ketones, alcohols, lactones, acids and unsaturated hydrocarbons (Labuza, 1971; Frankel, 1984). These secondary products of lipid oxida— tion are responsible for the off—flavor of oxidative rancidity in red meats, poultry and fish (Sato et al., 1973; Pearson et al., 1983). Malonaldehyde was reported by Frankel (1985) and Igene et al. (1985b) to be the most important breakdown product of lipid oxidation. Factors such as light, heat, oxygen, fatty acid composition, radiation and catalysts are important initiators of lipid oxida- tion (Khayat and Schwall, 1983; Labuza, 1971). The content and composition of muscle lipids differ within an animal and influ- ence oxidation potential (Dawson and Gartner, 1983). The varia- bility in content and composition of muscle lipids depend upon the muscle function (Allen and Foegeding, 1981). Turkey breast muscle contains about half as much total lipid as thigh muscle (Wilson, 1974). Table 2 shows the proximate composition of chicken meat indicating the considerable variability in fat composition. Total saturated, mono-, di-, anui polyunsaturated fatty acids in beef and in both chicken white and dark meat triglycerides and phospholipids are presented in Table 3. These data demonstrate that the phospholipid fraction contains about 15 fold higher levels of polyunsaturated fatty acids than the triglyceride fraction. Thus, the phospholipids are responsible for the susceptibility of the membranes to oxidation during cooking, flaking or other processes that cause membrane disrup- tion. Phospholipids are much more unsaturated than triglycerides 23 Table 2 - PROXIMATE COMPOSITION of chicken meat (raw) Types of Meat Composition Breast Thigh Moisture (%) 74.30 75.48 Fat (%) 1.63 3.61 Protein (%) 22.20 18.74 Ash (%) 1.10 1.04 Source: Data adapted from USDA, 1979. 24 in beef and in white and dark chicken meat (Igene et al., 1981). There are virtually no polyunsaturates in beef triglycerides and only 1.4% in chicken dark meat triglycerides. In contrast, polyunsaturates make up 15.5% of beef phospholipids and 22.7% of chicken dark meat phospholipids. Since most phospholipids are found in the membranes, they become exposed to oxygen and subject to oxidation during any processimg step that disrupts membrane integrity (Pearson et al., 1977). Hemoproteins (Tappel, 1955) and free transition metals (Wills, 1965) are powerful oxidation catalysts of unsaturated fatty acids in muscle and model systems. Hemoproteins have been reported to be the major catalysts of lipid oxidation in beef, chicken, turkey, fish (Tappel, 1952; Lee and Toledo, 1977). Igene et al. (1985a), Sato et a1. and Love and Pearson (1974) conclude that non-heme iron was the principle catalyst of lipid oxidation in cooked meat. Tappel (1955) reported that hematin compounds catalyze the oxidation of unsaturated fatty acids, and that iron is the active factor in catalytic activity. The ferric form of heme is the active catalyst of lipid oxidation in muscle (Younathan and Watts, 1960). In a 1975 review, Green and Price concluded either Fe2+ or Fe3+ hemes might function as catalysts of lipid oxidation; but the Fe3+ hemes may be necessary for rapid catalysis. 25 «>urm u . nosu-wmwo: 0* n:- m-ncwunon. 3036.. on.. can vopn.n nosnosn - as none can .3 nsmnwos csmno can coax :n-n damn—«now‘ao can vsonvso—*umn uncommon» .q turkey > chicken > pork > beef > lamb. Salih et al. (1989a) and Salih et al. (1989b) reported that lipid oxidation is higher in dark meat than light meat of chicken and turkey. 28 Table 4 - Unsaturated Fatty Acid Content of Lipids in Some Muscle Foods Melton, 1983 Content (%) Fatty acid Lamb Beef Pork Chicken Fish C18:1 9.51 33.44 12.78 20.25 19.59 C18:2 18.49 10.52 35.08 14.20 5.88 C18:3 0.43 1.66 0.33 0.90 8.07 C20:2 0.34 0.69 ---- ---- 0.20 C20:3 0.62 2.77 1.31 1.30 0.36 C20:4 13.20 8.51 9.51 11.60 3.75 C20:5 ---- 0.76 1.31 1.55 7.16 C22:4 ---- 0.88 0.98 2.10 0.65 C22:5 -—-- 0.92 2.30 5.75 2.39 C22:6 ---- ---- 2.30 5.75 2.39 29 Measurement pf Lipid Oxidation Many techniques, ranging from sensory evaluation to chemical and physical methods, are available for assessing the extent of oxidation in lipid-containing foods. However, the most widely used test for measuring the extent of oxidative deterioration of lipids in fat containing foods, especially muscle foods, is the 2-thiobarbituric acid test or TBA test (Melton, 1983; Gray, 1978; Rhee, 1978). The TBA test is a colorimetric analytical technique in which the absorbance of a pink chromogen formed between TBA and malonaldehyde (MA) is measured (Gray, 1978; Tarladgis et al., 1960). The red pigment obtained in the reaction occurs as a consequence of the condensation of 2 mole of TBA with 1 mole of MA (Sinnhuber and Yu, 1958). The intensity of color is a measure of malonaldehyde concentration (Tarladgis et al., 1960, 1964) and has been correlated organoleptically with rancidity (Zisper et al., 1964). However, the TBA procedure should be used mainly to assess the extent of lipid oxidation rather than to quantitate malonaldehyde, as malonaldehyde may only contribute a part of the total color complex. Kakuda et a1. (1981) suggested that the TBA reagents may react with a variety of compounds, other than malon- aldehyde, present in oxidized foods and thus lead to production of various colored compounds. Jacobson et a1. (1964) has shown that other products of lipid oxidation, such as alka-2,4-dienals, also react with TBA to form a red complex with the same absorp- tion maximum (532 nm) as the malonaldehyde-TBA complex. However, 30 Igene et al. (1985b) reported that the major TBA—reactive sub- stances in the distillate of cooked chicken was malonaldehyde. Kosugi and Kikugawa (1985) reported that there were several kinds of TBA reactive substances (TBARs) in the autoxidized chicken fat and methyl linoleate: the TBARs at 532 nm which were unstable in autoxidation process and slowly liberated malonaldehyde; and TBA-reactive substances at 455 nm which liberate other aldehydyes unstable after reaction with TBA. The result of the TBA test is expressed as the malonaldehyde content of foods in mg/kg of sample, or TBA.number. The TBA test can be performed on: 1) the whole food followed by extraction of the red pigment formed (Sinnhuber and Yu, 1958), 2) distillate of the food (Tarladgis et al., 1960) and 3) extract of the food (Witte et al., 1970). The distillation method is the most popular one of the three methods (Rhee, 1978), but that fact does not necessarily mean that it is the most accurate or repro- ducible method. Salih et a1. (1987) reported that TBA values determined by the distillation method were twice as large as those determined by extraction. High correlations between the two methods were observed by these authors. M However, several possibilities for the differences between the two TBA methods were suggested by Witte et al. (1970). The heat of distillation may increase the quantities of aldehyde and disrupt certain carbonyl compounds formed by reaction between malonaldehyde and amino acids or proteins (Buttkus, 1967). Heat during distillation is used to free malonaldehyde from its bound state with protein, whereas per chloric acid is used to release 31 malonaldehyde in the extraction method. Heat used in distilla- tion may also speed up the oxidation process (Witte et al., 1970). In the extraction method, the filtration may have given incomplete extraction of malonaldehyde, since no heat was in- volved in the method. The TBA test has been used by many re- searchers to follow lipid oxidation in cooked beef, pork and poultry (Melton, 1983; Huang and Greene, 1978; Igene et al., 1979), during refrigerated and frozen storage of beef, pork and poultry (Drerup et al., 1981), in freeze-dried beef and pork (Chipault and Hawkins, 1971) and in fish (Lee and Toledo, 1977) as well as to study lipid deterioration in relation to warmed- over flavor of beef and poultry (Greene and Cumuze, 1981; Igene et al., 1980; Sato and Hegarty, 1971; Sato et al., 1973; Wilson et al., 