u v . v union. ... u . 3...: . .-. . v u 4.1 4 IA» .oovl . A a '.-H m ¢ vl'alvvvomlvd. o .bWhVIONO. 1.031.“: u -mfifwknd 1.: 1.; 31 “1‘27. U, _ .H': 1‘ '[)u‘.;. “.1, .. L‘E :g'ru’m ‘ . t n .‘d’ln'f'A «‘1' t, J . “WU- . 1| 4‘53“}, :‘ .‘T :1 3' ’ . :53: Vim)“ --, - - n. ._. , .. 1 ‘- 4-. ‘ O o J.“- 311 354-",jfi." ‘ A '.“' Cu2+> fine flake salt (FFS) > control (CNT). Treatments had no detec- table effect on lipid oxidation as monitored by changes in fatty acid profile, but the proportion of PE and PC was significantly affected (P 81' + H' (free radicals) radical (Rfl mechanism was described in detail by Swern (1961). Propagation involves the combination of the first free radical with molecular oxygen to form peroxide free radical (ROOTL. These compounds are then able to abstract another hydrogen from a non-oxidized unsaturated fatty acid (828) to form further free radical and hydroperoxide which is capable of propagating the chain reaction. The hydroper- oxide is one of the major oxidation products that decom- poses to form compounds responsible for off-flavor (Gaddis et al., 1961; Horvat et al., 1969). R1. + 02 ---------------- > R100' (peroxide free radical) 8100' + RZH ------------------ > Rooa + 82° (hydroperoxide) 13 Another important mechanism for the oxidation of unsaturated lipids which involves activated species of oxygen was reviewed by Frankel (1985). Singlet oxygen, produced by photo-oxidation in the presence of sensitizers such as chlorophyll, is an important reactant. Singlet oxidation of methyl oleate takes an entirely different stereochemical course than free radical autoxidation. Singlet oxygen adds directly to either unsaturated carbon by an ene addition reaction, and results in a change of configuration of double bond from cis to trans. As linoleate reacts with singlet oxygen at a rate at least 1588 times faster than triplet oxygen, it was suggested that this very fast reaction is important way of initiating the free radical autoxidation of unsaturated lipids. Hydroperoxides formed by singlet oxidation can decompose thermally or in the presence of metal catalysts into alkoxyl and peroxyl radicals that can accelerate free radical autoxidation. Deactivation of free radical with the creation of non- radical end products is considered to be the termination of the chain reaction. Free radical inhibitors (R1) include antioxidants that may form inert end products as a termi- nation step. 14 R.+R. .................... >RR R. + R00. .................. >ROOR ROO' + ROO' ---------------- > ROOR + 02 R‘ + R1 ---------------- > RR1 Lipid Oxidation Products Hydroperoxides are the primary products of unsaturated fatty acid oxidation. Hydroperoxides are colorless and odorless and do not contribute to the off-flavor associated with lipid oxidation (Watts, 1954; Sato and Herring, 1973). Degradation of hydroperoxides through a series of scission and dismutation reactions yields the secondary products of lipid oxidation. The secondary products include aldehydes, ketones, acids, lactones, alcohols and unsaturated hydrocarbons (Frankel, 1984L. The secondary products are directly responsible for the off-flavor of oxidative rancidity (Lea, 1962; Sato and Herring, 1973). Herz and Chang (1978) reported that carbonyls are the predominant members in any class of compounds in the meat flavor concentrates. Fat soluble carbonyls are the ones primarily involved with meat flavor. Love and Pearson (1976) found that hexanal is one of the important products of lipid oxidation which contributes to the development of WOF. Heptanal and n-nano-3-6-dienal were found to be associated with WOF in turkey meat (Ruenger et al., 1978). 15 Catalysis of Lipid Oxidation Porphyrins of hemoglobin, myoglobin and the cytochromes are catalysts for unsaturated fat oxidation in meat because of their iron content (Robinson, 1924; Tappel, 1952; Younathan and Watts, 1959). The reaction between lipid and hemoprotein is believed to destroy the pigment and oxidize the fat (Watts, 1954). Tappel (1955) demonstrated 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, 1959). However, Brown et a1. (1963) reported that heme with iron in either the ferrous or ferric states were effective catalysts of lipid oxidation. Labuza (1971) disagreed with the theory that only ferric hemes are capable of catalyzing lipid oxidation in meat and stressed the importance of knowledge of the electron orbital structures in understanding the role of heme pigments in lipid oxidation. He suggested that the protein portion of hemoprotein molecules may cause steric hindrance of the iron, preventing it from catalyzing oxidation. When meat is heated, denaturation of the protein portion of the molecule might facilitate exposure of iron to unsaturated fatty acids. In a 1975 review, Green and Price concluded that either Fe2+ or Fe3+ hemes might function as catalysts 16 3+ hemes may be necessary for of lipid oxidation, but the Fe rapid catalysis. The most important mechanism involved in heme-catalyzed lipid oxidation has been considered to be the catalytic decomposition of hydroperoxides to generate free radicals. However it was reported by Kanner and Harel (1985) that the interaction of hydrogen peroxide with Meth or MetHb led very rapidly to the production of an activated cata- lysts which initiated membranal lipid peroxidation. The activity of this catalyst as a prooxidant decreased rapidly during the first 5 minutes, and within 18 minutes was down to 58% of the fresh preparation. Kanner and Harel (1985) have suggested that the prooxidant effect was obtained when the ratio between heme group and hydrogen peroxide is almost 1:1, consequently the maximum prooxidant effect was reached at 5 to 7 micromoles MetHb vs. 38 micromoles Heth (keeping hydrogen peroxide at 38 micromoles), because MetHb molecule contains four heme groups instead of one in Heth. Tappel (1962) reported that the most probable mechanism involved the formation of a coordinate complex between the heme compound and lipid hydroperoxide, followed by hemolytic scission of the peroxide bond. In this mechanism there would be no changes in the valence of heme iron. The findings of Sato and Hegarty (1971) were surprising in that neither ferrous nor ferric iron at relatively high concen- 17 tration had much effect on TBA values while the addition of iron powder resulted in an approximate doubling of TBA values. Accordingly they thought that the concentration of ionic iron or a balance between ferrous or ferric ions was critical to the reaction. Their results also showed that ferrous iron could play an important role in WOF develop- ment. Hirano and Olcott (1971) and Kendrick and Watts (1969) have demonstrated that heme compounds may act as either accelerators or inhibitors of lipid oxidation. The action depends on the ratio of heme to unsaturated fatty acids. Waters (1971) and Love and Pearson (1974) reported that there is strong evidence that ferrous iron is a more active catalyst of lipid oxidation than ferric iron. Love and Pearson (1974) and Igene et al. (1979) have demons- trated that nonheme iron is the major prooxidant in cooked meat and is released from heme pigments during cooking. Schricker et al.(1982), Schricker and Miller (1983) and Chen et al. (1984) have confirmed these results. It has been demonstrated in a variety of systems that the hemopro- teins are ab1e~to act as(antioxidants with certain ratios of fatty acid to heme, but Liu and Watts (1978) suggested that the ratio in red meats should favor catalysis rather than an antioxidant effect. Kunsman et a1. (1978) suggested that the polyunsaturated fatty acid / hemoprotein ratio in mechanically deboned red meat may be in the range where 18 hemes exert an antioxidant effect, thus explaining the observation that lipids in this product do not oxidize more than in ground beef. Cu2+ has been reported as having an inhibitory effect on the oxidation of hydrocarbons (Ingold, 1962). The inhibitory properties of cupric salts in lipid oxidation was attributed to chain termination by the catalyst (Ingold, 1962; Sato and Hegarty, 1971). However, a complex formation of free radicals with Cu2+ during propa- gation may also result in reduced rate according to Ingold (1962). The data reported by Tichivangana and Horrissey (1985) indicate that Cu2+ catalyzed oxidation followed a 2+ catalysis, but Cu2+ was pattern similar to that for Fe slightly less effective. The rates of oxidation show that prooxidant activity was in the order : Fe2+) Cu2+> Coz+) myoglobin (Mb), and that differences in activity between Fe2+ and Cu2+, Fe2+ and Co2+ and Fe2+ and Nb were signifi- cant at the P< 8.85, P<8.81 and P<8.88l levels, respective- ly in all muscle systems. The susceptibility of the raw and heated muscle systems to lipid oxidation catalyzed by the various prooxidants was in the order : fish) turkey) chicken) pork) beef) lamb, which is generally consistent with polyunsaturated fatty acid content of these tissues. Between 1 and 12% sodium chloride the rate of lipid oxidation increased in proportion to the concentration of the salt added (Castell et al., 1965). Castell et a1. 19 (1965) concluded that pure sodium chloride has no direct effect on the oxidation of lipids but acts only on non- lipid components which activate the lipid oxidation. Hills and Conochie (1946) proposed that fat peroxide reacts with chloride ions, resulting in the formation of chlorine. The free chlorine formed brings about further oxidation of the fat. Of particular interest is the ability of salt to denature actomyosin and other proteins in the muscle. Salt probably disrupts the lipoproteins or other complexes con- taining lipids (Dyer, 1953). It was found by Castell et al. (1965) that the prooxidant activity of sodium chloride in fish muscle was the result of sodium ions, and that cations of other metal salts had a similar prooxidant effect. When they used equivalent concentrations, the relative activity was in the followdng order : Cu2+) Fe3+) C03+) Cd2+) Li+) Ni3+) HgZ+> Zn2+) Ca2+) Ba2+. They found that sodium chloride and many other metal salts acted as a prooxidants, 2+ but they have a strong inhibitory effect on Cu -induced rancidity. Antioxidants Antioxidants are substances that slow down oxidation of fats or fat-containing foods to prolong their wholesomeness and palatabilityu The concentration of an antioxidant used is important for reasons of cost, safety, sensory 28 properties and functionality. Some antioxidants provide increased protection as their concentration increases, whereas others have optimal levels, and at higher levels may be prooxidant (Dugan, 1976). Dugan (1976) indicated that the most effective antioxidants in food systems function by interrupting the free radical chain mechanism. This group [which includes