l ll Illillllllllllllllllllllllll 8”" fiflfll'I-E- ~ -.—4-.a-v-.--.:~_~=-~ unveil)???" -..V.._v. u... 4.... . ‘1! 300639 4617 .15 v; 3‘ ‘ 7 31,15. 1 f Mines .' .v £3§® I This is to certify that the thesis entitled SOME FACTORS INFLUENCING THE NON-HIME IRON CONTENT AND LIPID OXIDATION ‘I‘N MEAT presented by CHUIN -‘CHIEH CHEN has been accepted towards fulfillment of the requirements for M_. SE degree in _EQQD_SQLEN CE Amati giajor professor Date DEC. 21., 1082 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution MSU RETURNING MATERIALS: Place in book drop to _mm§myl remove this checkout from LIBRARIES your record. FINES will be charged if book is N (A) Experiment B. . . 27 Experimental System . . . . 27 Addition of Antioxidants. . 27 Moisture Analysis . 29 Fat Analysis. . 29 Cooking Treatments. 30 Measurement of Lipid OxidatiOn by the TBA . Test. 30 RESULTS AND DISCUSSION. . 32 Experiment A. . . 32 Contents of Pigment Extract . 32 Influence of Various Methods for Analysis Of Non— heme Iron . . . 33 Concentration of Non- heme IrOn in Muscle - Tissue. . 35 Effect of Fast and Slow Heating On the Stabili- ty of Heme Iron . . . . 37 Effect of Nitrite on the Stability of Heme. Iron. . . . . . . . . . 4O Experiment B. . . 43 Effect of Heating On TBA Number Of Muscle Tissue. . . . 43 Effects of Salt and AntiOxidant On the TBA Number of Cooked Meat . . . . . . . 45 SUMMARY AND CONCLUSIONS . 49 LIST OF REFERENCES. 52 Table LIST OF TABLES Lipid composition of lean beef muscle. Iron compounds and their distribution in the adult human body . Design of experiment for testing effect of anti- oxidant-coated salt on cooked meat The pH values and concentrations of non-heme iron of muscle pigment extract analyzed by three different methods. . . . . . . . Concentration of total iron and non-heme iron in muscle tissue. Concentration of non-heme iron in slow heated muscle pigment extracts. Concentration of non-heme iron of fresh and nitrite-treated meat pigment extracts at varying heating times. . . . . . . . . . . TBA numbers of raw meat and heated meat with and without antioxidants and salt. vi Page , lO , 28 , 34 36 . 38 , 4l .44 INTRODUCTION The term warmed-over flavor (NOF) was first used by Tims and Watts (I958) to describe the rapid development Of oxidized flavor in refrigerated uncured cooked meat. The rancid flavor usually becomes apparent within 48 hours at 4°C. NOF has been recognized by consumers for years as evidenced by their aversion to "warmed-over" steaks, roasts and other "leftover" meat items, yet not until the last l5-20 years have scientists paid any attention to its significance. The need for a goOd understanding of NOF is generated because changing social patterns and eating habits have greatly increased the demand of precooked quick-frozen meals which frequently are plagued by this problem. NOF was considered to reSult from lipid oxidation by Tims and Watts (l958), who showed that the loss of meat flavor was paralleled by an increase in TBA values. Sato and Hegarty (l97l) first proposed that non-heme iron played a major role in accelerating lipid oxidation. Love et 11. (1974) confirmed the observations of Sato and Hegarty (l97l) by showing that TBA values were not accelerated when metmyo- globin was added at level of l to lO mg/g of meat. However, levels of ferrous iron as low as 1 part per million (ppm) resulted in enhanced lipid oxidation in water-extracted cooked meat. By using a model system, Igene 33 al (1979) demonstrated that the concentration of free iron in a cooked meat extract was 4.l8 ug/g meat compared to 1.80 u9/9 meat for the uncooked extract; Nitrite at a level of l56 ppm was shown to inhibit NOF in a meat model system by Igene gt 11. (l979) and in meat system by Fooladi gt 31. (l979). Although several mecha- nisms have been proposed for the antioxidant activity of nitrite (Sato and Hegarty, l97l; Zipser _t‘_l., l964; Kanner, 1979), none of them have been proven. Therefore, it is the purpose of the present study to investigate the mechanism by which nitrite prevents NOF. It is also of great importance to test the effects of sodium chloride, and of antioxidants such as a-tocopherol and Tenox 4 on development of NOF in refrigerated cooked meat. Salt has been shown to be a prooxidant in some cases and either have no effect or be an antioxidant in other foods (Lea, l939). Both prooxidant activity (Ellis t al., l974; Kimato et al., 1974) and antioxidant activity (Lips, 1957; Nitting, l975) of a-tocopherol has also been found in carcass lipids and meat products. But the effect of antioxidant-coated salt on cooked meat has not yet been studied. Specifically, the Objectives Of this investigation were: and (l) (2) To determine whether nitrite reacts with myoglobin to inhibit the release of free iron from myoglobin during cooking. To ascertain the effects of NaCl on the develop- ment of NOF in cooked meat; To investigate the effects of various antioxidants on develOpment of NOF in cooked meat which contains 2% added salt.