1976). Kakuda et a1. (1981) used a high performance liquid chromat- ographic technique (HPLC) to measure the extent of lipid oxida- tion in the distillate from freeze-dried chicken samples. These investigators reported that there was a linear relationship with a simple correlation coefficient of 0.946 between the TBA absorb- ance at 532 nm and HPLC peak height of the MA. Csallany et a1. (1984) also used a HPLC method to measure free malonaldehyde in rat liver and beef, pork and chicken muscle. They found that the malonaldehyde level measured by the TBA assay' generally were four- to five-fold higher than those estimated by the HPLC meth- od. 32 The determination of peroxides to assess lipid oxidation in muscle foods has been applied by many researchers. Melton (1983) has reviewed the application of the peroxide assay as a measure of lipid oxidation in pork, beef and poultry products. She reported that the peroxide assay is capable of detecting lipid oxidation in muscle foods, although it may not be useful for ground meat samples stored for prolonged periods. Apparently, the relationship of peroxide value (PV) to oxidized flavor in meat systems varies with the type of meat and the manner in which it has been processed. Jermiah (1980) used PV to investigate the lipid oxidation in frozen pork in different types of packaging wrap. He concluded that PV increased for up to 140 days of frozen storage for fresh pork cuts, but only up to 56 days for cured meat products. Sinnhuber and Yu (1977) reported a: linear relationship between peroxide value and malonaldehyde concentrations during oxidation of various classes of polyunsaturated acids. Significant (P(.05) correlation coefficients of r= -0.57 and r = -0.51 were found between TBA values and sensory scores for beef and chicken white meat model systems, respectively (Igene and Pearson, 1979; Salih et al., 1987). Igene et al. (1985b) studied the relationship between TBA numbers and panel scores for WOF in cooked chicken white meat and dark meat with and without chelators and/or antioxidants. They reported a correlation coefficient of 0.87 between the two measurements, which was statistically significant (P<.01). This showed that the TBA numbers were closely related to warmed-over flavor (WOF) panel 33 scores. Turner et a1. (1954) observed that pork patties made with ground raw pork having a TBA number of 0.46 were judged to be of "borderline" quality, while those with a TBA value over 1.20 were found unacceptable by a test panel. Among the other methods for following lipid oxidation in foods is to quantitate the carbonyl compounds formed by the degradation of lipid hydroperoxides (Gray, 1978). One of the most sensitive chemical assays for lipid hydrop- eroxides that is of use to biochemists is the measurement of conjugated dienes. In the formation of a: hydroperoxide from a: 1,4 non—conjugated diene during autoxidation, the diene moiety rearranges into conjugation. Four isomeric conjugated diene hydroperoxides can be formed on rearrangement. Chan and Levett (1977a, b) used HPLC to separate the mixture of four isomers that are formed in the air oxidation of methyl linoleate. Both cis- trans and trans—trans isomers were isolated and showed absorption maxima at 236 and 223 nm. A typical procedure to measure conjugated dienes would be to extract the lipids from the samples to be tested using 2:1 (v/v) chloroform/ methanol and evaporate this extract to dryness under a stream of nitrogen at ambient temperature (Buege and Aust, 1985). The dried extract is then redissolved in hexane or cyclo- hexane of spectroscopic quality, and its absorbance at 234 nm measured against the solvent blank or by difference spectroscopy versus nonperoxidized lipid. 34 This sensitive assay can be used to measure lipid peroxida- tion both in vivo and in vitro. When studying model systems comprised of aqueous suspensions of purified lipid, it is often possible to analyze for conjugated dienes directly without the need to extract the lipid into organic solution (Pryor et al., 1976). Antioxidants Antioxidants are substances capable of slowing the rate of oxidation in fat or fat containing foods. Their action may be due to donation of hydrogen or an electron which reacts with free radicals to form inert products terminating the chain reaction mechanisms (Anonymous, 1986). Labuza (1971) divided the antioxi- dants into three categories: Type I: free radical chain termi- nators, which donate hydrogen to the free radical and thus stop the chain reaction. This group is composed primarily of phenol- ic-type compounds, like butylated hydroxyanside (BHA), butylated hydroxytoluene (BHT) and tocopherol; Type II: radical production preventers, mostLy water soluble chelating agents, like citric acid, ascorbic acid and ethylenediaminetetraceticacid (EDTA); and Type III: environmental factors such as temperature, water activity regulators and packaging materials which affect the partial pressure of 02 and affect the rate of oxidative reac- tions. Desirable features of an antioxidant are that they (1) be effective at low concentration, (2) do not react with food compo- 35 nents, (3) be safe and inexpensive and (4) tunme a good carry- through property. The term carry-through refers to the ability to survive baking and frying operation. Pearson and Gray (1983) pointed out that antioxidants are of two types, naturally occur- ring and synthetic ones. Application of various antioxidants to meat and other food systems have been investigated by several researchers. The syn- thetic phenolic antioxidants, BHA and BHT, have been widely used in meat systems and in general have been shown to be effective in retarding lipid oxidation. Butylated hydroxyanisol has proven to be effective in a number of foods. A combination of antioxidants is frequently used in fat containing foods. Numerous combinations of BHA, BHT, n-Propyl- gallate and citric acid are available commercially for use in fat and fatty foods. Chen et al. (1984b) reported that Tenox 4, which contains BHA, citric acid and propylene glycol, coated on salt was an effective inhibitor of lipid oxidation. Pikul et a1. (1984) also have shown that addition of BHT to cmicken samples during sample homogenization, distillation or extraction step of the TBA assay prevent sample autoxidation. Smith (1987) reported that Tenox II which contains BHA, PG and citric acid prevented lipid oxidation and subsequent oxidized-lipid-protein interaction which caused myofibrillar protein denaturation during frozen storage. Acidic compounds such as citric acid, ascorbic acid and EDTA delay the onset of oxidative rancidity by inactivation of proxi- dant metals, and are often used in combination with phenolic 36 antioxidants. These chelating agents exhibit synergistic effects which increase the effectiveness of many antioxidants (Anonymous, 1986; Dugan, 1960). The naturally occurring antioxidants also can be added to foods to retard oxidation. Plant or vegetable proteins are known to contain natural antioxidants. Rhee and Ziprin (1981) studied the effectiveness of various oil seed proteins in retarding lipid oxidation in stored cooked meats when they are used as an ingredient of gravy or sauce for these products. YOunathan et a1. (1980) also showed that rancidity in turkey was effectively controlled by hot-water extracts of eggplant tissue and peelings of yellow onions and potatoes. Lipid-Protein Interactions One of the most common concerns associated with muscle lipids is their lack of stability to oxidation and associated effects on meat odor, flavor and the functionality of the muscle proteins. Oxidation of unsaturated fatty acids leads to the formation of hydroperoxides and their secondary breakdown products, such as carbonyl compounds, acids, etc. Exposure of proteins to peroxidizing lipids or their secondary breakdown products can produce changes in proteins including insolubiliza- tion, polymerization, decomposition to lower molecular weight products, browning, production of toxic