butylated hydroxyanisole (BHA), butylated hydroxytoluene (BHT) and tocopherol] consists mainly of phenolic compounds, capable of donating a labile hydrogen ion. Acidic compounds such aslcitric acid, ascorbic acid and EDTA delay the onset of oxidative rancidity by inactivation of prooxidant metals, and are often used in combination with phenolic antioxidants.Aucombination of antioxidants is commonly used to obtain better effect than the various individual antioxidants (Dugan, 1968). Citric acid was more effective against iron and nickel while ascorbic acid and its derivatives were effective against copper, but not iron (Morris et al., 1958). Sato and Hegarty (1971) showed that nitrite can comple- tely eliminate WOF at 228 PPM and will inhibit its develop- ment at 58 PPM" IFooladi (1977) indicated that 156 PPM of nitrite prevents WOF in cooked meat and poultry; It has been reported by Kraybill et al. (1949) that BHA was very valuable in the stabilization of food which has been heat 21 processed. According to Marion and Forsythe (1964), a significant delay of autoxidation of turkey lipids occurs with the addition of 8.84% BHA based on weight of lipids. Lower tocopherol content of turkey meat compared to chicken meat might be responsible for better storage stability of chicken meat (Mecchi et al., 1953). TBA, Sensory Scores and Bacterial Count The reaction of malonaldehyde (MAL) with 2-thiobarbitu- ric acid (TBA) has been widely used for measuring the extent of oxidative deterioration of lipids in muscle foods (Gray, 1978 and Rhee, 1978). The TBA test expresses lipid oxidation in mg MAL per Kg of meat sample. MAL is a secon- dary oxidation product of polyunsaturated fatty acids which have three or more double bonds (Dahle et al., 1962; Pryor et al., 1976). The structure of the adduct of TBA-MAL.was reported by Nair and Turner (1984) who indicated that the combined spectroscopic data are totally consistent with two spectrally equivalent tautomeric structures 1 and 2. TBA-MAL Adduct 22 They also concluded that formation of the 2:1 adduct of TBA and MAL probably is limited by neucleophilic attack invol- ving carbon-5 of TBA onto carbon-l of MAL followed by dehydration and similar subsequent reaction of the interme- diate 1:1 adduct with second molecule of TBA. Siu and Draper (1978) showed that the absorption spectrum of the pink solution in the visible range generated by adding the TBA reagent to meat distillates or filtrates was identical to that of the complex formed with the pure 1,3,3,- tetramethoxypropane (TEP). The meat extracts exhibited a single pink spot corresponding to that of the TBA—MAL complex when chromatographed on silica gel G (Rf-8.83). They concluded that these findings indicate that the TBA reaction is a valid indicator of the MAL present in meats. It was reported by Kakuda et a1. (1981) that there was a linear relationship, with an r2 of 8.946, between the TBA absorbance at 532 nm and HPLC peak height of the MAL. The HPLC method measured MAL levels from 1x18'11 to 4x18'11 mole/l8 microliters, compared to detection limits of 2 x 18'11mole/l8 microliters for the colored MAL-TBA complex. The TBA test can be performed directly on the food product, followed by extraction of the colored complex (Sinnhuber and Yu, 1958), or on a portion of the distillate of the food sample (Tarladgis et al., 1968), or on an 23 extract of the food (Witte et al., 1978.) Although the distillate method is the most popular one, it doesn't necessarily mean that i4: is the most accurate or reproducible method. Witte et al. (1978) reported that the extraction method has the advantage of simplicity and ease of use. It also has the advantage of being more specific and selective for carbonyl compounds as described by Tarladgis et a1. (1962) because it does not require heating. Witte et a1. (1978) reported that TBA values determined by the distillate method were twice as large as those determined by extraction. The differences may be due to the incomplete extraction of malonaldehyde in the extraction method, since no heat was involved. The heat of distillation may have increased the quantities of aldehyde from lipid precursors, or heat disrupted certain carbonyl addition products thought to occur by reaction between malonaldehyde and amino acids, pyrimidines or protein (Buttkus, 1967). Witte et al (1978) concluded that it is unlikely that the low values obtained in the extraction method were due to color development with TBA at room temperature rather than by Bm53g of the color heating since it has been shown that the developed at room temperature is greater than that developed by boiling (Tarladgis et al., 1964). A high 24 correlation between the two methods was observed by Witte et a1. (1978). There was a linear relationship between peroxide value and MAL concentrations during the oxidation of various classes of polyunsaturated fatty acids: linoleate, arachidonate, pentaenuate and hexaenuate (Sinnhuber and Yu, 1977). 1 Significant (P<8.85) correlation coefficients of r:— 8.57 and r=-8.51 were found between TBA values and sensory scores for beef and cMicken white meat model systems, respectively (Igene and Pearson, 1979). A correlation coefficient of -8.77 between flavor ratings and TBA number for turkey meat was reported by Jacobson and Koehler (1978). Microbial and/or oxidative changes limit the acceptable shelf life of meat and result in significant flavor changes in the final products (Dawson et al., 1975). Lipid oxidation is accelerated and bacterial reproduction occur more rapidly in ground beef resulting in off-flavor. Keskinel et al. (1964) reported that the TBA number increased in beef, pork, lamb and turkey meat after grinding. Freshly ground chicken and turkey have a bacterial count from 188,888 to 1,888,888 per gram (Maxey et al., 1973), and storage for 4 days at 5 C increases bacterial numbers to a level which is approximately 25 associated with unacceptable products. Dawson et a1. (1975) concluded that a comparison of bacterial growth with panel scores indicates that the rapid bacterial growth the first 18 days of storage had little effect on the flavor of fried patties. This work seems to indicate that the grinding has a greater effect on TBA than the initial increase in bacterial population. Fatty acid profile It has been reported by Lea (1953) awd Keller and Kinsel la (1973) that changes in fatty acid composition of lipids are important indirect indicators of lipid oxidation. Igene et a1. (1981) demonstrated that the fatty acid profiles in the triglycerides of beef, chicken dark meat and chicken white meat were not significantly changed during 13 months of frozen storage or cooking. These findings are in agreement with the results of Chang and Watts (1952), and Campbell and Turkki (1967). However, Igene et a1. (1981) reported that there were significant changes in fatty acid composition in phospholipids of beef and chicken dark and white meat. Unsaturated fatty acids particularly polyunsaturated fatty acids (PUFAS) participate in the development of oxidized flavor in frozen and/or cooked meat. Most of the losses in unsaturation were found in arachidonic acid. These results were verified by Igene and Pearson (1979) with model meat 26 systems. They have shown that the PUFAS of phospholipids were major contributors in the development of rancidity during frozen storage or in development of WOF in cooked meat. Fishwick (1968) investigated changes 2h: lipids of turkey muscle during chilled and frozen storage and reported that the plot of the loss of PE against the increase in free fatty acids (FFAS) was linear. The slopes of the lines calculated on a molar basis correspond to PE/FFA ratios of 36:188 and 38:188, respectively for leg and breast muscle. These values are of the same order of magnitude as the respective concentrations (38% and 32%) of PE in the total phospholipid, excluding SP. He has reported that the decrease in the PC content of the tissue during storage could not be followed directly because PB produced by enzyme hydrolysis was not separated from PC on thin layer plates. A linear relationship was, however, established between the increase in LPC and increase in FFA. The molar ratio of LPC/FFA was found to be 41:188 and 59:188 for leg and breast muscle respectively, compared with concentrations of PC in the total phospholipid, excluding SP, of 59% and 66%. The latter figures, however, are too high, probably by about 18 units because of the inclusion of PI in the PC fraction from the thin-layer plates. These findings indicate that lipid hydrolysis at - 27 18 C and -28 C was mainly due to the action of phospholipase A on the glycerophospholipids of the muscle to give corresponding lyso compounds and fatty acids. There was no evidence of any loss of SP. However Davidkova and Khan (1967) have published data on changes in the lipids of broiler chickens stored for two years at -18 C. The authors found that PC and PE decreased and FFAS and LPC increased. About 78% of the total increase in FFA content of the muscle was probably due to phospholipase B activity and the remaining 38% to breakdown of triglyceride. The fatty acid composition as percent of the lipids in fresh turkey muscle has been reported by Fishwick (1968) as shown in Table l. The data show that phospholipids were comparatively rich in polyunsaturated fatty acids, especially 28:4 and 22:6, whereas these acids were virtually absent from the triglyceride fraction. Leg FFA TC 06 PL 28 Breast FFA TG DG Fatty acid composition percent of the lipid of fresh turkey muscle * PL Fatty Table 1. acid 2079-,“44166644066 oooooooooooeooooo 103600688100agggl 2 22 6 4.700426807266311 oaooooooooosooooo 6 .820089900150903 01.. 