~ LITERATURE REVIEW Distribution and Composition of Animal Fats Lipids in meat, poultry and fish are commonly classi- fied as depot or adipose tissue and as intramuscular or tissue lipid (Watts, 1962; Love and Pearson, 1971). The depot fats are largely localized as fat globules within the individual cells, and consist mainly of triglycerides (Watt, 1962). The tissue lipids contain proportionately larger amounts of phospholipids, which occur largely, if not entirely, in association with proteins as lipoproteins and proteolipids (Watts, 1962). Tissue lipids are an integral part of various cellular structures, such as the cell wall (Kono and Colowick, 1961), the mitochondria (Holman and Widmer, 1969), the sarcoplasmic reticulum (Newbold 23.31., 1973), and the microsomes (Macfarlane gt al., 1960). Although the amount of tissue lipids in meat is highly variable, ranging from 2.0% to 12.0% in beef (Orme 33 al., 1958), the level of phospholipids (O.5-l.0%) remains nearly constant when expressed as a percentage of muscle (Dugan, 1971). According to Hornstein gt 31. (1961), the phospho- lipids in muscle contain a larger percentage Of unsaturated fatty acids than the neutral lipids, with particularly high levels of linoleic and arachidonic acids. The lipid composition for the beef lean tissue is given in Table 1. Variation in the phospholipid content was found to vary between species by Kaucher _t _t. (1944) and from location to location within the same Species by Gray and Macfarlane (1961). Luddy gt gt. (1970) found that porcine light muscles contained 20% more lipid and that these lipids contained 20% more triglycerides and 40% less phos- pholipids than those from dark muscles. They also found the level of fatty acids in the phospholipids from light muscles were higher in monoenes while the dark muscle phospholipids predominated in polyunsaturates. A similar report was established with poultry by Peng and Dugan (1965) and by Acosta gt__l. (1966). Katz £3.11- (1966) have shown that dark meat (legs) from the chicken contains only about half as much phospholipids as white meat (breast). Component phospholipids expressed as a percentage of the total phospholipids are somewhat similar in most animal tissues (Body gt gt., 1966; 1970). Keller and Kinsella (1973) reported the composition of beef phospholipids as being 53-58% phosphatidyl choline (PC), 23-25% phosphatidyl ethanolamine (PE), 5-7% sphingomyelin (SP), 5-7% phospha- tidyl inositol (PI), 1-4% phosphatidyl serine (PS) and 1-6% all others. Table 1. Lipid composition of lean beef musclea Lipid Fraction % Triglycerides (per cent of tissue) 2-4 Phospholipids (per cent of tissue) 0.8-1.0 Fatty acids with 2 and 3 double bonds' Per cent of triglycerides 6.1 Per cent of phospholipids 25.2 Fatty acids with 4 or more double bonds Per cent of triglycerides 0.1 Per cent of phospholipids 19.2 aHornstein t 1. (1961) Role of Phospholipids in Lipid Oxidation All of the early work which had been done on rancidity in meats was concerned with the oxidation of adipose tissue (Watt, 1954). Tims and Watts (1958) first noted that rapid deterioration in the flavor Of cooked meat was correlated with the amount of phospholipids on using the TBA method to measure the amount of oxidation. They postulated the possible denaturation of protein during heating may free the phospholipids, and thus make them more susceptible to oxidative attack. On fractionating the total lipids into triglycerides and proteolipids, it was demonstrated by the TBA test that the latter are responsible for the intensive oxidative reaction induced by heating porcine muscle tissue (Youna- thans and Watts, 1960). Campbell and Turkki (1967) have shown that the phospholipid concentration is higher in cooked meat than in raw meat because the neutral lipids are lost from the meat more readily on heating than the phospholipids. Igene gt gt. (1981) found that the drippings collected upon cooking contained largely triglycerides, whereas PE was essentially absent, indicating that it was bound to the membranes. The increased proportion of PE in cooked meat coupled with its susceptibility to oxidation indicate that it may play a key role in the autoxidation of cooked meat. Igene _t_gt. (1981) also found there was a significant decrease during frozen storage in the amount of PE and PC, but the decline was muCh greater upon cooking and storage. Decreases in the component phospholipids may be due to either autoxidation, hydrolytic decomposition, the lipid- browning reaction of lipid protein copolymerization as outlined by Lea (1957). Tissue Iron Monier-Williams (1950) stated that blood contains about 70% of the total iron in the body, while the other 30% is located in the tissues in various forms, partly as heme iron and partly as non-heme iron. Among the heme compounds, "they stated that hemoglobin and myoglobin are the major iron-containing porphyrins present in blood and muscle, respectively. Iron is also associated with the flavin nucleotide enzymes, cytochrome oxidase, xanthine oxidase and succinic dehydrogenase (Merkel, 1970). Iron is not only stored in the non-heme compounds as ferritin and hemosiderin, but it is also present in transferrin (Merkel, 1970). According to Thompsett (1934) the non-heme iron,may be present in both states of OXidation. Ferric iron appears to combine with the non-diffusible phosphorus compounds, such as phosphatides and