compound, poor digesti- bility and damage to specific amino acid residues and formation of lipid—protein complexes (Nakhost and Karel, 1983, 1984; Funes and Karel, 1981; Funes et al., 1982; Matoba et a1, 1984; Yanagita 37 et al., 1973; Shimada and Matsushita, 1978; Pokorny, 1977). These chemical changes contribute to the deterioration of food proteins during processing, storage and cooking. Proteins can react with either the free radicals, hydroperoxides or secondary products of lipid oxidation (Schaich and Karel, 1975). Lipid-protein interactions alter the fUnctional properties of meat. Loss of solubility due to protein-lipid interaction may be caused by the formation of lipid-protein complexes, by protein aggregation (H: by reactive) decomposition products of lipids. Polymerization can occur by protein-protein and protein-lipid crosslinking (Roubal and Tappel, 1966). The most widely reported mechanism involves reaction of proteins with malonaldehyde. Gamage et a1. (1973) have shown that malonaldehyde can react with proteins through free radical and nonradical mechanisms to form intra- and inter-molecular cross-linked products via Schiff base formation. In nonradical reaction, malonaldehyde reacts with the amino group of histidine, arginine and the epsilon-amino group of lysine in protein to form polymerized products (Shin et al., 1972). The reaction of malonaldehyde with protein is not the only process which can lead to denaturation. Olley and Duncan (1965) reported that proteins in solution have a strong affinity for fatty acids and detergents. The effect of fatty acids and detergents on proteins is complex and can lead to precipitation, disaggregation as well as to stabilization. The interactions between protein and lipid can occur by hydrogen bonding, ionic attraction or hydrophobic interaction 38 (Karel, 1973). And bonds can be affected by factors such as temperature, water activity, presence of catalysts, storage time and reactant concentration (Nielsen et al., 1985b; Funes et al., 1982). Reaction gf Amino Acids with Oxidized Lipids and Subsequent Effect 22 Nutritional Quality 2; Proteins. Most foods are cooked, industrially-processed or stored before consumption. During such treatments protein can react with other food components or chemical additives (Hurrell, 1984). These reactions can lead to flavor and color formation, loss of nutritional value and occasionally to potentially toxic com- pounds. For many years, nutritional research has concentrated on the stability of amino acids during food processing and storage. There have been several publications concerning the damage to amino acids by oxidized lipids. Many investigators have studied the reactions of proteins with oxidizing lipids and other food components such as reducing sugars and food additives. The reaction of proteins with oxidizing lipids may occur during the storage of cereal and fish products (Harrison et al., 1976; Khayat and Schwall, 1983) leading to serious losses in digesti- bility and bioavailability of some essential amino acids (Nielson et al., 1985a). Due to the presence of reducing sugars in some foods, Maillard reactions have been reported to occur in such foods (Hurrell and Finot, 1983). Food additives (alkaline mate- rials) which are used to improve protein solubility or other physical characteristics (Tannenbawm et al., 1970; Provansal et al., 1975) may also react with food pmoteins. These reactions 39 are also reported to cause changes in protein digestibility and bioavailability of some essential amino acids (Nielsen et ad., 1985a). These changes in turn may lead to severe losses in the quality of proteins. Modification of amino acid side chains can result from lipid-protein interaction. The main sites of peroxide reactions are in the sulfur containing amino acids, cysteine and methio- nine. Cysteine can be oxidized or undergo additional reactions with malonaldehyde and hydroperoxides (Gardner et al., 1977). Methionine can undergo oxidation to methionine sulfoxide, or it can react with a carbonyl compound in a Maillard reaction (Tuft amd Wartheson, 1979). The epsilon-amino group of lysine forms Schiff's base condensation products with aldehydes during reac- tions with reducing sugars (Mauron, 1981). Histidine is also susceptible to reaction with lipid hydroperoxides (Matoba et al., 1984). Since methionine and lysine are the most limiting amino acids in certain foods, their oxidation can affect the nutritive value of foods. A few reports have indicated, however, that degradation of tryptophan during food processing and storage could be of nutritional importance (Kanazawa et al., 1975). These authors found large losses of tryptophan (n1 reaction of protein with oxidizing lipids. Tryptophan has also been shown to be sensitive to oxidation (Nielsen et al, 1985a). Methods 32 Measure Nutritional Quality gf Proteins. Even though the amino acid profile is important in evaluat- ing the nutritive quality of a protein, the digestibility of that 40 protein is also important for determination of protein quality. The digestibility and the amino acid profile both can be measured by using rat bioassay, but this is an expensive and time consum- ing procedure. Several in vitro methods for the measurement of protein digestibility have been developed. Akeson and Stahmann (1964) found that a pepsin—pancreatic enzyme system gave a reasonably accurate approximation of protein digestibility. Saunders et a1. (1973) developed a papain-trypsin system, which correlated well with in vivo digestibility (r = 0.91). Hsu et a1. (1977) de- veloped a multienzyme system consisting of trypsin, chymotrypsin and peptidase which correlated well with in vivo digestibility (r = 0.90) also. The concept of biological availability as applied to amino acids in food proteins has been studied and a number of biologi- cal, chemical, microbiological and enzymic methods for determin- ing amino acid availability have been proposed. Numerous in vivo and in vitro methods for determining amino acid availability have been evaluated. One such method was described by Stott and Smith (1966) who developed a pmocedure, based on the method of Fernell and Rosen (1956), which uses Tetrahymena pyriformis W for measuring the availability of ly- sine, methionine, arginine and histidine in intact protein sources. Nutritional labeling requirements are concerned with nutri- ent retention during storage, and the effect of processing on the 41 nutritive quality of protein products have created a need for an assay method to measure the quality of protein which is less time consuming and less expensive than official bioassay for protein efficiency ratio (PER) (AOAC, 1970). The Tetrahymena pyriformis bioassay appears to be the most suitable since it is simple, rapid and inexpensive, and the amino acid requirements of the organism are in reasonable agreement with human (Rolle, 1975) and rat (Kidder and Dewey, 1961) requirements. Tetrahymena assays have also been shown to correlate well with rat-PER bioassay of commercial foods, r = 0.90 (Evancho et al., 1977). Effect 22 Freezing and Frozen Storage 22 Protein Functionality Freezing and frozen storage are widely used for meat pres- ervation. By freezing and in frozen storage many undesirable changes, such as microbial growth and metabolic processes, are inhibited. However, some chemical changes still can occur which affect the quality of meat products. Technological problems induced by freezing include drip loss, and with prolonged stor- age, toughening of the muscle tissue causing reduced acceptabili- ty and economic loss (Sikorski, 1978; Warrier et al., 1975). Denaturation of muscle proteins, especially myofibrillar proteins, play a major role in the deterioration in functionality of meats of beef, pork, poultry and fish during frozen storage. Functional changes during frozen storage have been related to protein insolubilization in the intact muscle of chicken (Khan and Van den Berg, 1967; Yamamoto et al., 1977), beef (Wagner and