12 1 4 88001949095502“ oSooooooooooooooo 1 0840085800540013 11 22 2 53029890 789663 8 a ooooooooroooooo {87.441822158.4a.tnuOZJnu118 1 112 2 680G364gagg 00 coco-cocoooooroor 1 o77ga7771000t06t 02 22 6066087017gg4g707 oooooooooooo so. 00420084gg0026926 13 1 5399g5918492970 9 oooooooooooooooro _035500157200220t0 2 121 918118.1018isayflqznu9.8012 rose-00.00.000.00 tn9931a7242112nuau8919.8.1nv .l. 112 4 14:8 l6:A 16:8 16:1 17:8 18:A 18.8 18:1 18:2 18:3 28:8 28:3 28 28:5 24:8 22:5 22:6 free fatty acid (06); saturated aldehyde (A) diglyceride (PL): triglyceride (TG); Phospholipid (FFA); From Fishwick (1968) 'k BXPBRIHBRTAL PROCEDURE Materials Chemicals and Solvents Chemicals and solvents used were of analytical or reagent grade. Fresh and glass distilled solvents were used directly, otherwise they were redistilled before use. Methyl Bsters Fatty acids and methyl ester mixtures used as standards for determination of the relative retention time of fatty acids in phospholipids and neutral lipids by gas liquid chromatography (GLS) were purchased from Supelco, Inc., Bel lefonte, PA. and Applied Science Lab., State College, PA. TLC Standards For the TLC analysis two types of standard mixtures were used. The polar lipid mixture contained cholesterol, phosphatidyl ethanolamine, phosphatidyl choline and lyso- phosphatidyl choline, while the serum lipid mixture con- tained specified quantities of lysolecithin, sphingomye- line, lecithin, phosphatidyl serine, phosphatidyl inositol, 29 38 phosphatidyl ethanolamine, phosphatidyl glycerol, cardio- lipin and phosphatidic acids. Thin Layer Plates Redi-Coats-G, 28x28 cm thin layer silica gel (8.58 mm) glass plates were purchased from Supelco Inc. Plates were activated before use at 118C for 38 minutes, and used for phOspholipid fractionation. Silicic Acid Silicic acid was obtained from Mallinckrodt, Inc. The fines were washed out with deionized distilled water followed by anhydrous methanol. Silicic acid was activated at 118 C for 24 hours before using for separation of polar lipids. Antioxidants Two types of antioxidants were used. An antioxidant coated salt (Diamond Crystal Salt Co. St. Clair, Michigan) was composed of 99.5% fine flake salt, 8.845% ascorbyl palmitate, 8.815% active vitamin E (mixed tocopherols, 8.8215% total), 8.8885% citric acid and 8.8425% silicone dioxide (anti cake). At 2% fat in meat, 2% of this salt is equivalent to 8.86% active antioxidant in the fat. This salt.was mixed with pure fine flake salt to achieve 8.82% antioxidant based on fat content of meat and 2% salt in the meat. The other antioxidant was Tenox 6 which contains 31 butylated hydroxyanisole(BHA) 18%, butylated hydroxytoluene (BHT) 18%, propyl gallate (PG) 6%, citric acid 6%, corn oil 28%, glyceryl monooleate 28% and propylene glycol 12%. (Eastman Kodak Co. Kingsport, Tennessee). Tenox 6 was mixed with salt not to exceed 8.82% antioxidant, based on the fat content of meat and 2% salt in the meat. Source of Salt Two types of pure salt (PS), calcium magnesium free salt (CMF) and fine flake salt (FFS), and two types of impure rock salt (RS), northern rock salt (NRS) and south- ern rock salt (SRS), were obtained from Diamond Crystal Salt Co. St.(3lair, Michigan. Analytical composition of the CMF and FFS was obtained from the same Company; (Tables 1 and 2 , Appendix A). Metal Ions Ferric chloride, ferrous chloride, cupric chloride and magnesium chloride were purchased from Sigma Chemical Com- pany, St. Louis, Missouri. Antibiotic Chlortetracycline hydrochloride was purchased from Sigma Chemical Company, St. Louis, Missouri. 32 Plate Count Agar The BBL Standard Methods Agar with lecithin and polyso- rbate 88 was used for aerobic plate count and was purchased from BBL Microbiology Systems, Becton Dickinson and Co., Cockeysville, MD. TEP Standard Solution The 1,1,3,3—Tetraethoxypropane (TEP) used for making standards for thiobarbituric acid analysis was obtained from Sigma Chemical Company. St. Louis, Missouri. Source of Meat Turkey meat used in this study was obtained from Bil- Mar Foods, Inc., Zeeland, Michigan. The white meat was represented by pectoralis minor, breast muscle, and the red meat represented by the deboned thighs. Thoroughly chilled samples were brought to the laboratory in nylon bags in insulated ice boxes with a minimum of delay. Methods Extraction of Total Lipids Total lipids were extracted by the method of Folch et a1. (1957) with some modifications. A specified weight of ground turkey meat was homogenized at 23,888 rpm for one minute in Virtis homogenizer ( The Virtis Co. Gardiner, 33 N.Y.) with 4 volumes chloroform:methanol (2:1). The homoge- nates were filtered through Whatman #1 filter paper using Buchner funnel. Four more volumes of chloroform:methanol (2:1) were used for reextraction of the residue. The combined filtrate was transferred to a separatory funnel and washed with 2 volumes of 8.74% potassium chloride solution and left at -25 C overnight to allow the chloro- form and aqueous layer to separate. The chloroform layer was collected and evaporated under reduced pressure at 48 C using a Rotavopor-R ( Buchi, Switzerland). Traces of chlo- roform were removed by a stream of nitrogen and the'crude lipid was redissolved to a constant volume in chloroform and transferred under nitrogen to vials with air tight teflon lined screw caps to be stored at —25 C for further analysis. Separation of Neutral and Phospholipids Total lipids were separated into neutral and phospho- lipids according to the method of Choudhury and Arnold (1968) with slight modifications. One half gram of crude lipid was added to 18 g of silicic acid in a 125 ml Erlen- meyer flask. Twenty five milliliters of chloroform was added, the mixture shaken and the flasks sealed under a stream of nitrogen and kept in the freezer 8 to 16 hours. The content of each flask was filtered through a.sintered glass funnel under vacuum. The silicic acid remaining in 34 the funnel was washed with five 38 ml aliquots of chloro- form. This filtrate contained the neutral lipids. The chloroform was evaporated and the neutral lipid sample dried and preserved in the same manner as total lipids. A known volume from the neutral lipid solution was heated in the oven to a constant weight at 185 C. The percentage of neutral lipids equals the weight of the extracted lipid over the sample weight times 188. The phospholipid content was determined by washing the silicic acid residues with three 25 ml aliquots of methyl alcohol. The methyl alcohol was evaporated and the phos- pholipid sample redissolved in chloroform to a specified volume. Phospholipid was quantitated and preserved for further analysis as in neutral lipids. Classification of Phospholipids The TLC plates were heat activated, then cooled for 38 minutes. A five microliter aliquot of phospholipid sample (58 mg/ml) or standards were spotted with a microsyringe, The spots were dried with a stream of nitrogen and the plates were placed in the developing chamber. The chambers were lined on all sides with saturation paper saturated with solvent just before use in order to obtain reprodu- cible results with straight solvent fronts. Two hundred ml 35 of fresh chloroform/methanol /water mixture (65/25/4, v/v) and new chamber liners were used for each run. The developed chromatograms were nitrogen dried for several minutes. Chromatograms were visualized with sulfu- ric acid—potassium dichromate reagent prepared by dissol- ving 8.6 g of potassium dichromate in 288 m1 of 55% reagent grade sulfuric acid. Spots were developed in an oven at 175 C for 38 minutes. Quantitation of phospholipid classes from the spots was accomplished according to the densito- metric method described by Rouser et a1. (1964). A Shimad- zu Dual-Wavelength Thin Layer Chromato Scanner Model CS-938 and Data Recorder Dr-Z were used in this investigation. Preparation of Methyl Esters Boron-trifluoride/methanol was used for converting neutral and phospholipids to methyl esters according to the procedure described by Morrison and Smith (1964). Gas-Liquid Chromatography (GLC) A Varian (Model 3788) chromatograph equipped with hyd— rogen flame ionization detector (FID) and CDS 111 integ- rator was used for GLC analysis of fatty acid methyl esters. Two meter x 2mm i.d. glass columns packed with 18% SP—2338, 68% cyanopropyl on 188/128 supelcoport (Supelco, Incu)‘were used. The carrier gas was nitrogen at 28 ml/ minute. The hydrogen flame was supplied with 388 ml air/- (. V J I“) at; w»- . - , - valued: 2p} L-;1 , " , y 1 - ,.“ .&;sb flflfljfl- ,‘x 'l.i;J-.~ fl A T.*L.‘_ , ' . , ,'-(liffl 1.." ’_’:;.:¢1) 'n'l .7':.: ‘7" . - ' l (!.bxu ifli'gba‘l ‘5! 3: Ln: 0‘ - x , d.‘ 7:: Germ.) 'H, -. (jg-31.)... .. ._ H - ALIIJE 0' asaasi: 5.4:Zw’14vr; - »‘ ' '. a" :0} O -Osianab on! u} in:;; ._~; 1 , _.P ,_ .‘qa an: obsmtna A .(EBGI) .2? s * ehsn {1 ,ad ::.3L (Ccfflfl 01.; 8E2 33 Isbo H Iarmsaa o1emoxdu .u{sl nknY dipasiavsW+i .nqlragizsivni aim: a; Diiu answ Sara £9930g)3 aging . . a ass-3a m in new -._’ _ . _-, r eniazivnoo xblfhasv law lona¢r:a\cabuootitzaqas°';; «:5 a “4mm hung“ a “glam.“ mt: 36 minute and 38 ml hydrogen/minute. The injection port temperature was 288 C and the detector temperature was 388 C. The column temperature was programmed at 158 C for 3 minutes, followed by an increase of 5 C/minute to a final temperature of 228 C. Qualitative identification of the fatty acid peaks was done by comparison with the retention times of the known fatty acid methyl ester peaks. Peak areas were quantitated by the integrator, and the results were expressed as perc- entage of the total area. Identification of dimethyl acetal of Hexa- and Octadecanal Phospholipid samples prepared as previously described were esterified with either sodium methoxide or boron tri— fluoride-methanol (BF3-MeOH). To verify that the compounds were not plasticide or other artifacts from the extraction procedures employed, phospholipid samples were also saponi- fied with alcoholic KOH solution. The nonsaponifiable part was separated and washed out with petroleum ether. The free fatty acids were extracted from the saponifiable portion with hexane after acidification with HCl. The volume of the remaining solution was reduced with nitrogen and 4-16 mg were dried and esterified with BF3-MeOH as previously described. GLC—MS spectrophotometric analysis for identification and quantitation of the two dimethyl 37 acetals was performed at Northern Regional Research Center (USDA, Peoria, Illinois, courtesy of Dr. Edwin Frankel). Thiobarbituric Acid Test (TBA) An improved extraction TBA method was used in this investigation (Salih et al., 1986). It is a modification of the extraction method of Witte et a1. (1978) with further modification as recommended by personal communica- tion with J. Bowers (1984). Ten grams of ground meat were homogenized with 35 ml of 3.86% perchloric acid in a Virtis homogenizer at a speed of 13,888 rpm (speed setting at 68) for one minute. The blended sample was filtered through Whatman no. 2V filter paper into a 58 ml Erlenmeyer flask. The Virtis flask was washed with 5 ml distilled water and filtered. Two 5 ml portions of the filtrate were pipetted to 58 m1 test tubes. Five ml of 8.82 M thiobarbituric acid in distilled water was added, the tubes were capped, the contents mixed thoroughly» and then incubated at room temperature in the dark for 15-17 hours. The absorbance was determined on Perkin-Elmer Spectrophotometer at 531 nm against a reagent blank in which 5 ml of distilled water was added in place of the filtrate. The standard stock solution of l,l,3,3-tetraethoxy- propane (TEP) was prepared by dissolving) 8.233 g TEP in a 38 1888 m1 of distilled water. This is a 18’3 M Tap solution. Ten ml of this solution were diluted to 588 ml in distil- led water. This stock solution contained 18’7moles TEP/Sml (and the one used for