phosphoproteins. Ferrous iron apparently does not form such complexes if the iron becomes reduced. since it is liberated in the ionizable form. Tompsett (1935) gave values for non-heme iron ranging from 100 to 150 ppm in liver, spleen and bone down to roughly 5 or 10 ppm in other tissues, such as the brain and kidney. The concentration of non-heme iron was reported to be 1 ppm in beef top round steak by Sato and Hegarty (1971), and 1.8 ppm or 8.7% of the total iron of the pigment extract from beef longissimus dorsi muscle by Igene gt gt. (1979). In contrast, Cook and Monson (1976) reported that the total non-heme iron comprised from 40-50% of the total iron in beef muscle. Similar results of 61.6% for beef muscle were found by Schricker gt gt. (1982). Table 2 shows the distriOution Of iron in the human body. Effect of Heating on Myoglobin Myoglobin and other heme-containing proteins undergo denaturation during heating. Denaturation Of these proteins causes the rapid release of the heme moiety from the globin part Of the molecule leaving free heme, which is very sensi- tive to oxidation (Lawrie, 1966). Cooking, if sufficiently thorough, destroys the porphyrin complex and liberates the available iron (Monier-Williams, 1950). Igene t 1. (1979) demonstrated that cooking released a significant amount of non-heme iron from bound heme pigment, which they demonstrated increased the rate of lipid oxidation in cooked meat. Youna- than and Watts (1959) have attributed the rapid oxidation of lipids in cooked meat to the conversion of ferrous iron from 10 Table 2. Iron compounds and their distribution in the adult human body. Iron in Per cent of compounds grams total iron Iron porphyrin (heme) compounds Blood hemoglobin 3.0 60-70 Myoglobin 0.13 3-5 Heme enzymes Mitochondrial cytochromes c 0.004 0.1 a ,a,c],b - - - MTcrosomal cytochrome b5 - - Catalase 0.004 0.1 Peroxidase - - Nonheme compounds -Flavin-Fe enzymeS' Succinic dehydrogenase - - Xanthine oxidase of liver - - NADHC-cytochrome c reductase - - Iron chelate enzyme aconitase - - Transferrin 0.004 0 l Ferritin 0.4-0 7-15 Total available iron stores 1.2-1 - Total iron 4-5 100 aGranick (1958). 11 the porphyrin to the ferric form during heating. Tims and Watts (1958) have suggested that denaturation of protein during heating may free phospholipids, and thus make them more susceptible to oxidative attack. Effect of pH on Myoglobin Apomyoglobin, which is myoglobin devoid of its heme group, can be obtained by lowering the pH of a myoglobin solution to pH 3.5 according to Stryer (1972). This author stated that the heme group binds only weakly to the protein at this acidic pH, and thus can be removed by extraction with an organic solvent. Lewis (1954) found less than 10% of the prophyrin was cleaved from Meth at pH 3.6. In contrast, Snyder (1963) showed that at pH 6.6 40% of the heme present in Mb was extracted with acetone but negligible amounts of hematin were extracted from Meth. Heme Compounds as Prooxidants The catalytic effect of iron prophyrins on the oxidative decomposition of polyunsaturated fatty acids was first described by Robinson (1924), who attributed the catalysis to the iron content of the molecule. According to Greene (1971), the main forms in which myoglobin (Mb) exist in meat are as reduced Mb, as oxymyoglobin (MbOZ) and as nitric oxide ferrohemochrome, in-which the iron porphyrin is in the ferrous (Fe++) state. The other forms Of myoglobin 12 include ferrihemochrome and metmyoglObin (Meth), both Of which contain iron in the ferric (Fe+3) state (Kendricks and Watts, 1969; Greene, 1971). Ferric hemochromogen, the denatured form of this pig- ment in cooked meat, is postulated to be the active catalytic form of the muscle pigments (Younathan and Watts, 1959; Tappel, 1953). Tarladgis (1961) attributed the catalytic activity of ferric hemOproteins to the paramag- netic character of the prophyrin-bound iron, He suggested that the presence of five unpaired electrons in metmyoglObin produces a strong magnetic field that would favor the initi- ation of free radical formation. A number of investigators (Watts, 1954; Tappel, 1952; Younathan and Watts, 1959; Liu and Watts, 1970; Greene, .1975) have also indicated that hematin compounds are involved in lipid oxidation in meat. Heme Compounds as Antioxidants The antioxidant activity of heme compounds was esta- blished in fatty acid model systems but there have been no studies on meat products. Kendrick and Watts (1969) reported the linoleate-to-heme ratio for maximum catalysis of lipid oxidation to be 100 for hemin and catalase, 250 for metmyoglObin, 400 for cytochrome c and 500 for methe- moglobin. At heme concentrations of two to four times the Optimum catalytic amount, they noted that lipid oxidation 13 did not occur. They theorized that a stable lipid hydro— peroxide-heme derivative was formed at inhibitory heme concentrations. At lower heme concentrations, it was postulated that the heme may be unable to contain the lipid radicals, and oxidation results. Mechanism Of Heme Catalysis Catalysis by iron porphyrinsis characterized by rapid initiation and propagation Of the lipid oxidation chain reaction according to Tappel (1962). He suggested that hematin