Anon, 1986), turkey (Smith, 1987) and fish (Matsamoto, 1980; 42 Shenouda, 1980). Wagner et a1. (1986) reported that the denatu- ration of myofibrillar protein occurred during freezing and frozen storage of beef when the myosin head region unfolds, followed by a weakening of the actin-myosin interaction as indi- cated by mg2+ and Ca2+ ATPase activity losses. These changes caused protein aggregation and decrease solubility of protein. Changes in ATPase activity and water absorption capacity due to freezing rate were also observed by these authors. Decrease in water-holding capacity during freezing is due to the fact that during freezing water-protein associations are replaced by protein—protein association or other interactions (Fennema, 1976; Hamm, 1975). Protein denaturation during freez- ing and frozen storage have been shown to influence gel proper- ties in protein from chicken (Smith, 1987), beef (Wagner and Anon, 1986) and fish (Kim et al., 1986). Protein denaturation during frozen storage of meat may be caused by one or more of the following factors: 1) formation of ice crystals (it has been shown that freezing causes the forma- tion of inter- and intra-cellular ice crystals which damage the cells and rupture the membranes), 2) enzymatic activity, 3) dehydration of protein molecules (i.e., migration of the water molecules of hydration to form ice cmystals which results in a disruption of the hydrogen bonding system as well as the exposure of a hydrophoic or hydrophilic surface of protein molecule, and consequently would leave these regions unprotected), 4) an in- crease in solute concentration in the unfrozen water phase and 5) 43 reaction of proteins with oxidizing lipids (Matsamato, 1980; Shenouda, 1980). Lipid oxidation occurs extensively in the refrigerated and frozen storage of nneat products (Dawson and Gartner, 1983) and may be one cause of myofibrillar protein denaturation (Sikorski, 1978; Buttkus, 1967). Lipid-protein interactions alter the functional properties of meat and may cause deleterious changes in final product quality (Sikorski, 1978). Igene et a1. (1979) reported that frozen storage increased malonaldehyde concentration in chicken meat. Pikul et a1. (1984) also reported a significant increase in malonaldehyde concentra- tion, when they stored frozen chicken at -18°C. Freeze-drying pf Meat and Meat Products Freeze-drying is the mildest method known for drying meats (Gooding et al., 1957; Regier and Tapple, 1956) and has become an important process for the preservation of foods, because the process involves a minimal structural change in the food (Anony- mous, 1980). But even this process causes undesirable changes in meat quality. The texture of rehydrated freeze-dried meat is drier and tougher than the original meat. The decrease of ten- derness and juiciness is caused by reduction in water-holding capacity' of the Inuscle proteins (Hamm and Deatherage, 1960). These undesirable changes possibly result from denaturation of muscle protein. The possibility of denaturation is indicated by the fact that the ATPase activity of actomyosin is partially reduced by freeze-drying of meat (Hunt and Matheson, 1959). They . 44 found that 20 to 60% of the ATPase activity and some of the contractility had been lost during drying. In freeze-dried food, loss of protein quality due to the interaction of protein with oxidizing lipid is a serious problem. Lipid oxidation is one of the most important problems in dehy- drated food, because of the highly porous nature of the product, making the lipid more accessible to oxygen and because of the low moisture content, which tend to promote oxidation. Relatively small increases in,the water content of dehydrated foods often retard oxidation (Maloney et al., 1966). Funes en: al. (1982) reported that freeze-drying promotes protein polymerization in aqueous emulsions containing lyzozyme and peroxidizing methyl fatty acid esters and decreases lipid hydroperoxides and malonaldehyde concentration. Water Activity and Its Influence 92 Food Product Quality and Stability The water activity (aw) of many foods is an important ther- modynamic property which can be used to predict the state and relative stability of food with respect to physical properties, rate of deteriorative reactions and microbial growth (Labuza, 1980). Water activity (aw) is defined as the ratio of the vapor pressure (P) of water in the food to the vapor pressure of water (Ps) at the same temperature (Brockman, 1970). The state of water in foods has direct effect on quality and stability through its effects on chemical and enzymatic reactions. Most bacteria do not grow below aw = 0.90, and most mold and yeast strains are 45 inhibited between 0.88 and 0.80 (Scott, 1957; Troller and Chris- tian, 1978). The occurrence of enzymatic reactions in low mois- ture foods when the enzymes have not been activated by heating has been the subject of extensive studies (Acker, 1969; Multon and Guilbot, 1975). It has been shown that there is a correla- tion between the activity of enzymes and the water content of foods. Nonenzymatic browning and lipid oxidation are the major chemical deteriorative mechanien that occur 2h) most dehydrated and intermediate-moisture foods. Oxidation is often at minimum when a food surface is covered by a monolayer of water (Labuza, 1971). Oxidation may increase to a maximum rate in an intermedi- ate moisture range, which corresponds tx>ia water activity range of 0.55 to 0.85. Water activity of a food has various effects on lipid oxida— tion. Water affects the rate of lipid oxidation in at least three ways: 1) an antioxidant effect due to hydration of metal catalysts, which decreases their catalytic action, 2) an antiox- idant effect due to bonding of hydroperoxides, which reduces their activity and 3) a proxidant effect due to increasing the mobility of reactants and catalysts (Labuza, 1971). Labuza (1969) and Labuza et a1. (1971, 1972) reported that as both water content and water activity (aw) increase well above the Brunau-Emmet-Teller monolayer (BET), oxidation rate increases due to mobilization of catalysts. At very low water activity levels, food containing unsaturated fats and exposed to atmospheric oxygen are highly susceptible to development of oxidative rancid- 46 ity. This high oxidative activity occurs at water activity levels below the so-called monolayer level of moisture. As water activity is increased, both the rate and the extent of autoxida- tion decrease until an aw in the range of 0.3 - 0.5 is reached, depending upon the system investigated, then the rate of oxida- tion increases through the intermediate moisture food (IMF) range. Malony et a1. (1966) reported that water had an inhibitory effect on the oxidation of a freeze-dried model system consisting of micro-crystalline cellulose and methyl linoleate, varying with water activity up to 0.5, and then at intermediate moisture levels, presumably aw 0.5 to aw 0.75; the lipid oxidation of model systems becomes accelerated (Labuza et al., 1970, 1969). Moisture Sorption Isotherm Water is a key component in foods, not only because it is the constituent in the highest concentration, but also because it strongly influences the physical structure, palatability and technical handling capability of the food material. Even more important, almost all deteriorative processes taking place in foods are in some way influenced by the concentration and mobili- ty of water. The graphical relationship between water content of a food product and its corresponding water activity at a given tempera- ture is expressed by a sorption isotherm (Labuza, 1968). Changes in temperature affect the relationship markedly (Wolf et al., 1973). Moisture adsorption isotherms have been used to calculate 47 the Brunaue, Emmet and Teller (BET) monomolecular layer value, and they have been used to predict the optimum moisture content of numerous foods (Rockland, 1969). The practical importance of sorption isotherms results from their usefulness in various areas of food technology. The knowledge of the water-vapor sorption isotherm of food products enables the food technologist to calcu- late the sorption and desorption enthalpies, to determine "bound water", to facilitate the calculation and operation of drying, mixing and packaging processes, to evaluate the physical, chemi- cal and microbiological stability of the product and finally to predict its shelf life (Stamp et a1. 