standard each time). Both TEP solu- tions were kept refrigerated. With each group of meat samples a standard curve was made with concentrations of 1, 2, 4, 6 and 8 x 18'8 moles TEP/Sml. To 5 ml TEP in each reaction tube 5 ml TBA reagent were added. The contents were mixed and tubes were capped and stored for 15—17 hours. TBA values were derived from the regression equation for the standard curve. The regression equation for the standard curve is: sample concentration = {[absorbance- intercept estimateJ/concentration estimate}/l8"8. The r2 should be not below 8.95 otherwise something may be wrong with the standard solution. Concentration is determined from the regression equation and multiplied by the constant (K), which is 8.774, to determine mg MAL/Kg meat which is the TBA value. K is derived from 93% recovery'of TEP to MAL, which weighs 72 g/mole. The recovery of MAL was deter- mined by adding known amounts of TEP to meat homogenates. The 5 ml filtrate analyzed was equivalent to lg meat . Thus the sample concentration OH x 18"8 moles of MAL is the same as A x 18"5 moles MAL/Kg meat. This is equivalent to 72 A x 18'5g MAL/Kg meat or 72 A x 18"2 mg MAL/Kg meat. 39 After correction for recovery (8.72 A x 188/93) the con- stant is 8.774. Nonheme Iron Nonheme iron was determined as described by Schricker et a1. (1982) with some modifications. Modification of the Schricker method included centrifuging the sample after incubation for 18 minutes at 27,888 x g to precipitate interfering pigments. Calculation of nonheme iron was based on weight of the sample after 28 hours incubation at 68 C to correct for any evaporation which may have occurred. Total Iron and Copper The concentration of total iron and copper were deter- mined using an atomic absorption spectrophotometer (Perkin- Elmer Model 2388). Iron and copper standards were used. A wet ashing procedure was used for preparation of samples. (Schricker et a1. 1982). Aerobic Plate Count A pour plate method was used as described by Deibel and Lindquist (1981). Raw and cooked turkey breast meat treat- ments analyzed include control (CNT), rock salt (RS). antibiotics (ANT) and cupric + ferrous ions (CuFe). Maxi- mum refrigeration and frozen storage times were 21 days and 12 months, respectively. Raw and cooked turkey thigh meat treatments included control (CNT), pure salt (PS), ferric 48 ions (Fe) and cupric ions (Cu). Maximum refrigera- tion and frozen storage times were 14 days and 3 months, respectively. Plates were incubated at 37 C for 48 hours. Moisture The A.O.A.C. (1975, 25.883b) procedure for the determi- nation of moisture was used. Total Fat The total solids left after moisture determination were used for total fat determination. The Goldfisch ether fat extraction method (A.O.C.S., 1974, 24.885b) was used. Protein The Micro Kjeldhal procedure was used to determine the total protein (A.O.A.C., 1975, 23.889). Buchi distillation unit (model 322), Buchi control unit (342), Buchi titrator (model E 526) and an integrator were used for automatic distillation, titration and calculation. Cooking The samples were vacuum packaged in polyethylene mylar laminate bags and cooked in water bath at 75 to 78 C to an internal temperature of 71C. 41 Experimental Design Experiment A The objectives of this experiment were: 1. To evaluate the effectiveness of an improved extraction thiobarbituric acid test (TBA) for monitoring lipid oxida- tion in poultry products. 2. To determine the correlation between TBA methods and sensory scores. For objective no.1, the TBA of cooked turkey breast meat, refrigerated at 4 C for 72 hours, was determinedby both the improved extraction and the distillation methods. The effect of BHA (125 micrograms/g fat) was comparedand the critical TBA value at which wormed-over flavor (WOF) could be recognized by sensory evaluation was determined for the two methods. For objective no. 2, freshly deboned unfrozen chicken breast was ground by a single pass through the'7 mm plate of a food grinder (Kichen Aid Stand mixer KSA with plastic grinding attachment, Hobart Corp”, Troy, OH). Sixty grams of ground meat were stuffed into 58 m1 polystyrene centri- fuge tubes and sealed with a screw cap. Ground meat was cooked in a water bath to different final temperatures (Table 2%. The water bath and center product temperature during cooking were monitored using thermocouples and a Honeywell recording potentiometer (Model F 2157). When 42 final temperature was reached the tubes were cooled in ice for 38 minutes before analysis and storage. The cooked Table 2. Experimental design for the heat treatment of ground chicken breast (1) ground chicken breast Sample Water bath Final center Cooking time number temp.(C) temp. of sample (C) (min) 1 95 56 5.6 2 = 65 7.2 3 = 78 8.7 4 = 88 19.9 5 = 98 28.8 (1) All heating treatments were evaluated in duplicate meat samples were reground and mixed immediately before testing to eliminate variation within the sample» Lipid oxidation was measured using the extraction method desc- ribed by Salih et a1. (1986) and distillation method of Tarladgis (1968). TBA tests were performed within two hours of cooking and after 48 hours of storage at 4 C. Cooked ground chicken was evaluated for warmed-over flavor (WOF) after 48 hours of storage at 4 C by sensory test. A panel composed of 8 males and 4 females, evaluated the samples. All panelists had previous training and expe- rience in evaluating WOF in meat. Ground chicken was served in covered one ounce plastic cups coded with random three digit numbers and presented in 43 a balanced block design. The ground meat was heated for 15 seconds to a temperature of 78 C in a microwave immediately before serving. Five samples were evaluated at each ses- sion. A freshly cooked sample was used as a reference. An unstructured 188 mm scale with ends of the scale labeled "no warmed-over flavor" and "strong warmed-over flavor" was used. A typical scoring form and instructions are shown in Table 2, Appendix B. Analysis of variance (ANOVA) was used for testing the significant differences attributable to main effects. Main effects tested were treatments, judges, replications and two-way interactions. Duncans new multi- ple range test was used to calculate significant differen- ces between treatments (Steel and Torrie, 1968). Experiment B The primary objectives of experiment B were to study the effect of two types of pure salt, two types of rock salt, a broad spectrum antibiotic, iron, copper and magne- sium on the development of oxidative rancidity during ref— rigerated and frozen storage of raw and cooked ground turkey breast meat. Turkey Breast Meat Processing: A 75.8 Kg batch of freshly separated pectoralis minor turkey breast muscle»was trimmed of external fat and ten- dons and cut into two halves. Consequently two identical 44 portions of meat were obtained, each portion weighed appro- ximatelyr36.8 Kg. (All meat was packaged under vacuum in polyethylene mylar laminate bags (Koch Equipment Co. Kansas City, MO), approximately 4.8 Kg each, and frozen rapidly to -25 C for one day. Packages of one portion were thawed in the cold room within 6 to 8 hours and ground twice through 3/8 inch plate using a Hobart meat grinder. The ground meat was divided into 18 portions, 3.6 Kg per portion and each portion was mixed with one of the treatments indicated in Table 3 using a Hobart Kitchen Aid mixer for one minute. Each treated portion was divided into 9 portions, wrapped in PVC film and stored as described in Table 4. The other 36.8 Kg in packages were thawed, ground, and mixed with the same treatments (Table 3), cooked, wrapped, and stored as in Table 4. Before analysis, each wrapped sample was reground by a single pass through the 7 mm plate of a food grinder (Kichen Aid Stand mixer KSA with a plastic grinding attach- ment). 45 Table 3. Treatments of experiment B 533e Description 1. CNT Control 2. CMF 2% calcium magnesium free salt (CMF)1 3. FFS 2% fine flake salt (rrs)2 4. NRS 2% northern rock salt 5. SRS 2% southern rock salt 6. ANT 188 PPM Chlortetracycline 7. FeCu 25 PPM ferrous ions + 25 PPM cupric ions + 2% CMF 8. Cu 58 PPM cupric ions + 2% CMF 9. Fe 58 PPM ferric ions + 2% CMF 18. Mg 58 PPM magnesium ions + 2% CMF 1CMF type high grade salt (Diamond Crystal Salt Co”,St. Clair, Michigan) 2Alberger Process Fine Flake Salt (Diamond Crystal Co" St. Clair, Michigan) 46 Table 4. Design of experiment B Raw or cooked turkey breast A samplee(3.6 Kg) for each treatment shown in Table 3 was divided into nine equal subsamples, wrapped in PVC film and refrigerated or frozen Refrigerated at 4 C for the Frozen at -25Cfor the indicatedstorage times indicatedstorage times 1. 8-day l. l-month 2. 2-days 2. 3-months 3. 7-days 3. 6—months 4. 14-days 4. 12-months 5. 21-days Experiment B Analysis: 1. TBA 2. Fat extraction 3. Fat separation into neutral and phospholipids 4. GC analysis of fatty acids 5. TLC phospholipid classification 6. Heme iron analysis 7. Total iron and copper monitoring by atomic absorption spectrophotometer 8. Fat, protein and moisture determination 9. Aerobic plate count 47 Experiment C The objectives of experiment C were to study the effect of fine flake salt, iron, copper and two types of antioxi- dants on the development of oxidative rancidity during refrigerated and frozen storages;of raw and cooked ground turkey breast and thigh meat. Turkey breast and thigh meat processing: A 36.8 Kg batch of each of freshly separated pectoralis minor turkey breast muscle and deboned turkey thigh meat were trimmed fromlexternal fat and tendons. All meat was packaged under vacuum in polyethylene mylar laminate bags and frozen rapidly to —25(L Each meat type was thawed in the cold room within 6 to 8 hours and ground twice through 3/8 inch plate using Hobart meat grinder and divided into two halves. One half was divided into 18 portions, 1.8 Kg per portion. Each portion was mixed with one of the treat- ments listed in Table 5 and divided into subsamples which were wrapped and stored as reported in Table 6. The second half was divided and treated like the first half, but the meat was cooked. 48 Table 5. Treatments of experiment C 56de Description 1. CNT Control 2. FFS 2% fine flake salt 3. Fe 58 PPM ferric ions + 2% FFS 4. Cu 58 PPM cupric ions + 2% FFS 5.,VE1 2% FFS coated with 8.845% ascorbyl palmitate, 8.815 active vitamin E (mixed tocopherols, 8.8215 % total), 8.8885% citric acid 6. VEFe VE + 58 PPM ferric ions 7. VECu VE + 58 PPM cupric ions 8. T6 2% pure salt coated with tenox 6 (BHA 18%, BHT 18%, PG 6%, citric acid 6%, corn oil 28%, glyceryl monooleate 28% and propylene glycol 12%) 9. T6Fe T6 + 58 PPM ferric ions 18. T6Cu T6 + 58 PPM cupric ions 1An antioxidant coated salt prepared by Diamond Crystal Co. 49 Table 6. Design of experiment C Raw or cooked turkey meat Breast or Thigh A sample (1.8 Kg) for each treatment shown in Table 5 was divided into seven equal subsamples (breast) or six subsamples (thigh), wrapped in PVC film and refrigerated orfrozen Refrigerated at 4 C for the Frozen at ~25Cfor the indicatedstorage times indicatedstorage times 1. 