involves the formation of a lipid peroxide-hematin compound and its subsequent deComposition into free radi- cals, which prOpagate the chain reactions with the concomi- tant destruction of the catalyst. Tappel (1955) suggested the mechanisms for hematin-catalyzed unsaturated lipid oxidation as shown in Figure 1. The hematin com- pound (a)' and the lipid peroxide (LOOH) are postulated to form an activated compound (b). Subsequent scission of the peroxide bond occurs, resulting in production of a lipid radical (L0‘) and a heme radical (c). Abstraction of a hydrogen atom (H) from a lipid molecule (LH) regenerates the hematin and produces a lipid radical (L'). Tappel (1962) also suggested that a direct attack on the lipid by the heme compound could result in generation Of lipid radicals according to the following mechanism: 2+ + LH + hematin - Fe3t—9~L' + hematin - Fe + H 14 L0° + LH——-—> LOH + L. L. + 02 >:L00° L00- + LH -————e> LO0H + L. Where: LH = Linoleic acid LOOH = Hydroperoxide linoleate Figure 1. Mechanism for hematin-catalyzed lipid oxidation (Tappel, 1955). 15 Role of Non-Heme Iron in Lipid Oxidation Although the catalytic effect of hemoglobin and other porphyrins on lipid oxidation has been a generally accepted phenomen in the past, the importance of non-heme iron on lipid oxidation has been pointed out by scientists in recent years. Kwoh (1971) reported that heme was the dominant catalyst in cooked meat, but that significant lipid oxi- dation still occurred in cooked meats in which the heme had been destroyed by H202. Wills (1966) indicated that non-heme iron was a more active prooxidant at acid pH values, whereas, hemOproteins were less pH-sensitive. Sato and Hegarty (1971) reported that beef muscle, which had been thoroughly extradted with water, did not develop warmed-over flavor indicating that the substance($) respon’ sible for initiating the reaction was/were water soluble. Heme compounds were found to have little effect on the devel- opment of warmed—over flavor in this system. The reaction was apparently catalyzed by ferrous iron and ascorbate. They suggested that ascorbic acid functions by keeping at least a portion Of the iron in the ferrous state. At higher levels ascorbic acid inhibited the reaction, possibly by upsetting a balance between ferrous and ferric iron. The observations Of Sato and Hegarty (1971) were con- firmed by Love and Pearson (1974), who showed that metmyoglO- bin at concentrations from 1 to 10 mg/g meat did not influence 16 TBA values of water extracted cooked beef, but that Fe2+ was effective as a prooxidant. Love and Pearson (1974) also pointed out that the prooxidant activity was located in the low molecular weight fraction of the extract. Igene gt_gt. (1979) demonstrated that the increased rate of lipid oxidation in cooked meat is due to the release of non-heme iron from the meat pigments during cooking. Addition of 2% EDTA was shown to effectively chelate the non-heme iron and significantly reduced lipid oxidation. Metallic Ions as Catalysts of Ligid Oxidation Ingold (1962) pointed out that metals, such as iron, cobalt, and copper which possess two or more valency states with a suitable oxidation-reduction potential between them, are particularly important catalysts. Ing01d (1962) also stated that the effect of metals can be reflected in an altered rate of chain initiation, propaga- tion, or termination, as well as by an altered rate of hydrOperoxide decomposition. The basic function of the metal catalyst is to increase the rate of formation of free radicals (Ingold, 1962). Uri (1956) has described a commonly accepted mechanism for metal catalysis involving the oxidation of a metal ion with hydrOperoxide decomposition resulting as follows: M” + ROOH—>M+("”) + 'OH + RO' l7 Ferrous iron has been shown to have greater prooxidant activity than ferric iron in a number of experimental systems (Brown gt gl., 1963; Sato and Hegarty, 1971). Measurement of WOF in Cooked Meat WOF was considered to result from lipid oxidation by Tims and Watts (1958). They showed that the loss of meat flavor was paralleled by an increase in TBA value, which measures the amount of malonaldehyde - a product of lipid oxidation. Zipser _t _t. (1964) found a high correlation between TBA values and oxidative off flavor in cooked meat. Thus, TBA test has been used routinely to measure off flavor development in cooked meat. Younathan and Watts (1959) concluded that the difference in flavor between nitrite-cured and non-nitrite cured meat soon after cooking is due to the development of warmed-over flavor caused by the rapid oxidation of unsaturated fatty acids in the uncured meat. 0n the other hand, nitrite protected the meat against oxidation. Action of Antioxidants on Food Lipids Antioxidants may interfere with or delay the onset of oxidative breakdown of fats and fatty foods (Blanck, 1955). Primary or phenolic antioxidants (such as tocopherols, butylated hydroxyanisole or butylated hydroxytoluene) Function by breaking the oxidative reaction chains 18 (Shelton, 1959). In support of this viewpoint, Cort (1974) reported that the phenolic antioxidants act as electron or hydrogen donors to quench electron mobility with subsequent