1984; Labuza, 1980; Iglesias et al., 1979). Many investigators (Taylor, 1961; Rockland, 1969; Rockland and Nishi, 1980) have described the procedure for ob- taining water vapor isotherms for foods. In one general method the dehydrated food material is placed in a vacuum desiccator containing a specific saturated salt solution known to provide a definite equilibrium relative humidi- ty necessary for determination of the isotherms. Typical iso- therms in foods are S shaped, as shown in Figure 2. According to the theory of Brunaue et al. (1938), water is bound (strongly adsorbed) in a monomolecular layer in the region of the first slope of a moisture adsorption isotherm, up to point a, Figure 2. Above point a, in the linear section, bi- or multimolecular adsorption occurs. From about point b on, water is condensed in capillaries with increasing water activity (Acker, 1969). The ~bound water in food is strongly bound to hydroxyl groups of polysaccharides, the carbonyl and amino groups of proteins, and Figure 2. 48 HOISTURE. PER CENT OF SOLIDS o 1 . l l I I 0 5 I0 I! 20 25 RELAIIVE HUMIDITY Moisture isotherm curve of food products. 49 others on which water can be held by hydrogen bonding or by stronger interactions. The most effective way of estimating the contribution of adsorption at specific sites to total water binding is the use of BET. This mathematical relation is useful as an estimate of the "monolayer value", which can be considered as equivalent to the amount of water held adsorbed on specific sites. The amount of water which represents a monomolecular layer according to the BET theory may be regarded as a protective film which protects the food particles from attack by oxygen. The statistical monomolecular layer may run: ha fact represent a continuous film but rather corresponds to the number of available reactive adsorption sites in the protein, carbohydrate and fat components of the food. When the amount of water is adequate for combining with these functional group, they are pmotected from reaction with oxygen. In other terms, the relative humidity or moisture vapor pressure at this point represents a partial pres- sure of water vapor, which is competitive with the oxygen partial pressure to the extent of being protective. The equilibrium moisture content of a food sample depends on sorption mode, temperature and composition. With reference to composition, the sorption characteristic of food sample may be modified by the relative protein and fat contents. Proteins adsorb much more water at low aw's than do fatty materials (Labuza, 1968), and subsequently the presence of fat may depress the sorptionability of a protein sample (Hermansson, 1977). The presence of fat modifies the water sorption capacity of foods; 50 i.e., the higher the fat content, the lower the equilibrium moisture content for a specific water activity (Saravacos, 1969; VanTwisk, 1969; Heldman et al., 1965). For this reason and to allow a comparison between similar foods but differing in fat content, moisture content should be given on a percentage non-fat dry basis or grams of water per 100 gram of non—fat dry material. This procedure is based on the assumption that fat does not adsorb water. For the reasons discussed, it is clear that there is not a single isothenn for a given product. Pretreatment, composition and chemical changes may all somehow influence the shape of the isotherm. For this reason, the researchers should select the sorption data that most closely resembles their particular inter- est. EXPERIMENTAL PROCEDURE Experimental Design This study was conducted in four experiments, the first of which was designed to investigate the effect of processing (freezing and freeze drying) on lipid oxidation and protein solubility of chicken breast myofibrillar proteins. Three groups of samples were prepared as follows: 1) con- trol sample with no methyl linoleate, 2) sample + 15% fresh methyl linoleate and 3) sample + 15% oxidized methyl linoleate (methyl linoleate was air oxidized for 24 hrs). The degree of freshness and rancidity in methyl linoleate was measured by the conjugated diene method (Pryor et a1, 1976). The samples were frozen in liquid nitrogen. Frozen blocks were freeze dried for 24 and 48 hrs. Freeze dried samples were analyzed for solubility and lipid oxidation. The second experiment was designed to study the moisture sorption isotherm and calculate the monolayer value of freeze- dried chicken protein under different water activities. Freeze-dried chicken myofibrillar proteins were prepared as described subsequently herein. Desiccators equilibrate to six different water activities, aw 0.08, aw 0.11, aw 0.22, 3w 0.43, aw 0.65, aw 0.85 were used for this study. Freeze dried chicken proteins were humidified in ‘vacuum desiccators each of which contained a specific saturated salt solution for the desired water activities (See table 5). 51 52 The third experiment was conducted to investigate the effect of lipid oxidation, water activity and storage time on pmotein solubility. Freeze-dried chicken myofibrillar proteins were exposed to five different water activities aw 0.11, aw 0.22, aw 0.43, aw 0.65, aw 0.85 in this experiment. All freeze-dried samples were stored for three weeks in desiccators containing saturated salt solutions to give the above water activities. The fourth experiment was conducted to study the effect of lipid oxidation, storage time and water activity on gelling, water-holding capacity and nutritional quality of protein. For this experiment only, two water activities, aw 0.43 and aw 0.85, were prepared for this study. All freeze-dried chicken protein samples were stored for three weeks in desiccators con- taining desired saturated salt solutions. Materials The materials used in the experiment and their sources are listed below: 1. Chicken meat was purchased from local retail estab- lishments (East Lansing, MI). 2. Methyl linoleate (methyl ester of linoleic acid) (Sigma Chemical Company. St. Louis, MO). 53 3. Amino acids and other chemicals used in protein digesti bility test (Sigma). 4. Enzymes used in the in vitro digestibility study (Sigma). They were porcine pancreatic trypsin (Type IX) with 14,190 BAEE units per mg protein; bovine pancreatic chymotrypsin (Type II) 60 units per mg powder; and porcine intestinal peptidase (Grade III), 40 units per 9 powder. 5. Tetrahymena Pyriformis (American Type Culture Collec- tion, Rockville, MD). 6. Liquid Nitrogen (MSU campus) 7. Double Beam absorption spectrophotometer (Lambda 4B, Perkin-Elmer, Norwalk, CT). 8. All ingredients used for specific water activities (Sigma). 9. Automatic refrigerated centrifuge (Sorval Co. RC2—B) l0. Electronically Speed-Controlled Stirrer (Heller GT-21) equipped with a LM Jiffy Mixer Stirrer Shaft (Thomas Scientific Apparatus, Cat. 8634—520). 11. Instron Universal Testing Machine Model 4202, (Instrom, Canton, OH) 12. Virtis pilot plant freeze-drier (Model FF D42 WS) 13. Brinkman Polytron (Brinkman Instrument, Westbury, NY). 