8-day l. 1-month 2. 2-days 2. 3-months 3.7-days 3.6-months(breast) 4. l4-days Experiment C Analysis: 1. TBA 2. Fat extraction (for thigh meat only) 3. Fat separation into neutral and phospholipids 4. GC analysis of fatty acid 5. Heme iron analysis 6. Total iron and copper monitoring by atomic absorption spectrophotometer 7. Fat, protein and moisture determination 8. Aerobic plate count 58 Statistical Analysis Statistical analyses were performed using Statistical (Analysis Systems (SAS, 1985):for the five factor analysis of variance (ANOVA) with nested design of TBA results of experiments B and C together. Statistical Package for the Social Sciences (SPSS, 1984) was used for the 3 factor ANOVA of fatty acid profile in turkey breast neutral and phospholipids. MSTAT (1985) was used for ANOVA for aerobic plate count, neutral and phospholipid and phospholipid classes. Microstat (Ecosoft, 1984) was used for analysis of taste panel scores, and to obtain standard deviations, simple correlation and regression coefficients. The signi- ficance between treatments was determined using either the Tukey test or Boonferroni t-test for multiple comparison analysis, after a significant F was determined. Graphs were plotted using Plotit (Eisensmith, 1985). RESULTS AND DISCUSSION TBA Methods and Sensory Scores The relationship between TBA measurements by an improved extraction method or the distillation TBA method of Tarladgis et a1. (1968) and taste panel scores for monitoring lipid oxidation and WOF in poultry meat were investigated. TBA numbers obtained using the distillation method are 1.4 to 2.8 times larger than those of the improved extrac- tion method (Table 7). These results are comparable to the findings of Witte et a1. (1978) who reported that the TBA values determined by the distillation method were approxi- mately twice as large as those determined by the extraction Table 7. Comparison between distillation and improved extraction TBA method1 Meat type Distillation Extraction Ratio raw breast 8.39 1 8.38 4.25 I 8.11 2.8 X cooked breast 18.87 1 8.32 7.18 i 8.26 1.4 X raw thigh 11.52 i 8.31 6.18 1 8.19 1.9 x cooked thigh 14.16 1 8.37 7.89 1 8.21 1.8 X 1Means represent six determinations 51 52 method. The heat used in distillation may increase quanti- ties of aldehydes, disrupt certain carbonyl products thought to occur by'a neaction between malonaldehyde and amino acids, pyrimidine or protein (Buttkus, 1967L. Heat used in distillation may speed up the oxidation process (Witte et al., 1978 and Siu and Draper, 1978). The latter was substantiated by the decrease in TBA number when adding BHA before blending. The decrease in TBA number was about 18 and 15% in distillation and in the improved TBA methods, respectively (Tables 8 and 9). These findings agree with results of Pikul et a1. (1983) which demonstrate a signifi- cant decrease in TBA value when BHA was added. Table 8. Effect of BHA on TBA values measured by the distillation methodl'2 Meat type Control + BHA % Decrease raw breast 8.39 :_8.38 6.44 :_8.27 23.8 cooked breast 18.87 1 8.32 8.76 I.0°29 13.8 raw thigh 11.52 i 8.38 9.14 1 8.28 21.8 cooked thigh 14.16 1 8.37 11.94 1 8.38 15.7 1BHA: 125 micrograms/mg fat 2Values represent means of six determinations 53 Table 9. Effect of BHA on TBA values measured by the improved extraction method1 Meat type Control +BHA % Decrease raw breast 4.26 :_8.11 3.45 1 8.12 18.9 cooked breast 7.18 i 8.25 6.35 i 8.22 18.6 raw thigh 6.18 1 8.19 5.86 1 8.27 17.1 cooked thigh 7.89 i (5.21 6.91 i 8.22 12.3 1 Values represent means of six determinations Heat in the distillation method is used to free malon- aldehyde from its bound state with protein, while perchlo- ric acid is the releasing factor in the extraction method. The standard deviation in the distillation method is gene- rally higher than that in the improved extraction method (Table 7). The relationship between the improved extraction and distillation TBA methods is good. The correlation coeffi- cient was higher when the cooked chicken breast meat was analyzed after 48 hours of refrigeration than when done 2 hours after cooking (r = 8.934 vs. 8.854, Figures, 1 and 2). Witte et a1. (1978) reported that the correlation coefficient between TBA values determined by the extraction method and the distillation method was represented by an r value of 8.845. 54 8 *r I ** r '* r ** r l ”r I ' i"* 1 ' 7.. r a 0.9663 - ‘ r2= 0.9337 . 6.. o - 0.4998 _ b - 0.3165 A“ TBA NUMBER BY EXTRACTION METHOD T o U I f I r r T j I ‘U 1 o 2 4 6 a 10 12 1'4T1161118 20 TBA NUMBER BY DISTILLATION METHOD Figure 1. Relationship between the improved extraction and distillation TBA methods for chicken breast meat refrigerated at 4 C for 48 hours. 55 34 TBA NUMBER BY EXTRACTION METHOD N L r =8 0.8536 .. r2= 0.7286 c - 0.0625 b - 0.2995 Figure 2. U I' j I I i f j T 2 4 6 8 10 TBA NUMBER BY DISTILLATION METHOD Relationship between the improved extraction and distillation TBA methods for chicken breast meat refrigerated at 4 C for 2 hours. 56 The correlation coefficient between the improved extra- ction or distillation TBA values and the panel scores is demonstrated by Figures 3 and 4 (r = 8.858 for the extrac- tion and 8.835 for the distillation method). The relation- ship of both methods with panel scores was high indicating that TBA is a valid objective test used with sensory eval- uation for monitoring warmed over flavor in poultry meat. Data showed that the TBA values which could be consi- dered as an indication of warmed over flavor development in cooked turkey breast meat as proved by the triangle test are 1.28 and 3.41 for the improved extraction and distil- lation methods, respectively (Table 18). Table 18. Relationship between TBA methods and taste panel for cooked turkey breast1 TBA Values TTBA Values Differenceby Meat type Extraction Distillation Triangle test freshly cooked 8.32 i 8.82 1.87 1 8.87 -(p<8.85) 24 hr after cooking 1.28 1 8.81 3.41 1 8.15 +(p<8.8l) 48 hr after cooking 4.31 :_8.18 8.56 :_8.17 +(p<8.81) 1 Means are averages of 5 determinations - = Difference is not significant at the indicated level. + = Difference is significant at the indicated level. S7 ' l I “r '1 r r ' .4 ' 4 . r - 0.8504 . 1 L 1 r2- 0.7232 . O J .. 3 5 o - 2.5612 .. 23‘ J b 0441a ' ' l u. A " ° DJ 0: 4 ' . 8 m ‘d d _J . uu E 4 - 0. + J I'<" d " 4 i J . 2 - . r . . . o i E 3 1 E 5 TBA NUMBER (MG MAL/KG MEAT) Figure 3. Relationship between taste panel scores and TBA numbers for the improved extraction method. 58 6 . ‘r B" T I ‘T I I r I* r I '4’1 . 1 ‘1 1 r =- 0.8347 . A LI. 1 12" 0.6968 - CD fiifi A 3 o - 2.2443 M UK ‘ a E b - 0.1690 ‘ J m 1 n: 1 ‘ . 8 m H ‘ .J A DJ E 4 . 0. J . hJ '6 3-4 .1 S . . fi 1 '1 1 2 1* I*"* T . I r I " I' . I ”T TI *E‘WT 1 0 2 4 6 8 10 12 14 16 18 Figure 4. TBA NUMBER (MG MAL/KG MEAT) Relationship between taste panel scores and TBA numbers for the distillation method. 59 Dawson et al. (1975) assumed that TBA values above 2 (using distillation method of Tarladgis et a1, 1968) would be associated with the development of rancid flavor in a turkey pattie. In conclusion, the improved extraction TBA method was found to be fast, precise, easier to perform and as accu- rate as the distillation method. The following points should be considered when using the improved extraction procedure. 1. High fat meat or high levels of microbial contamination may cause a turbid filtrate. 2. Before the improved extraction method is used for new samples, an absorbance scan of TBA-MAL complex should be taken to check for interfering compounds. 3. Boiling to develop the TBA—MAL complex may cause the formation of an orange colored complex if carbohydrate is present in the sample. Lipid Oxidation, Factors and Variables For Experiments B and C Together Meat Type (Light vs. Dark) Turkey thigh muscle was more susceptible to lipid oxi- dation than turkey breast muscle for each of the treatments (Figure 5) as measured by TBA test. Maximum TBA values are 68 V _. EEZT_::—::::::::::::: :::: _:_::::::E_::: r mg _ =:::_m::_2::22:_:::_::::: H .II E_E::__T:::_H:::_::: é U _ S—EZJFEEE::::::::22:—_::: H EEEHIH::«EZEEEHHEEZZ Em: ox\._<2 02V mmmzaz (E _ mm ::_Et:_:_::E: . 4 . q + . . a . _ 4 B . . . wt .0 .a 4. 1. a; a: nu cm VE ‘Wfi \Emu TB 'flFb 1300 CNT l$3 Fb FIgure 5. Effect of treatments and turkey meat type on TBA values. (Fe); ferric ions cupric ions «an; vit.Eh ascorbyl palmitate, citric acid (VE): fine flake salt (FFS); (CNT); tenox 6 (T6) control 68 _. E::1::::::::ESE—HES:— r h :_::1:::_:_:::::::: 1&1 VB ‘Wfb _ =::::_::::::_::::::::::: r m =::E::_!::::::_:::: gash. _ =_:_d.::_:::::::_E::::::Z:— p = _ ::::: _ im: 21.: 2E : :_: mm 1 r _ mm EEZETEZEZEEC . 1 1 . a 1 _ . a . . . . . a . at .0 .0 4. 1. 9. 1. nu CE: ox\._<2 02v mmmzaz <9 1 mam 13 “5% FUR! CNT l$3 Fb FIgure 5. Effect of treatments and turkey meat type on TBA values. (Fe); ions ferric cupric ions «an; vit.Eh ascorbyl palmitate, citric acid fine flake salt (FFS); (CNT); control tenox 6 (T6) (VE); 68 EZZIBREAST 'EEEE'flflGH U _. F:—:1_::::::H:::::E::: gamma .71 :2:—“::::_::_:::_::: _ 52.:___:::::__:::::=:::::: r p _:_::::_1:::_::::::: guy _ EHEHETHEH:::::::E::H::FFS—E . _ E:_::::_E:::—EE _::: r CNT IWS Fb _ E: : ET: :EHE _:: J! 7. I d d .- d d d - J q 8 J- d .0 .D t. 1g 0‘ 1. nu CE: 83% 02V mmmzaz <9 \Ebu TB 'flFb lion 1&1 VB “5% FIgure 5. Effect of treatments and turkey meat type on TBA values. ions (Fe); ferric cupric ions Khfl; vit.Eh ascorbyl palmitate, citric acid (VE); fine flake salt (FFS); (CNT); control tenox 6 (T6) 61 reached after 7 and 14 days of refrigeration at 4 C, and three months and six months of frozen storage at - 25 C. The difference in TBA values for the two types of meat is highly significant (P<8.88l). These findings agree with Marion and Forsythe (1964) who reported that the TBA value was higher for the turkey red meat stored at 4 C for l to 7 days than for turkey white meat stored similarly. They attributed this difference to the higher total lipid con- tent of red meat. In an earlier work Barron and Lyman (1938) attributed these differences to the larger amount of myoglobin in red muscle compared to the white muscle. The changes of TBA values for both thigh and breast meat with storage time are shown in Figures 6 and 7 for refrigerated and frozen meat, respectively; The peak of lipid oxidation as measured by TBA is reached after 2 weeks of refrigeration (after which a decrease in TBA was noticed in a preliminary study) or 3 months of frozen storage. Cooking Meat stored for a short time after cooking develops warmed-over flavor much faster than uncooked meat. Labuza (1971) suggested that the rapid rate of oxidation in cooked meat is due to the denaturation of myoglobin during the cooking process. The increase in lipid oxidation as repre- sented by TBA values was proportional to the storage time between 8 and 14 days of refrigeration at 4 C. 62 3 0| TBA NUMBER (MG MAL/Kc MEAT) .p 3- .. 