interruption of free-radical chain reac- tions. According to Uri (1961), the mechanism Of antioxidant action is as follows: ROO’ + AH (antioxidants) - >=ROOH + A' The radical may be stabilized by recombination in either of two ways: 0 + . > A A A2 ROO' + A' 7: ROOA This means that during autoxidation, the antioxidants are converted into dimers and other products (Uri, 1961). It is also possible that the antioxidant is oxidized directly by oxygen. This is the case for tocopherol, which is partly oxidized to tocoquinone in fats (Tappel, 1962). At the end of the induction period, the antioxidants disappear with little being known as to their exact fate (Cort, 1974). a-Tocopherol as a Lipid Antioxidant The tocopherols are products of synthesis by plants, and may be deposited in animal tissues in the non-saponi- fiable portion of the lipid fraction, usually together with sterols, vitamin A, vitamin K and other naturally occurring antioxidants (Mervyn and Morton, 1959). The lg tittg antioxidant activity of the four known tocopherols increases in order of alpha, beta, gamma and delta, while the activity 1g tttg increases in the opposite order (Parkhurst gt gl., 1968). According to Aruand and Woods (1977), vitamin E (tocopherols) is not destroyed by acid, alkali, the process of hydrogenation, or by high temperature, but is oxidized slowly by air and rapidly in the presence of rancid fats. The chemical basis of the antioxidant action of vitamin E is its combination with free radical intermediates of lipid oxidation and lipid peroxides, thus inhibiting further lipid peroxidation (Tappel, 1962). A number of studies have demonstrated the beneficial antioxidant activity of a-tocopherol in improving the stability of carcass lipids of poultry (Marusich _t _l., 1975), of beef (Kimoto gt gt., 1974) and of pork (Astrup, 1973). Nevertheless, some workers regard a—tocopherol as a poor antioxidant, particu- larly in products containing highly unsaturated fatty acids (Lips, 1957; Witting, 1975). 20 At low concentrations, a-tocopherol functions as an antioxidant, but at high concentrations may become a pro- oxidant (Chipault, 1961; Labuza, 1971). Witting (1975) suggested that an increase in tocopherol concentration results in increased peroxide formation through free- radical initiation, an increased rate Of autocatalysis and an increased rate of destruction. Effect of NaCl on Oxidation of Meat Sodium chloride, a common meat additive, has a puzzling effect on oxidative changes in meat. The role of NaCl in initiating color and flavor changes in meat is well known but poorly understood. Some of the studies on salt are complicated by the fact that salt may contain metal conta- minants, which may serve as catalysts of lipid oxidation. Rancidity may still develop in the fat of dry cured hams, even though a low metal containing salt (0.1 ppm copper and 0.4 ppm iron) is used (Olson and Rust, 1973). An increase in the concentration of NaCl from 2 to 6% has been reported to result in a rapid rate of monocarbonyl formation during freezer-storage of pork tissue by Ellis gt _t. (1968). Lea (1937) suggested that NaCl influences lipid oxida- tion by promoting the activity of lipoxidase in meat. Later work by Banks (1961) and Tappel (1952) showed that meat does not contain lipoxidase. 21 Chang and Watts (1950) attributed the prooxidant acti- vity of salt to the possibility that salt may affect the physical state of meat in such a way that the hemoglobin would be brought into closer contact with the fat. Nitrite as an Antioxidant in Cooked Meat Nitrite has been shown to be an effective antioxidant in cooked meat by many researchers. Younathan and Watts (1959) reported that nitrite and sodium chloride acted synergistically to retard oxidation of lipids in cooked meat stored at refrigerated temperatures. Sato and Hegarty (1971) showed WOF in cooked ground beef was eliminated by nitrite at a level Of 220 ppm and partially inhibited at 50 ppm. Swain (1972) found the TBA values of nitrite- treated hams were initially lower than comparable samples without nitrite and remained lower during storage up to 2 weeks. Fooladi gt gt. (1979) and Igene gt_gt. (1979) demonstrated that the addition of nitrite at a level Of 156 ppm protected against oxidative changes during the storage of cooked meat. Several mechanisms by which WOF is inhibited have been proposed. Zipser and Watts (1967) reported that the lower level Of oxidation in stored, cured meat results from the conversion of the pigments to the catalytically inactive ferrous nitric oxide hemochromogen. Nitrite was also 22 suggested to inhibit WOF by stabilizing lipids in all membranes which are normally disrupted and exposed to oxygen by cooking or grinding (Sato and Hegarty, 1971; Pearson _t gt., 1977). Kanner (1979) demonstrated a potent antioxidant, s- nitrosocysteine, was generated during curing of meat with nitrite, and thus serves as an inhibitor of WOF in cured meat. EXPERIMENTAL Materials Source of Meat Beef was obtained from the Michigan State University Meat Laboratory. Portions of semitendinosus muscle were excised from carcasses of 9 month old cattle immediately after slaughter. The meat was placed in a 4°C cold room until 