54 Methods Meat Sample Preparation Fresh unfrozen chicken breast was skinned, hand deboned and most of the fatty tissue and connective tissue were physically removed prior to grinding the muscle. The muscle was minced twice through a chilled grinder (Kitchenaid stand mixer KSA with a plastic food grinder attachment [Model FG-A]) and stored in a cold room (2°C) for further processes. Isolation pf Mypfibrillar Proteins Myofibrillar proteins were extracted in a 2°C cold room following a procedure described by Eisele and Brekke (1981) and Smith et a1. (1988) with some modifications. The ground muscle was blended for 30 sec in a Warning blender at maximum speed with four volumes of 0.1M KCL and 0.05M K phosphate buffer, pH = 7.1. The suspension was stirred for 30 min with a motorized propeller, but without foaming. Any connective tissue which accumulated on the propeller was discarded, and the suspension was transferred to 250 m1 centrifuge plastic bottles and centrifuged at 6000 x g for 10 min at 0°C. The supernatant, containing fat and sarco- plasmic protein, was discarded and the pellet resuspended in the original volume of fresh buffer and stirred for 60 min. The suspension was centrifuged and pellet resuspended in two volumes of cnilled distilled water and stirred for 30 min. The suspen- sion was strained through two layers of cheese cloth to remove excess connective tissue. The extraction procedure of stirring, centrifugation and resuspension was repeated two times with 55 chilled distilled water. The volume of final pellet was measured, and the pellet was analyzed for protein by Micro-Kjeldahl procedure (AOAC, 1984). A flow diagram of the extraction sequence is shown in Figure 3. Model System Myofibrillar proteins and 15% methyl linoleate (as percent- age of drip protein weight) were mixed with a motorized propeller for 2 min. The resultant slurry was then distributed in equal amounts in aluminum pans and placed in an insulated box on a perforated suspended floor at which liquid nitrogen was main- tained at a level of 2 cm below to 1/2 cm above in order to freeze samples within 3 min. The pans were covered with filter paper. Frozen blocks in the pans were transferred to a Virtis laboratory freeze-dryer (Model FF D42 WS). 11 control batch of myofibrillar protein and water was prepared in the same manner but without methyl linoleaUe. Figure 4 shows the experimental flow diagram for preparation of the model system. All freeze-dried samples were resuspended in 0.6M KCL prior to analysis. Saturated salt solutions which were used in adjust- ing water activity are presented in Table 5. Saturated salt solutions were placed in bottom of the desiccators which were held at room temperature in the dark. 56 Figure 3 - Flow Diagram of Myofibrils Extraction supernatant Chicken breast skinned, hand deboned Grind twice through a 4.8 mm plate 1 Blend with 4 volumes fresh, (.1 M KCl, chilled buffer 15 seconds 1 Stir 30 min i E Centrifuge (sarcoplasmic protein 6000 X g 10 min and fat) . 4, v Discarded pellet Resuspend in 4 vol fresh buffer stir for 1 hr Supernatant 11, Centrifuge ‘ swag x g 40 min Discarded pellet distilled diionized dd H20 Resuspend in Stir 30 min Strained Centrifuge 6000 X g 10 min b pellet ~V Resuspend in ddHZO two times Washed myofibrillar protein 57 Figure 4 - Model System Flow Chart Washed myofibrillar protein? 1 Methyl linoleate added 15% based on protein content __Experiment I Frozen in Liquid Nitrogen \A. Freeze dried for 24 and 48 hr | 3 i i E \J '1 Freeze dried for 48 hrs.__; Stored for 2 weeks in 6€/>/4 desiccators containing ‘ desiccators containing saturated salt solutions m saturated salt solutions aw = 0.08, 0.11, 0.22, D. aw = 0.43, 0.85 0.43, 0.65, 0.85 ‘ Stored for 3 weeks in desiccators containing saturated salt solution aw = 0.11, 0.22, 0.43, 0.65, 0.85 Stored for 3 weeks in vP.L ¥ % *EXP - experiment 58 Table 5 - Relative Humidities of Saturated Solutions at 25°C Saturated Salt Relative Humidity (%) by solution Reference at 25°C Sodium Hydroxide 8 (NaOH) Lithium Chloride 11 (LiCl) Potassium Acetate 22 (KC2H302) Potassium Carbonate 43 (K2CO3) Cobalt Chloride 65 (COC12) Potassium Chloride 85 (KCL) Source: Data adapted from Labuza, 1984. 59 Protein Solubility Protein solubility of freeze-dried chicken protein was measured following a procedure described by Morr et a1. (1985) with small modifications. About 500 mg of freeze-dried protein were weighed accurately in a beaker and mixed with 20 ml of 0.6M KCl, .05M K phosphate, pH 7.0. The mixture was homogenized with a Brinkman polytron (Brinkman Instrument, Westbury, NY) for 5 sec. Additional 0.6M KCl solution was added to bring the total volume of the dispersion to about 40 ml. The beaker was placed on a magnetic stirrer and dispersion was stirred overnight in 2°C cold room. The pH of dispersion was determined and adjusted to 7.0 with NaOH solution. The dispersion was quantitatively trans- ferred into a 50 ml volumetric flask diluted to the volume with additional 0.6M KCl solution and mixed by inverting and swirling, and then centrifuged 30 min at 27300 x g. The protein content of the supernatant was determined by Micro-Kjeldahl (AOAC, 1980). The solubility of the protein was calculated as: Protein solubility % = supernatant protein (mg/ml) x 50 Sample wt (mg) X decimal % protein in sample 100 6O Gelation Preparation pf Myofibril Gels Gels were prepared by adjusting a suspension of freeze—dried chicken myofibrillar proteins to 4.0% (w/v) protein, 0.6M KCl, pH 6.0. The suspension was mixed with a Brinkman polytron (Brinkman Instruments, Westbury, NY) for 5 sec and stirred overnight in 2°C cold room. The resultant protein suspension was transferred to 16 x 100 mm disposable culture tubes (approximately 10 g per tube) and placed in a 80°C water bath for 10 min. Heat-treated protein solutions were transferred to an ice-water bath, permit- ted to cool for 30 min and kept at room temperature for one hour. Water-holding Capacity Water-holding capacity of the gels were measured following a procedure described by Jauregui et al. (1981) with small modifi- cations. One and a half gram of gel were added to preweighed filter paper folded into a: 50 m1 conical centrifuge tube. The weight of the sample and filter paper was recorded. The sample was then centrifuged at 30000 x g for 15 min. The gel was removed from the filter paper, and the paper was reweighed. The expressi- ble moisture was measured according to the following formula and reported as percent weight lost from original sample. Expressible moisture (%) = Pw - P x 100 PS-P When: Pw is the weight of filter paper after centrifugation and after removing the gel P = is the weight of paper P5 = is the weight of paper and sample before centrifuga- tion. 61 Gel Strength The work required to penetrate the gel was evaluated as the back extrusion apparent viscosity using an Instron Universal Testing Machine (Model 4202, Canton, OH) equipped with a 50N load cell and coupled with a udcrocomputer (Hewlett-Packard 86B)(Lever, 1988). The speed of the plunger was 20 mm/min, travel distance 30mm, load calibration cell 50N. Distance-force data were used by the computer to calculate the back extrusion apparent viscosity and the apparent elasticity using the proce- dure described by Hickson et a1. (1982). Thiobarbituric Acid Test (TBA) The TBA distillation method of Tarladgis et a1. (1960) was used to measure lipid oxidation with minor modification. Two grams of freeze-dried protein was used instead of the prescribed 10 grams of fresh meat. This amount was used based on estimation of 20% protein in chicken meat. TBA numbers were expressed as mg malonaldehyde/g protein. Apparent I2 vitro Protein Digestibility The digestibility of freeze-dried chicken protein, casein, navy bean protein and treated casein was measured following the in vitro procedure described by Hsu et a1. (1977). This proce- dure utilizes a multienzyme system consisting of trypsin, chymo— trypsin and peptidase. Fifty nu. portions cu? aqueous protein suspension (6.25 mg protein/ml) were adjusted to pH 8.0 with 0.1N HCl or NaoH, while 62 stirring in a 37°C water bath. The multienzyme solution (1.3 mg trypsin, 3.3 mg chymotrypsin and 0.52 mg peptidase/ml) was main- tained in an ice bath and adjusted to pH 8.0 with 0.1N NaoH and/or HCl. Five ml of the fresh prepared multienzyme solution were then added to the protein suspension which was being stirred at 37°C. The pH drop was recorded over a 10 min period using a pH meter. Percent digestibility was expressed relative to ca- sein. Digestibility (%) = pH