2 .. J . 1- d O I If I’ T ‘I I I I II 1 —. I ‘I ‘I‘I I 0 2 4 6 8 1O 12 14 16 18 STORAGE TIME (DAYS) Figure 6. Effect of meat type and refrigerated storage on TBA values of turkey meat. 63 8 f I fi I f F I A E 5 '1 o d i .. _| < 2 0 q 2 V 1 0: LIJ .. m 2 . D Z - < m4 . '— 1d a q d o I l T l f I T 0 2 4 6 8 STORAGE TIME (MONTHS) Figure 7. Effect of meat type and frozen storage on TBA values of turkey meat. 64 Lipid oxidation developed more in the cooked than in the raw turkey meat (PK8.881) as shown in Figure 8. The TBA values for the cooked meat reached a maximum after 3 months of frozen storage at -25 C while the raw meat reached a maximum TBA value after 6 months (Figure 9). Lipid oxidation in cooked meat was higher than raw meat throughout storage (Figure 9). Sato and Hegarty (1971) proposed that cooking disrupts the muscle membrane and results in exposure of labile lipid components to oxygen and other catalysts. Igene et a1. (1985) suggested that cooking might I iberate most of the heme iron which acts as an active prooxidant. These assumptions might explain the higher TBA values after cooking in turkey red meat than in turkey white meat (Figure 18). TBA changes for raw and cooked turkey breast and thigh with storage for refrige- rated and frozen storage are represented in Figures 11 and 12, respectivelyu An interaction between meat type» coo- king and storage time could be observed from these Figures. 65 a " l I I“ I’ I I r I I I I I -— 7‘ com-1 E :6-1 .. g2 RN! 4 ES- 4 :2 1 1 o. .J :4 V 15‘ ‘ J 00 .3 4 :1 4 :3 22 .. E. 1 1 Oj‘TTt T j ‘r‘j ‘fi 6 2 4 éja 161121411613 STORAGE TIME (DAYS) Figure 8. Effect of cooking and refrigerated storage on TBA values of turkey meat. 65 a I’ I ”I I I I I I I I I r J 7‘ coach 1 g . 2 6- .. (3 I RN! 4 x E 5- 4 I 1 I o . .J 2 4 V J E 1 00 13‘ 4 z . . D 2: :2 _ < E 1 14 '1 a. q 0 II"I II IF’I II II* I 4. 1*‘j" I I I 1 II o 2 4 6 a 1012 1141618 STORAGE TIME (DAYS) Figure 8. Effect of cooking and refrigerated storage on TBA values of turkey meat. 66 8 ' T I I r T J 74 - ’5‘ . 2 6d OOOKED ' o J 4 § .1 .J 5 - m ' < 2 1 .1 o ‘d .1 2 V E 3. m "'1 2 .1 . 3 1 Z 2 < ._ E 1 ‘ 14 - O I f I r f 1* I 0 2 4» 6 8 STORAGE TIME (MONTHS) Figure 9. Effect of cooking and frozen storage on TBA values of turkey meat. 67 3 '71 6 v 6 3%» k\‘ \\\ '0 6 \\ 6 O .6 \ W\ v.v .v .v‘v o’ (A?! 1 8 v.v.v; 2.? O 6 .9113 g1 .’ TBA NUMBER (MG MAL/KG MEAT) :- 11. £4 \\\\\\‘\\\\ \\\\\\\\ O .' 41 3 ‘ .6 O %’ '6 .6 o 94’ H’ 1% BREAST THIGH Figure 18. Effect of cooking and turkey meat type on TBA values. 1 Bars represent standard errors. 68 8 I ' I ' I I ' l I I l I . menu-11611 . 7-1 .. E ‘ ‘ mm 2 5" - 2.. - o . ' mum ) X E 5. mm .1 z 1 «1 o 4. - 3 95 / m 3-1 .. 4 2 «I D z 2 .. é . 1_ 4 1 ‘l 0 . 4 6 8 1O 12 14 16 18 STORAGE TIME (DAYS) (2 Figure 11. Effect of cooking, meat type and storage on TBA values of refrigerated turkey meat. 69 I I I I I I 1 . 94 4 A '1 u E ‘ . 3 7‘ COOKEDTHGH '1 E _ < 2 1 RAM THGH « (D «1 :5 (KKIEDEREMW ‘-’ 1 O: RNIIREEW 4 hJ m 2 D - z d g 1 1 , 4 ll ‘ O , , . T . , . 0 2 4 6 8 STORAGE TIME (MONTHS) Figure 12. Effect of cooking, meat type and storage on TBA values of frozen turkey meat. 78 Treatments Lipid oxidation in foods can be catalyzed by certain divalent cations, even when present in trace amounts. Metal cations may come from packaging materials, proces- sing equipment or be present in added ingredients such as salt, spices or flavorings. In general, TBA results indicated that the relative prooxidant effect of divalent cations and salts on turkey meat were in the following order: Fe3+> Cu2+> rock salt (RS)> fine flake salt (FFS) > control (CNT) (Figure 13). A significant difference (P<8.85) was detected among treat- ments. These findings agree with Younathan and Watts (1959) that ferric heme was the active catalyst of lipid oxidation in muscle system. However, Brown et a1. (1963) and Hirano and Olcott (1971) reported that heme with iron in either the ferrous or ferric states were not different in promoting lipid oxidation. The most important mechanism involved in heme catalyzed lipid oxidation has been consi- dered to be the catalytic decomposition of hydroperoxides to generate free radicals» In contrast, Sato and Hegarty (1971), Love and Pearson (1974) and Igene et a1. (1979) concluded that nonheme iron has a stronger prooxidant effect than heme iron. Heme compounds were reported to have little influence on the development of off-flavors or TBA reactive materials in meat. Ferrous iron has been 71 f??? ?????????, gain]. 1’ 4 I ??.?..???? ?????/????, V»? // ?????????. ////////////////////////////////////////A 1. ”fig/17g??? Twme/////////////////////// ‘11 d 1 d 8 -1 4 d )1 11 d1 d .3 4. iv n4 1. nu 962 93% 95 Emznz <9 ferrous cupric ions (Cu); rock salt (RS); vit.E, ascorbyl palmitate, citric acid (MI lg VE \EHIWEMIB ‘flfibflflm ferric ions (Fe); pure salt (PS); meat. 6 (T6) CNT!!! RS lfiCuFb tenox Bars represent standard errors. control (CNT); + cupric ions (FeCu); Figure 13. Effect of treatments on TBA values of turkey magnesium ions (Mg); (VB); 1 72 reported to have greater prooxidant activity than ferric iron (Smith and Dunkleyy 1962; Brown et al., 1963; Wills, 1965; O'Brien, 1969 and Bidlack and Tappel, 1972). However it has been reported by Govindarajan (1973) that catalysis by nonheme iron appeared to be independent of the ionic state of the metal, although the type of model system used and the concentration of iron or a balance between ferrous or ferric iron affect results (Sato and Hegarty, 1971). The effect of cooking on the release of nonheme iron was investigated in both turkey breast and thigh meat in order to determine the effect of this catalyst on warmed- over flavor development (Table 11L.The presence of fine flake salt (2%) in meat samples had no effect on the release of nonheme iron (P<8.85). Breast meat samples which had 58 micrograms/g added iron, were found by atomic absorption to have 41.38 1 1.82 and 58.81 1 2.13 micrograms/g total iron in the raw and cooked meats, respe- ctively. The corresponding values for thigh meat contain— ing 58 micrograms/g added iron were 56.87 1 1.96 and 53.8 1 2.17 micrograms/g total iron for the raw and cooked meats, respectively. The nonheme iron content of raw and cooked turkey breast meat was very large when compared to total iron content (93 to 95% nonheme iron). The values for turkey 73 Table 11. Total iron, nonheme iron and percentage of nonheme iron in turkey meat.1 Total iron Nonheme iron Nonheme iron Meat type (microgram/g) (microgram/g) (percent) raw breast 5.9918.38 5.6818.23 93.7915.92 cooked breast 5.9318.29 5.6318.25 95.8211.82 rawthigh 15.1318.82 l3.7811.47 ELJ918.25 cooked thigh 15.1318.59 l4.l918.67 93.8615.48 1Values represent means of 6 replicates nonheme iron are in the range reported for other specie as pork, lamb and beef of 5.2, 7.8 and 9.8 micrograms/g non- heme, respectively (Schricker et al., 1982). A similar proportion of nonheme iron (91 to 94%) was found forthigh meat although the total and nonheme iron content in turkey thigh meat were about three times greater than breast. The percent of nonheme iron in turkey meat is much higher than that reported in other species. This may be explained by the low myoglobin content of poultry meat. Yamauchi (1972) showed that chicken breast contained 8.263 mg myoglobin/g meat compared to 8.642, 3.169 and 4.387 mg/g in pork, mutton and beef, respectively. Percent of nonheme iron in cooked and raw turkey meat was not significantly different (P<8.85). The method of Schricker and Miller (1982) used for this analysis may be not sensitive enough 74 to detect the changes in nonheme iron as it is present at low concentrations in turkey meat. The method of Igene et a1. (1985) was used in this investigation and found to have almost the same or lower sensitivity. In conclusion, the nonheme iron level is very high (about 93%) in turkey meat. It could be one of the major factors which contri- bute to the lower stability to lipid oxidation in turkey meat, especially thigh meat which has about 3 times more total iron than the breast meat. 'However the nonheme iron is 28.9 micrograms/g in heated fresh beef as reported by Schricker et a1. (1982) and yet beef is one of the more stable meats to lipid oxidation. Thus in a complex meat system there is no one single factor which controls meat stability. Polyunsaturated fatty acids and the ratio of reduced to oxidized forms of iron might be major factors. Copper acted as a prooxidant and produced TBA values close to that produced by iron in turkey meat (Figure 13, Page 71). Results agree with the data reported by Tichiva— ngana and Morrissey (1985) which indicated that copper catalyzed oxidation followed a pattern similar to that for iron catalysis, but copper was slightly less effective. ResultSIOf copper analysis are presented in Table 12. Breast meat with 58 micrograms/g copper added analyzed by atomic absorption was found to contain a lower level. the results for thigh meat were similar except there was 75 greater total copper, particularly in the cooked samples. In meat without added copper the level in thigh was 2.5 to 3 times greater than in breast. Table 12. Copper concentration in turkey meat1 Copper Added copper Meat type (microgram/g) (58 micrograms/g) raw breast 8.8118.35 38.421l.21 cooked breast 1.8918.43 37.8311.53 raw thigh 3.8618.37 39.1711.38 cooked thigh 2.4918.39 47.2811.18 1Values represent means of 6 replicates The difference between the copper content in breast and thigh might be one of the factors which make the thigh meat more labile to lipid oxidation as monitored by the TBA method. Castell and Spears (1968) reported that small amounts of copper ions play an important part in the deve- lopment of oxidation defects in fat-containing food products. Cupric salts, in very special cases, were reported to have inhibitory effects in lipid oxidation and this was attributed to chain termination (Ingold, 1962). Many aspects of lipid oxidation need to be investigated in order to fully comprehend the prooxidant and inhibitory role 76 played by copper ions. These might include oxidation of cell components other than lipids, the role of phosphorous- containing compounds, especially nucleotides, PH effects and the effect of high temperature. Fine flake salt had a significant prooxidant effect (P<8.85) in these