24 hours postmortem, then it was wrapped, frozen and Stored at -200C for future use. Solvent and Chemicals Experiment A - All chemicals and reagents were of analytical grade except acetone which was of reagent grade. Experiment B - All chemicals were of reagent grade Unless otherwise specified. The sodium chloride was of analytical grade. Tenox 4 (Eastern Chemical Products, Inc., Kingston, TN, 37662) consists of 20% butylated hydroxyanisole (BHA), 20% butylated hydroxytoluene (BHT) and 60% corn oil. The a- tocopherol-coated salt was Obtained from Diamond Crystal Salt Company, St. Clair, Michigan. It contained 3.25% a-tocopherol coated on 96.75% salt. The other antioxidants used in this experiment included BHA-citric acid-propylene glycol-coated salt, which contained 3% BHA, 1.6% citric acid and 2.2% 23 24 propylene glycol coated on 93.2% salt. Methods Experiment A Preparation of Meat Model System A meat sample weighing approximately 500 grams was thawed at room temperature. After removing all visible fat and connective tissue, the lean tissue was homogenized with one volume of deionized water in a Waring blender. The homo- genate was saved and the pigments were extracted by adding two additional volumes of deionized water and stirring con- stantly with a magnetic stirrer at 4°C for 24 hours. The extract was then separated from the meat residue by filtering through cheese cloth. Extraction with three volumes of water was repeated four times. At this point residue was practically colorless. The extracted meat pigments were concentrated in a Virtis II freezer-drier, until each milliliter Of the extract represented 2 grams of raw meat sample. The cured pigment was produced by adding nitrite to the meat at 156 ppm level. A period of 24 hours was allowed to ensure the curing reaction. The cured pigment was then ex- tracted using the same procedure as for the raw meat pigments. Analysis of Total Iron in Muscle Tissue The concentration of total iron and non-heme iron in the sample were measured by using an atomic absorption spec- trophotometer (Instrumentation Laboratory, Inc., Lexington, MA). 25 To measure the concentration of total iron, nitric acid and perchloric acid digestion was carried out on the entire muscle sample. Five grams of meat were placed in acid-washed 150 m1 Erlenmeyer flasks with 20 m1 of concentrated nitric acid and 5 m1 of perchloric acid. The mixture was heated on a hot plate until the digest appeared clear. The samples were then diluted with glass-distilled water to appropriate volumes so that the mineral concentration fell within the linear range using the atomic absorption spectrophotometer. If 5 m1 of muscle pigment extract was used as sample, 5 ml of concentrated nitric aCid and 1 ml of perchloric acid would be needed for digestion of sample. Analysis of Non-Heme Iron by Different Methods Method 1 - Direct Acid Digestion. This procedure was a modification of the method of Schricker _t gt. (1982) who used direct acid digestion of the meat samples. Muscle pigment extracts of 2 ml were incubated in the presence of 2 ml of an acid mixture containing equal volumes of 6 N HCl and 40% trichloroacetic acid (TCA) in a 65°C water bath for 20 hours in loosely stoppered 15 ml centrifuge tubes. After centrifugation at 3000 rpm for 15 minutes, the clear supernatants were transferred to small tubes and directly analyzed using an AA Spectrophotometer. Method 2 - Chelation with EDTA and Precipitation with TCA. This method utilized the procedure described by Igene t l. (1979). TO 2 ml of pigment extract, 1 m1 of a 26 12% ethylenediaminetetraacetic acid (EDTA) solution was added to chelate any non-heme iron in the sample. Then the heme iron was removed by precipitating heme compounds with 1 m1 of 40% TCA. After centrifugation at 3000rpm for 15 minutes, the supernatants were transferred and analyzed by AA spec- trophotometry. Method 3 - Chelation with EDTA and Precipitation with Acetone. This procedure was a modification of the method of Igene _t gt. (1979) with the TCA being replaced by acetone. Two ml of the pigment extract weretreated with 1 ml of 12% EDTA to chelate the non-heme iron. Then 1 ml of acetone was added to precipitate the heme iron (myo— globin). For all three methods, the amount of heme iron was obtained by the difference between total iron and non-heme iron. Heating Treatments of Muscle Pigment Extracts The pigment extract was placed in 50 m1 test tubes. To show the effect of incomplete cooking, the uncooked extract was heated for time periods of 5,105nu120nfinutes Dian 85-870C water bath. Temperature change of each sample was monitored by thermocouples in the center of the tubes. Cleaning Of Glassware All of the glassware Twas cleaned by soaking in dilute HCL (1:3, v/v) for more than 30 minutes then rinsed with deionized water to eliminate iron contamination. Caution 27 was also taken to avoid contamination of the sample with air. Exgeriment B Experimental System This part of the study was designed to test the effects of salt and antioxidants on the development of WOF. A sample weighing 1300 grams was thawed and ground through a 3/8-inch plate then through a 3/16-inch plate. The ground meat was divided into five different