sample x 100 pH casein The samples were allowed to continue to digest for three hours and the three hour digests were then used to determine protein quality as assessed by the Tetrahymena bioassay. The digests were adjusted to pH 7.1, and the total nitrogen content of the digest was measured by using Micro-Kjeldahl procedure. The digest was diluted to contain 1 mg nitrogen/ml with distilled water. Tetrahymena Bioassay of Protein Quality All samples analyzed for protein digestibility were further assayed for protein quality. The protein quality was measured by the Tetrahymena bioassay procedure as described by Stott and Smith (1966) and Shorrock (1976) (described in detail in Appendix B). Preliminary tests were conducted to verify that the Tetrahymena bioassay procedure would detect differences in protein quality (amino acid profile and amino acid availability) and that Tetrahymena growth was 63 proportional to nitrogen concentration. Casein (high nitrogen, United States Biochemical Corporation, Cleveland, OH) and glucose were mixed (1:1 ratio) with 20 ml water, then heated in an oven (oven dry) at 90°C for 2 hr. Casein was also incubated with 100 m1 of .15M NaOH and autoclaved at 120°C for 4 lug The pH was adjusted to 7.0 with HCl. Casein, navy bean protein, heated casein plus glucose and heated casein plus NaOH were assayed at 0.025, 0.05, 0.075, 0.10 and 0.20 milligram nitrogen/ml. Growth was proportional to nitrogen concentration from 0.025 through at least 0.1 mg nitrogen/ml. Therefore, all samples were assayed at a concentration of 0.1 mg nitrogen/m1. Percent protein utiliza— tion was calculated according to the following formula: Protein utilization (%) = OD sample - sample blank x 100% OD casein - casein blank Determination pf Moisture Sorption Isotherms Moisture sorption isotherms for freeze-dried chicken protein were obtained as described by Labuza, (1984). Freeze-dried chicken proteins were mixed uniformly by mortar and pestle and humidified in vacuum desiccators each of which contained a spe- cific saturated salt solution for desired water activity. All samples were stored for two weeks and the moisture content of freeze-dried samples was determined on duplicate samples by the method described in AOAC (1980). After data had been collected, a plot of moisture on the Y axis vs aw on the X axis was made, and tnen the isotherm was used to calculate the BET mono- 64 layer. The BET monomolecular layer value of freeze-dried chicken proteins (control and treated samples) was calculated from the moisture adsorption isotherm data by the following procedure: BET equation: a = 1 + (c-l) a m: Grams of water per 100 g of dry matter at a water acti- vity a and temperature T. c: Constant related to the heat of adsorption. m0 = Monolayer value (grams of water equivalent to mono- layer absorbed on 100 g of dried solids. a = water activity (aw) This equation can be rearranged to give: a = I + (S.a) (1-a)m Where I = Intercept and S = slope. Thus, a plot of a/(l-a)m vs a gives a straight line. Then the monolayer from this plot is mo = l I + S. Statistical Analysis: Data for protein solubility and lipid oxidation (TBA numbers) in experiment 1 were analyzed as 3x4 factorial including: 1) control,+ 15% ML, + 15% oxidized ML; 2) either fresh, frozen, freeze dried for 24 hr or freeze dried for 48 hr. (The exercise repeated twice with three observations per subsample). Results for the moisture sorption isotherms (experiment 2) were analyzed an; 2x6 factorial. The factorial design 65 includes: 1) control vs ML treated myofibrillar proteins and 2) water activities of 0.08, 0.11, 0.22, 0.43, 0.65 and 0.85. For experiment 3, data on protein solubility and lipid oxidation were analyzed as a non-symmetrical 3 factor facto- rial which includes: 1) control vs ML treated myofibrillar proteins; 2) water activities of 0.11, 0.22, 0.43, 0.65 and 0.85; and 3) three weeks of storage. Results on WHC, gelation and the nutritional data (digestibility anui microbiological assay) 1n) experiment 4 were analyzed as a nonsymmetrical 3 factor factorial. The factorial design includes: 1) control vs ML treated myofi- brillar proteins; 2) water activities of 0.43 and 0.85; and 3) three weeks of storage. The significance between treatments was determined using either the Tukey test or Bonferroni t-test for com- parison analysis, after a significant F was determined (Gill, 1978; Woolf, 1968). Graphs were plotted using Har- vard Graphic. RESULTS AND DISCUSSION Effect pf Freezing and Freeze Drying 22 Chicken Myofibrillar Protein. Extracted myofibrillar protein slurries from chicken breast were allotted into three portions: 1) a control or untreated portion without added methyl linoleate, 2) a portion with 15% fresh methyl linoleate and 3) one with 15% oxidized methyl linol- eate. Aliquots of each of the above lots were frozen in liquid nitrogen and freeze dried for 24 and 48 hrs in a freeze drier. Freeze-dried samples were analyzed for moisture, protein solubil- ity and lipid oxidation. The percent solubility of chicken breast proteins during different stages of freeze drying is illustrated in Table 6. It is evident from the data that both freezing and freeze drying decreased (P405) protein solubility. Percent protein solubility dropped from 96.9% to 82.3%, 81.2% and 67.2% when the fresh isolate was frozen in liquid nitrogen, freeze dried for 24 hr and 48 hr, respectively. Freezing the samples in liquid nitrogen decreased solubility significantly, while freeze drying for 24 hr caused no further decrease in protein solubility in the control. However, freeze drying for 48 hr decreased (P(.05) protein solubility further as compared to freezing alone or freeze drying for 24 hrs. It seemed that the time of freeze drying had a significant impact on protein solubility. Similar results were obtained for the group of samples that 66 67 Table 6: Percentage solubility of chicken breast myofibrils during different stages of freeze drying Percent Solubility in 0.6M NaCl, pH 7.0 Treatment Control + 15% Methyl + 15% Oxidized Linoleate Methyl Linoleate Fresh isolate 96.9a 96.9a 96.9a Frozen in LN2 82.3b 82.0b ~ 79.6b Freeze dried for 24 hr 81.28 81.9b 71.6C Freeze dried for 48 hr 67.2c 33.8C 17.8d a,b,c,d Means in the same column followed by a common superscript do not differ (P(0.05). All values are the average of triplicate determinations. 68 were treated with 15% fresh methyl linoleate, but the results for the samples treated with oxidized methyl linoleate were differ— ent. In the presence of 15% oxidized methyl linoleate, protein solubility dropped from 96.9 to 79.6, 71.6 and 17.8% for frozen, freeze dried for 24 and 48 hrs. In this case, the decrease in solubility attributable to the first 24 hrs of freeze drying was also significant but smaller than that seen with an additional 24 hr of freeze drying. The drastic change in protein solubility after 48 hr in the samples treated with oxidized methyl linoleate may be attributed to the presence of oxidized lipid material itself and/or the duration of freeze drying. It is speculation that a greater amount of protein denaturation may have occurred during the latter stages of freeze drying. Also, the presence of unoxidized methyl linoleate may have been oxidized during freeze drying and led to insolubilition of chicken proteins. Lipid oxidation of chicken breast myofibrillar proteins during different stages of freeze drying is shown by the results of TBA tests in Table 7. It is clear from the data in Table 7 that rapid freezing in liquid nitrogen had no significant effect on lipid oxidation in the control or methyl linoleate “treated samples. Freeze drying for 24 hr increased (P<.05) lipid oxida- tion only in the presence of methyl linoleate. However, freeze drying for 48 hr increased (P(.05) lipid oxidation in the control and methyl linoleate treated samples. It is evident that the presence of methyl linoleate and the longer period of freeze drying increased lipid oxidation as measured by TBA numbers 69 