studies. The literature spanning oxida- tive deterioration in foods leaves conflicting conclusions insofar as the role of salts is concerned. The prooxidant activity of salt appears to vary with conditions. Under certain conditions salt has been reported to have inhibit- ory or antioxidant effect (Tarladgis et al., 1968 and Zipser et al., 1964). The effect of two antioxidants was studied in order to determine the most effective one in turkey meat. Tenox 6 was more effective in controlling the prooxidant effect of both iron and copper. The use of Tenox 6 resulted in signi- ficantly lower TBA numbers (P<8.8l) when mixed with the meat containing salt and added iron and copper. Tenox 6 was the more effective antioxidant in thigh meat containing pure salt (PS) than was the case for breast meat with PS (Figure 5). The antioxidant which contained vitamin E, ascorbyl palmitate and citric acid produced a slight inhi- bition of the prooxidant activity of iron and copper,but the effect was not significant compared to the treatments which contained iron or cOpper. Dugan (1976) indicated 77 that the most effective antioxidants in food systems func- tion by interrupting the free radical chain mechanism, which include BHA, BHT and tocopherol. This group consists mainly of phenolic compounds capable of donating a labile hydrogen. Tenox 6 may have exhibited more antioxidant effect as it contained 18% BHA, 18% BHT, 6% PG and 6% citric acid. It has been reported by Kraybill et al. (1949) and Marion and Forsythe (1964) that BHA was very valuable in delaying autoxidation in foods. BHA with BHT and citric acid provide good processing stability and help stabilize many processed foods (Dugan, 1976). This antioxidant control is achieved as a result of the syner- gistic interaction of antioxidants. Shown in Figure 14 is the effect of treatments and refrigerated storage on lipid oxidation of raw and cooked turkey breast meat. In raw breast the TBA number was about 2.5 to 4.25 on 8-day analysis for treatments which have Fe, Cu, VEFe, T6Fe and VECu. The treatments containing T6Cu, FFS, VE, T6 and CNT have a TBA number below 1. This indi- cates that lipid oxidation started immediately after mixing of treatments with the ground meat for the first group of treatments which have strong prooxidant effect. In the cooked breasts TBA values start between 1.25 and 3.2 and tend to cluster together throughout storage. The effect of treatments andfrozen storage on lipid oxidation of raw and 78 .umm0un awxusu pODODOOMHHOM mo mosam> «we so mucmpfixofiucm can m=o4 Honda .uamm .ommuOum no uomuum .va musmwm 6534 was 35.95 629 was 85.95 .....-4.P.4.«Lo ...m..n..w..m..w-~19.m.o ' ‘ ' l I II’IW r '5' <1 '16' u u E! I: \ .1. I1 J, . (Ivan {DI/"MI 911) 838mm V81 3 -\ r I. ' 1 (111311 ox/Tvn 911) 113mm V81 49;; . .1 a 3 <1: 0: 11 V omxooo T I! ‘ b L I- b I. I- n b P b n . 79 cooked turkey breast is Shown in Figure 15. Raw frozen turkey breast with added iron tended to have higher TBA values throughout storage than was the case with other treatments. In the cooked frozen breast the effect of treatment is less since TBA values are more similar. The maximum TBA inIall the cooked, breast treatments was rea- ched after 3 months of frozen storage vs. 14 days of refri- gerated storage. The decline seen in the sixth month of frozen storage could be explained by the fact that malonal- dehyde is not a final product of lipid oxidation and it is labile for further decomposition with prolonged storage. Figures 16 and 17 show the effect of treatments and storage conditions on lipid oxidation of raw and cooked turkey thigh. The general pattern is similar to the corre- sponding one of the breast meat, but in the thigh the range of TBA values is higher. Tenox 6 was a good antioxidant for both raw and cooked turkey meat containing copper and iron, but it was more effective in raw meat (Figure 18 and 13, page 71). Tenox 6 had better antioxidant control on fine flake salt in the raw meat, while the vitamin E, ascorbyl palmitate and citric acid combination had only slight antioxidant control on samples which have Fe, Cu or FFS, whether in the raw or cooked turkey meat (Figure 18). 88 muCflfimewucm 0C6 mCOw H6805 .ummmun moxusu cmnouu mo mw24m> dme co $922: was 8365 m r 4 LI) m I -O I 1‘ ((1 omxooo - - ¢ '4'" N v- ' I a (Ivan ox/Tm 9N) UBBINDN YBL .uamm .wmmuoum mo yummwm .mH musmflm $9.63 was 856.5 . 4 . 4 . H . J, , .r (Ivan ox/Tvu ON) aaannN V8.1 V I I h 81 .nmflnu aoxusu pmumuomfiuwwu mo mosam> ¢me co mucmpfixoflucm can mcofi Hoods .u4mm .mmmuoum mo u06uum .ma Ouzmwm 493. was gem 493V 9.: 8565 2.374.. a...a...4.4.4.m.o 2.....4...«...a..-a.~14.4149 (Ivan sax/1m an) aaannN V8.1 (man :1an on) UBBNDN V81 82 by .nmwnu moxusu cmN0uu mo mosao> dma co mucocfixofiuco can meow Hmuoa .uamm .omououm mo uDOuum .54 musmfim $92on was 88.5 4 LP » Ll H I rj ' ' V ' U r—' ‘ j r V r V (Ivan :1an 911) mm V81. $5203 ua=uo<¢oa f -- r 1 a F g 4 to j 1 g 1 a: \ r 1 r E r. . 89 f I IN 4 on! . r .6 10 . z .. 22m 14. r fl P b bl IIbI P P o“ (Nan ox/‘MI on) 1133111111 V81 83 d (1 1 d1 g - JG — d - d1 - I 4 H:_::::_:___::::::: E ::_::::_H::::—::: ~22 U T _ E::::::__::::_JI_:::_Zn r 5?:2_::—:::::::: _HHHHHHHHHHHHHHHHuHgHHHHHHHHHHHHHHHLLHHHLLHM. _§. _ _.._...................4................_... U _ =::::::r_:EHEEEE:2:: r T. :i:::_:::EE::_Wd:_ 8 1i #1 u 4 4 T d! 1 u 1 4|- 1 q (8 6 5 ‘v 3 2 1 0 CE: 93% 03 mmmzaz <9 "Eh cm VE ‘Wfii \EBU T6 “F3 CNT (*3 F5 Effect of treatments and cooking on TBA values Figure 18. of turkey meat. fine flake salt (FFS); . . . cupric ions (Cu); vit. E, ascorbyl palmitate, Citric ac1d ferric ions (Fe); (CNT); control tenox 6 (T6) (VB); 84 Lipid Content of Fresh Turkey Breast and Thigh Heat The composition of lipids extracted from turkey meat are presented in Table 13. The proportion of phospholipids in the total fat was higher in breast than in thigh meat (Table 13), but the total amount of phospholipids was higher in thigh meat, 0.66% vs. 0.25% of meat weight. Table 13. Lipid content of fresh turkey breast and thigh meat Meat type Lipids Breast Thigh Total lipidsl 0.75 _+_ 0.02 6.02 _+_ 0.06 Phospholipidsz 33.47 i 1.11 11.01 i 2.36 Neutral lipidsz 66.53 1 1.11 88.99 i 2.36 1Expressed as percentage of total tissues 2Expressed as percentage of total fat These findings are in agreement with Wangen et al. (1971) who concluded that on an absolute basis, turkey thigh had the larger concentration of phospholipids but as a percentage of total lipids, phospholipids comprised a larger percentage in breast lipids. A similar trend was 85 reported by Marion and Miller (1968) and Pikul et a1. (1984) for chicken breast and thigh tissues. Effect of Frozen Storage and Cooking on Turkey Lipids Content The proportion of phospholipids in the raw breast meat during frozen storage is shown in Table 14. A decline in phospholipid percentage during storage became significant (P<0.05) after one month of frozen storage. Table 14. Effect of frozen storage on turkey breast phospholipids (PL) and neutral lipids (NL) percentage of total lipidslv2 Lipids Cooking Storage time (months) at ~25 C i 1 3 6 12 PL raw 35.35a 33.61b 32.77c 32.73c 32.86d NL raw 64.65 66.39 67.13 67.27 67.14 PL cooked 33.17a 30.41b 30.19C 31.48d ' 31.09d NL cooked 66.83 69.59 69.81 68.52 68.09 1Values in the same lipid group and in the same row and bearing the same superscript are not significantly different (P<0.ES) 2Each value represents a mean of 10 determinations 86 Results also showed that cooking decreased the propor- tion of phospholipids in total lipids. Level of phospho- lipids in cooked breast meat were 33.17, 30.41, 30.19, 31.48 and 31.91% of total lipids at 0, l, 3, 6 and 12 months of frozen storage, respectively. The lowest ratio of phospholipids to neutral lipids in the cooked breast was observed after 3 months of frozen storage. The decrease of the phospholipid proportion with frozen storage and cooking corresponds to the variation in the levels of malonaldehyde as measured by the TBA test (figure 11 and 12 Pages 68 and 69) as well as the changes in fatty acid profile shown later. The results of this study showing decreased proportion of phospholipids due to frozen storageeand cooking are in agreement with the work of Acosta et al. (1966) who repor- ted an apparent decrease in the phospholipid content of turkey tissues frozen for 180 days at -25 C. ‘Davidkova and Khan (1967) and Fishwick (1968) reported a similar trend and concluded that the phospholipid decrease with time of storage at -10 C or below appeared to be largely indepen- dent of the actual temperature of storage. They explained the decrease in phospholipid content by the loss of PC and PB. The increased yield of triglyceride after storage was explained by Davidkova and Khan (1967) as resulting from deteriorative biochemical changes which make the 87 association or binding between triglycerides and proteins less strong, permitting greater extraction. Effect of Treatments on Composition of Turkey Lipids Treatments with pure salt, rock salt, antibiotic and metal cations have minor effects on the proportions of lipids, although some significant effects could be detected (Table 15). Treatments had no detectable effects on lipid Table 15. Effect of treatments on the percentage of turkey breast lipids Treatments Phospholipids Neutral lipids CNT2 32.48 67.52 cd CMF 32.79 67.21 e FFS 32.79 67.21 e NRS 31.99 67.91 b SRS 31.87 68.93 a ANT 32.74 67.26 e Fe2+Cu2+ 32.49 67.51 d Cu2+ 33.05 66.95 f Fe3+ 32.81 67.19 e Mg 32.26 67.74 bc 1Values in the same lipid groupbearing the same superscript are not significantly different (P<0.DS) 2 Definitions are in Table 3, Page 45 88 oxidation as monitored by changes in fatty acid profile. However, the TBA test did show significant treatment effects (see prior discussion and Figure 13). Effect of Storage, Cooking and Treatments on the Composition of Turkey Breast Phospholipids! The proportionate levels of phosphatidyl choline (PC), phosphatidyl ethanolamine (PE), sphingomyelin (SP) and lysophosphatidyl choline (LPC) in fresh turkey breast meat phospholipids were 54.17, 23.89, 14.01 and 7.10%, respecti- vely. These findings are in good agreement with the results reported by Fishwick (1968) and Wangen et a1. (1971) in turkey breast phospholipids. Fishwick (1968) found 60.9% PC + phosphatidyl inositol (PI), 29.2% PE, 8.2% SP and 1.7% phosphatidyl serine (PS) while Wangen et a1. (1971) reported 51.90% PC, 22.84% PE, 10.42% SP, 9.81% PI + PS and 5% LPC. The levels of PC and PE were found to decrease during frozen storage, while the proportion of SP and to a lesser extent LPC increased (Figure 19). The decline in PC and PE was significant (P<0.01) after one month of frozen storage of the breast meat at -25 C, and continued until the twelfth month of storage. The propor- tion of SP started to increase significantly even before the first month of storage. The increase in LPC was slight but was also significant (P < 0.01) after the first 89 60 ' l’ ' 1 ' l ' r ‘1 I r' I ' ‘r . 