groups and assigned to one of the five different treatments as shown on Table 3. One third of the samples in each treatment group was used as raw meat while the remaining samples were cooked. After cooking, half of the cooked samples were refrigerated at 4°C for 48 hours, while the other half were immediately used for the TBA test. The TBA test was carried out on the raw meat immediately after addition of the antioxidants and on the cooked meat at 0 and 48 hours refrigeration. Since the amount of antioxidants added to the samples was based on the fat content, it was necessary to determine the content of fat and moisture before adding the antioxidants to the samples. Addition of Antioxidants According to FDA regulations, when a single anti- oxidant is added, it may not exceed 0.01% based on the fat content of the food. When more than one antioxidant is added, the combined total may not exceed 0.02%, Of which no one antioxidant may exceed 0.01%. When antioxidants 28 Table 3. Design of experiment for testing effect of anti- oxidant-coated salt on cooked meat. Sample Number Treatment 1 Control- no added salt or antioxidants 2 2% NaCla w/o antioxidants 3 _ 2% NaCla coated with 0,01% oI-tocopherolb’C 4 2%.NaCla coated with 0.01% BHAb, 0.007% citric acidb’c and 0.005% propylene- glycolb’c 5 2% NaCTa with 0.01% BHAb’C and 0.01% BHTb’c. 'a% salt based total weight of sample. b% antioxidant based on fat content of sample. c Chemicals coated on the salt. 29 and synergists are added, the combined total may not exceed 0.025%, with no one antioxidant exceeding 0.01%. During the preparation of sample, additional salt was added to the original formula of antioxidant-coated salt to end up with 2% salt in the sample. The reason for choosing 2% as the level of salt to be tested is because 2% salt is commonly used for many commercial cured meat products. Moisture Analysis A variation of the A.O.A.C. (l975) procedure (24,003) was used for moisture analysis. Aluminum sample dishes were dried at least one hour before use, cooled and handled with tongs to avoid any fat and/or moisture contami- nation from the hands of the analyst. A sample of approxi- mately 4 g was weighed out into the previously tared, dried dishes. The sample were dried in an air convection oven at 100°C for 18 hours, then cooled in desiccator. The percentage weight loss of the sample was calculated and used as the moisture content. Fat Analysis A variation of A.O.A.C. (1975) procedure (24,005) was used for fat analysis with extraction by the Goldfisch apparatus. The Goldfisch beakers were dried in an air convection oven for at least one hour at 100°C then cooled in a desiccator. The dried samples after moisture analysis were used for determining the fat content. The dried samples were rolled together and then placed in the white 3O porous thimble. Care was taken to avoid spilling the dried material out of the dishes. Then 30 m1 of anhydrous diethyl ether was poured into each of the tared and dried beakers. The thimble containing the dried sample was put in a monel holder and clipped 0n the apparatus with the beaker being attached firmly to the condenser. The hot plate was raised to touch the bottom of the beaker and the heat was applied. Heating was set for about three hours to complete the extraction of fat. By replacing the monel holder with a reclaiming tube, nearly all the ether in the beaker was driven off and collected. The remaining ether was removed by placing the beaker in the drying racks of the apparatus. When no odor of ether remained, the beakers were dried in an air convection oven for one hour then cooled in a desiccator. The weight gain of the beaker was taken as weight of fat in the sample, and was used to calculate the percentage of fat. Cooking Treatments Meat samples were put in retortable pouches with the bags left open at one end, and cooked in boiling water until the internal temperature of the samples reached 700C. They were then cooled by tap water. Drippings were collec- ted and added back to sample before further testing. Measurement of Lipid Oxidation by the TBA Test The distillation method of Tarladgis gt gt. (1960) was used for measuring TBA numbers. The distillation 31 apparatus consisted of a 250 m1 round bottom flask, which was attached to a Friedrich Condensor with a three-way connecting tube. It was then placed in an electric heating mantle. A 10 9 sample of meat was homogenized with 50 ml of distilled deionized water for 2 minutes in a Virtis homogenizer at low speed. The homogenate was transferred quantitatively into a 250 m1 round bottom flask by washing with 47.5 ml of distilled, deionized water. The pH of the meat slurry was adjusted to 1.5 by the addition of 2.5 ml of 4 N HCl. Boiling chips were added and a small amount of Dow antifoam was sprayed into the flask to prevent foaming. The slurry was steam distilled using the highest setting on a power stat until 50 m1 of the distillate were collected. The distillate was mixed and 5 ml were transferred to a 50 ml test tube. Then, 5 ml of TBA reagent (0.02 M 2-thi0- barbituric acid in 90% glacial acetic acid) were added. The tubes were stoppered and the contents mixed. The tubes were heated in a