Table 7. Lipid Oxidation of Chicken Breast Proteins During Different Stages of Freeze Drying TBA Number Treatment + 15% Methyl + 15% Oxidized Control Linoleate Methyl Linoleate Fresh isolate 0.20a 0.23a 1.01a Frozen in LN2 0.23a 0.31a 1.06a Freeze-dried for 24 hr 0.29a 91.73b 1.69b Freeze-dried for 48 hr 0.52b 1.30C 6.70° a,b,c Means in the same column followed by a common superscript do not differ (P40.05) 70 Thus, the interaction of oxidized lipid with protein may be the reason for the decrease in the protein solubility mentioned above. Freeze drying for 48 hrs in the presence of oxidized methyl linoleate led to a 5-f01d increase in lipid oxidation as compared to freeze drying for 24 hrs. This clearly indicates that both the presence of oxidizable lipid during extended freeze drying and lipid oxidation per se cause decreases in protein solubility; The presence of fresh methyl linoleate during the longer (drying period effected a: 50% reduction. in solubility, while the use of already oxidized material resulted in a 74% decrease in protein solubility as compared to control samples. The results on protein solubility correspond to those re- ported by Wagner and Anon (1986). These authors found that freezing and frozen storage decreased protein solubility, altered the structure of myofibrillar proteins and rheological behaviour of the myofibrillar proteins. They also reported that freezing of muscle had a denaturation effect on myofibrillar proteins. The changes in muscle proteins during freezing or frozen storage have been regarded by many investigators as a denaturation phe- nomenon (Matsumoto, 1980; Wagner and Anon, 1985, 1986; Fennema et al., 1973; Shenouda, 1980 and Park et al., 1987). These changes are manifested mainly by a decrease in solubility or extractabil- ity of the myofibrillar fraction. The decrease in the amount of liquid water available to the proteins at freezing temperatures, the mechanical damage at various muscle structures caused by ice crystals and the induced increase in concentration of tissue 71 salts and other extractives have been regarded as major causes of protein denaturation :hi frozen foods (Sikorski, 1978). Other investigators also related the loss of protein solubility during freezing to the denaturation of proteins (Dyer and Morton, 1956 and Love, 1958). According to Dyer (1951), the proteins present in the sol form in the fresh tissue are converted to gel form, which is denaturated by salts in the muscle at the eutectic point. It, therefore, appears that the dehydration of muscle cells during frozen storage is the main factor favoring the condition for denaturation. The results obtained showing the effect of freeze-drying on lipid oxidation and protein solubility are in close agreement with the data of Kuo and Ockerman (1984) which showed that freeze drying caused undesirable changes in muscle proteins. Hamm and Deatherage (1960) reported that decrease of tenderness and juici- ness in freeze-dried post-rigor meat was caused by a loss of water-holding capacity and solubility of muscle proteins. They concluded that these undesirable changes resulted from denatura— tion of muscle proteins. The possibility of denaturation is indicated by the fact that the adenosinetriphosphatase (ATPase) activity of actomyosin is partially reduced by freeze drying of meat (Hunt and Matheson, 1959). The results on lipid oxidation agree with those reported by many investigators (Kanner and Karel, 1976; Schiach and Karel, 1975 and Funes and Karel, 1981). Kanner and Karel (1976) report- ed similar changes in lysozyme due to reaction with peroxidizing 72 methyl linoleate in a dehydrated model system. Kazuki et al. (1987) reported a big loss in solubility and protein digestibili- ty when they incubated casein with linoleic acid. Funes et a1. (1982) reported that freeze drying promotes protein polymeriza- tion in the presence of peroxidizing lipids. This effect may be due to concentration of protein and lipids by removal of the solvent, which results in a higher rate of protein-free radicals formation and subsequent protein polymerization. Moisture Sorption Isotherms The relation between water activity and moisture content of freeze-dried chicken myofibrillar proteins can be precisely described by the moisture adsorption isotherm depicted in Figure 5. These isotherms show the typical sigmoidal shape of a general moisture adsorption isotherm reported by other investigators (Konstance et al., 1983; Labuza, 1984; Mittal and Usborne, 1985). The curve for the methyl linoleate treated myofibrillar proteins is higher (P{.0l) than that for the control (no methyl linoleate added). Thus, this data indicates that methyl linoleate in— creased the water sorption at any given water activity. These results are, however, contrary to the general concensus that increasing the fat content of foods decreases their sorption isotherms (VanTwisk, 1969; Hermansson, 1977; Konstance et al., 1983; Mittal and Usborne, 1985). These investigators have worked on air dried model systems. Hence, their proteins were probably native and made the most contribution to the sorption properties of their model systems. 73 mm: II 00230.. .IT laser (0 CD I mm; ) d Teal mm Ms No .. 09 C0 0 El aw mwre .‘m I St Ia MW 9 ( .5 C) I _ _ _ _ L o c.» or 9m em a <<>._.mm >0._._<_._.< mwocnm m. onmdcnm mdmonpneo: Hmonrmna 0m firm nosnnop one nnmmnma A+ Hmw amnzww Hesopmmnmv mammnmldHemm arenwm: B._.mw >O._._<_._.< mwmcnm m. are wma Bosonme meow 0m dam mnmmumlmwwma Oreoxmn Unmmmn BKOmHUHHHHmH wnonmwnm mn Nmon. m u imnmn mnnwmm_z O>mm_Z+Q_ICOme Z><< wm>z O>mm_Z+Z>OI Hmmfismzh. mwncno He. owcmmnwcwwwnw 0m wnonmw: mHoa nwmmws. nw<< own: use wmcwnmnwnnm nmmmws. 101 caeehd as the only protein source, extensive weight losses oc- curred. Incubation of casein with glucose resulted in poor digestibility. This result is in close agreement with Nielson et al. (1985a). Maillard reactions between protein and reducing sugars are probably the most important reactions which occur in food proteins during processing and storage. Finot and Magnent (1981) reported that half the lysine in casein which had under— gone Maillard reaction with glucose was extracted in faeces of rats. In addition, in vitro enzymatic release of lysine after Maillard reaction has been found to be especially low (Scarbieri et al., 1973). Navy bean protein, as a source of lower quality plant protein, showed reduced digestibility compared to casein control. The second part of the experiment was done on the freeze dried chicken proteins. The extracted myofibrillar proteins were divided into two groups: .1) control group with no methyl linoleate and 2) treated group with 15% methyl linoleate. All samples were frozen in liquid nitrogen and freeze dried for 48 hrs. The freeze-dried chicken protein was placed in two different. water activities (0.43, 0.85) for three weeks, and samples were analyzed for protein digestibility and protein quality. Casein was used as a positive control. Figure 16 shows the percent protein digestibility of control and treated samples at different water activities. It was clear from the data that in vitro protein digestibility did not change 102 % RELATIVE PROTEIN UTILIZATION o>mm_z o+oEoOmm 9.280: z><< mm>z Hmm>._._<_mz._. mwocnm Hm. wmnnmcn wnonmw: CHHHHNmnHo: ow emnnwrwamsm wKHOmOHEHm mm vmmmnnmm ow Dwmmmnmsn mnonmwd mocnnmw. 103 PERCENT RELATIVE PROTEIN DIGESTIBILITY Zr v2 ._mo 30 mo mo .5 mo u nonnHoH u zmnvkw resonmnm n Smnmn >nnwm_mz 0 >23 gr >23 0 >58 5 >28 Hmm>._._<_m2._. MHmcmm Hm. UMommnHUwan< 0m mnonmws mm bmmmnnmm 6% anmwnamsn mad Smnmn >nnwmm_z 8 :5 022$ 3 >53 o><28 amm>H§mza mwocnm Ho. wmnnmzn wnonmw: cnwwwnmnwos UK emnHmIKEmsm wmmmnnmm UK enmmnamdn m Smnmn >nnwmm_z IT o>mm_2+oEoOmm 1T o>mm_2+2mo: 1m: z><< 2%: I ah m _ _ L h _ _ o m S m m so am wwocnm H. nzmsom e: w: o4 >23 m.m 1 IT o>mm_zr IT 8239 >56 m... 1m. 8230.. >58 .X. 2315. 52055 >28 Nm . N .l ! /-u//uuln (find: . 1w. WWJ 1%] U 1“ mm - j]. {aura _ _ _ _ _ _ m m so am o N ._.__<_m nzmsom H: w: o