boiling water bath for 35 minutes. After cooling in cold water for 10 minutes, absorbance was read on a Beckman DU spectrophotometer at 538 nm against a blank containing distilled deionized water and TBA reagent. Absorbance readings were multiplied by a factor of 7.8 (Tarladgis EE.El-’ 1960). TBA values are expressed as mg malonaldehyde per 1,000 g of sample. RESULTS AND DISCUSSION Exgeriment A The first part of this study was designed: (l) to determine the influence of various methods of analysis on the amount of non-heme iron in meat pigment extracts; (2) to investigate the effects of heating time on the non— heme iron in the pigment extract; and (3) to ascertain the effects of nitrite on the amount of free iron released during the heating of the pigment extract. Contents of Pigment Extract Most of the analyses done in the study were based on the water extract from raw meat samples. The extract of muscle tissue would be expgcted to contain sarcoplasmic proteins, inorganic constituents, some carbohydrates and their metabolic intermediates. According to Forrest _t _t. (1975) the sarcoplasmic proteins, which are readily extrac— table in water, include myoglobin, hemoglobin, and the enzymes associated with glycolysis, the citric acid cycle and the electron transport chain. Although the enzymes of the citric acid cycle and the electron transport chain are contained within the mitochondria, they are readily extrac- ted along with those found directly in the sarcoplasm. AS 32 33 shown in Table 2, the enzymes and other compounds containing iron are found in the pigment extract. Analysis for total iron showed that water extraction removed 86.3% of the total iron, or 21.22 ug/g out of 24.57 mg of iron/g of meat sample. Similarly, Sato and Hegarty (1971) reported that the hemin content of the water extracted tissue was reduced to about 25 ppm from a total of 240 ppm found in a 10 g unextracted muscle. This indicates that 89.6% of the total iron was water extractable. This value is comparable to the values obtained in the present study. Influence of Various Methods for Analysis of Non-heme Iron Three different methods were used in this study to compare the effects of different pH values and temperatures on the concentration of non-heme iron in the pigment extract. The results are shown in Table 4. The final pH values varied for the different methods, with pH values of 1.21, 1.57 and 5.43 for Methods 1, 2 and 3, respectively. The effects of pH valueon the non-heme iron content can be seen by comparing methods 2 and 3. If the pH values were to exert a significant effect upon the value for non-heme iron, the results for the two methods (2 and 3) should be different. However, the values are in good agreement (1.20 and 1.15 ug/g meat, respectively). Although the porphyrin ring binds weakly to globin at pH 3.5 according to Stryer (1972), the differences in pH value did not have any measur- able effects On the concentration of non-heme iron. EDTA 34 .mmewu m>we umpmow_amg we: m:_m> zoom m .mo.ov a pm «enurewcmwm appmowpmwpmpm mew mpgwcumcmazm pcmgmwwwc An vm30__oe cog? mam;-:oc Lee mwzpm>m .mFaEmm com pomcuxm ucmsm_a mFOmze mcvm:_ Awmgmwr mcopmo< Tnzaczv Po.mF o.o mo.onam_._ mm.Fuo_.m— vo.onm¢.m In ucmmmmm vogue: mpmcowumw>mv ugmucmpm mcwzozm muogpme pcmgmmewv macs“ an vaxpmcm uumgpxm pcmsmwa mpomae we cog? mamgucoc $0 mcowpmgucmucoo new mwzpm> :a cam: .v m_nme 35 was used in both methods 2 and 3 to tie up the free non-heme iron before any other protein-bound iron was precipitated by TCA or acetone. The effects of temperature and heating time upon the values for non-heme iron were also investigated using Method 1, which was a modification of the procedure of Schricker _t gt. (1982). Using this method, samples were incubated with acid in a 65°C water bath for 20 hours. This gave a value of 1.50 ug/g meat for the non-heme iron concentration of the pigment extract. This value is considerably higher than that of the other two methods, with Method 2 and 3 giving values of 1.20 and 1.15 ug/g, respectively. The low pH value and long-time heating for Method 1 may release more heme iron from the globin moiety than other two methods. Concentration of Non-heme Iron in Muscle Tissue The amount of non-heme iron in raw muscle tissue varies widely between different studies as shown in Table 5. Sato and Hegarty (1971) reported that there was about 1.0 ug/g of non-heme iron in muscle. Igene _t _t. (1979) reported the percentage of non-heme iron to be 8.7% , or 1.80 ug/g meat out of a total iron content Of 20.64 ug/g meat. In the present study, it was found that non-heme iron was 1.31 u9/9 meat, which amounted to 6.2% of the total iron (21.22 ug/g meat). These results agree closely with values 36 Table 5. Concentration Of total iron and non-heme iron in muscle tissue. Total Non-Heme % Non-Heme Iron Iron Iron Sato and 24.00a 1.00c 4.2 Hegarty (1971) lg3;§)_£._lo 20.64b 1.80b 8.7 Present study 21.22b 1.31b 6.2 (method of Igene gt gt. 1979) Schricker 23.40a 8.40a 35.9 et a1. TTEB?) aValue expressed as ug/g meat, using raw meat for sample. bValue expressed as ug/g meat, using extract for sample. CSample information unavailable. Approximate value reported by authors with rounding