1 . K 50- 4~—c PC 7 i? \_I 1 4 {3401 1 U) g (J 030- -+ E.- E; . }_"~‘-~_¥ J 3: PE (L. 20d . U) 0 1 IE . D. 10‘ 4190-1 r/i . J O ”T 1* r r”*’ r ' 1* r r .11 r ' ‘_r -1 1 3 5 7 9 11 13 STORAGE TIME (MONTHS) Figure 19. Effect of storage on the distribution of phospho- lipid classes in turkey breast phospholipids. phosphatidyl choline (PC); Phosphatidyl ethanolamine (PE); sphingomyelin (SP); lysophosphatidyl choline (LPC) 1 Bars represent standard errors. 90 month of frozen storage. These changes were explained by Corliss and Dugan (1970) and Tsai and Smith (1971) who reported that PE is the most reactive component in the phospholipids and thus changes in PE may indicate the extent of lipid oxidation. They also reported that PC does not exert a prooxidant effect. However Acosta et al. (1966) reported that PC was more active in the early stages of autoxidation. It is expected that PE is more susceptible to autoxidative degradation than PC (Tsai and Smith, 1971). This is due to the fact that PE contains more highly unsa- turated fatty acids than PC (Hornstein et al., 1961; Keller and Kinsella, 1973; Body and Shorland, 1974). On the other hand, the increasing levels of SP and LPC during frozen storage suggest that lipolysis had occurred. Similar fin- dings had been reported by Awad et al. (1968) and Braddock and Dugan (1972) in beef and salmon. Data in Table 16 show that changes in the proportion of oleate, linoleate,,linolenate and arachidonate were not sufficiently consistent to reflect the autoxidative degra- dation process throughout frozen storage of turkey breast meat. This has been supported by Lea (1953), Hornstein et al. (1961) and Keller and Kinsella (1973). They indicated that the unsaturated fatty acids disappear rapidly at adva- nced stages of autoxidation. Polyunsaturated fatty acids 91 Table 16. Effect of frozen storage on the proportion of unsaturated fatty acids in phospholipid 1'2 Fatty acids Storage 18:1 18:2 18:3 26?} time 0-Month l3.38:l.26a 18.62:0.6Sa 0.18:0.15a 12.0810.64ab l-Month 14.15:O.93a 18.64:1.56a 0.18:0.253 11.82_+_0.57ab 3-Month 13.19_+_0.64ab l7.88:0.35a 0.22:0.09a 12.27_+_0.75ab 6-Month 134110.73ab 18.1810.59a 0.24:0.70a 12.741033a lZ-Month 12.17:0.46b 16.1810.88b 0.19_+_0.60a 11.40:}.13b 1Each value represents a mean of 20 replicates 2Means in the same column bearing the same letter are not significantly different (P<0.0l) (PUFA) appear to be principally responsible for the deve- lopment of rancidity, therefore both PC and PE could play an important role in the development of rancid flavors in stored meat or meat products because of their high content of PUFAS. The losses in PE after cooking were greater than in PC (Figure 20), although the decrease in both was significant (P<0.01). Therefore, PE was found to be less stable to cooking than PC. It was indicated by Love and Pearson (1971) and Keller and Kinsella (1973) that the loss of 92 PHOSPHOUPID CLASSES (Z) Figure 20. Effect of cooking on the proportion of phospho- lipid classes in turkey breast phospholipid. 1Lipid classes bearing the same letter are not significantly different (P<0.05) phosphatidyl choline (PC); phosphatidyl ethanolamine (PE); sphingomyelin (SP); lysophosphatidyl choline (LPC) 28ers represent standard errors. 93 arachidonic acid was consistent with its greater propensity to undergo autoxidation, especially'when associated with PE. However, because of the lipolysis, the SP and LPC levels have increased after cooking. the proportion of PE and PC decreased significantly (P<0.01) during frozen storage of turkey breast meat treated with copper, 50 micrograms/g. meat and copper and iron, 1:1 50 micrograms/g meat, (Figure 21). The changes in PC and PE are large over the storage time and appear to be directly related to the development of oxidative deteriora- tion when monitored by the increase of malonaldehyde in samples containing added iron and copper (Figure 13, Page 71). The metal cations act as prooxidants mainly to the polyenoic acids of PE and PC which are extremely reactive and through oxidative degradation produce a number of car- bonyl compounds which greatly influence oxidized flavor (Lea, 1957; Younathan and Watts, 1960). Changes in Fatty Acid Composition in Phospholipids of Turkey Breast During Frozen Storage and Cooking The fatty acid composition of phospholipid fraction of raw turkey breast was calculated as a percentage of total fatty acids in the phospholipid fraction. Results are shown in Table»l7. The percentage of saturated, mono-, di- and polyenoic fatty acids in the fresh breast at 0-time were q 94 w .pwmfiaonmmonm unmoun moxusu cw mommmao camaaonmwonm mo cofiuuomoum on» so mucoEummuu can omououm mo uummmm .Hm ousmwm $5225 us: 8565 6:503 ma: gas e...m...n.m.e e..«...e.....a.~.m.e9 '7'1 8 0.30 (x) ammomms wumdsona (x) annouo wumason-Id 95 31.35, 12.74, 20.87 and 23.91, respectively (Table 17). These values are in good agreement with the corresponding values 31.90, 12.90, 23.30, and 21.00 reported by Fishwick (1968). Some changes occurred in the unsaturated fatty acids during frozen storage. The loss in unsaturation could be accounted for by the changes in 18:2, 20:2, 20:3, 20:4, 22:3 and 22:6 fatty acids. The initial levels of dienoic, polyenoic and total unsaturated fatty acids were 20.87, 23.91 and 57.52%, respectively. The proportion of these acids showed a relative decrease at the 12th. month of frozen storage. The reported values for these acids after 12 months storage were 18.90, 21.34 and 52.63%, respectiv- ely. Although these observed losses in unsaturated fatty acids would indicate the occurrence of lipid oxidation during frozen storage of raw turkey breast, there were no significant changes for these acids when stored for less than 12 months. These findings lead to the conclusion that fatty acid profile changes are not ideal for monitoring lipid oxidation in stored poultry meat. Nineteen fatty acids were identified and quantified in the phospholipid fraction from cooked turkey breast meat. The percentage of saturated, mono-, di-, and polyenoic fatty acids in the freshly cooked turkey breast were 37.00, 15.31, 18.83 and 23.66, respectively (Table 18). 96 Table 17. Changes in fatty acid composition as percent of the phospholipids of raw turkey breasf muscle for the indicated frozen storage times Months of storage at -25 C Fatty acid2 0 1 3 6 12 14:0 0.313 0.11 0.16 0.24 0.03 14:1 0.0 0.0 0.0 0.02 0.0 15:0 0.0 0.07 0.0 0.02 0.0 16:DMA 8.45 7.41 10.70 8.38 13.03 16:0 14.33 18.45 13.14 14.19 14.19 16:1 0.53 0.58 0.56 0.83 0.27 18:DMA 2.67 0.30 3.53 2.71 3.37 16:2 1.63 0.0 2.19 1.61 2.05 18:0 16.71 17.86 15.38 16.03 16.70 18:1 12.21 13.82 13.04 14.23 12.12 18:2 18.99 18.62 18.00 18.68 16.74 18:3 0.18 0.23 0.29 0.33 0.21 20:2 0.26 0.33 0.20 0.07 0.12 20:3 2.01 2.43 1.41 1.71 1.59 20:4 12.49 11.27 12.82 12.05 12.07 20:5 0.84 0.69 0.57 1.34 0.57 22:3 0.22 0.25 0.07 0.06 0.08 22:4 3.05 3.09 3.07 2.85 2.98 22:5W6 1.36 1.19 1.27 1.20 0.94 22:5w3 1.53 1.36 1.51 1.42 1.29 22:6 2.24 1.94 2.08 2.02 1.61 Sat. 31.35 36.49 28.68 30.48 30.97 Mono 12.74 14.40 13.60 15.09 12.39 Dienoic 20.87 18.95 20.40 20.36 18.90 Polyenoic 23.91 22.45 23.09 23.01 21.34 Unsat. 57.52 55.55 57.08 58.46 52.63 1Results represent means of 10 determinations 2 Number of carbon : number of double bonds 3Percent of total fatty acids 16:DMA s Dimethyl acetal of hexadecanal 18:DMA = Dimethyl acetal of octadecanal W = Omega 97 Table 18. Changes in fatty acid composition as percent of the phospholipids of cooked turkey breast muscle for the indicated frozen storage times Months of storage at -25 C Fatty acid2 0 1 3 6 12 14:0 0.233 0.23 0.20 0.16 0.01 14:1 0.09 0.00 0.00 0.00 0.00 15:0 0.00 0.00 0.00 0.00 0.00 16:DMA 4.08 9.32 8.81 9.32 14.51 16:0 17.68 15.11 14.90 13.88 15.05 16:1 0.69 0.41 0.35 0.27 0.20 18:DMA 1.14 2.73 2.68 2.75 3.60 16:2 0.53 1.55 1.37 1.61 2.17 18:0 19.04 19.52 19.66 18.53 17.52 18:1 14.54 13.37 13.33 12.21 11.91 18:2 18.25 17.17 17.76 17.69 15.62 18:3 0.19 0.11 0.14 0.15 0.17 20:2 0.04 0.04 0.11 0.04 0.09 20:3 2.06 1.76 1.60 1.73 1.24 20:4 11.67 11.46 11.72 13.43 10.74 20:5 1.13 0.59 0.52 0.54 0.58 22:3 0.59 0.36 0.30 0.11 0.13 22:4 2.94 2.59 2.74 3.14 2.81 22:5w6 1.42 1.00 0.96 1.18 0.92 22:5w3 1.59 1.18 1.16 1.40 1.23 22:6 2.07 1.52 1.70 1.86 1.52 Sat. 37.00 34.86 34.75 32.57 32.58 Mono 15.31 13.78 13.68 12.48 12.11 Dienoic 18.83 18.76 19.24 19.35 17.88 Polyenoic 23.66 20.55 20.84 23.54 19.34 Unsat. 57.79 53.10 53.75 55.37 49.33 1Results represent means of 10 determinations 2Number of carbon : number of double bonds 3Percent of total fatty acids 16:DMA = Dimethyl acetal of hexadecanal 18:DMA = Dimethyl acetal of octadecanal w = Omega 98 Important changes occurred in these fatty acids during cooking. The saturated and the monoenoic fatty acids inc- reased by 15 and 16.7%, respectively, due to cooking as compared to that in the fresh breast meat values, (Tables 17 and 18) indicating a shift in proportion due to loss, disappearance or failure to detect unsaturated fatty acids. The effect of cooking on the level of fatty acids in phospholipids, from turkey breast regardless of storage conditions is illustrated in Figures 22 and 23. The pro- portion of the total unsaturated, di- and tetraenoic fatty acids decreased significantly (P<0.05) as a result of cooking. No significant (P<0.05) changes were found in mono- and trienoic fatty acids. The level of total satu- rated fatty acids increased significantly (P<0.01). The main change in dienoic acids was due to the autoxidation of linoleate (18:2). Linolenate (18:3) is a polyunsaturated acid which was expected to be more labile to autoxidation, but no significant changes were detected. This might be a result of the extremely low quantities of this acid. However arachidonate (20:4) content was slightly decreased after cooking. The high level of arachidonic acid in phospholipids distinguishes this fraction from neutral lipid. 99 80 . Rflfi r r r r r '1 COOKED ‘ 70- 1 4 60a _ FATTY ACID (7:) 8 .\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\X~:‘ Unaat Mono Dienoic Trianoic Tatroanoic Sat Figure 22. Effect of cooking on the proportion of fatty acids in the phospholipid fraction. 1Fatty acid groups bearing the same letter are not significantly different (P<0.05) Bars represent standard errors. 100 2° _1 COOKED 22 . 1 EZZIRNW A5 92 kt: 18:2 18:3 20:4 18:1 the proportion of unsatura- 23. Effect of cooking on Figure ds in the phospholipid fraction. ted fatty aci 1Fatty aci bearing the same letter ds different (P<0.05) ent standard errors. repres 100 1:37:31» 1- q I! ‘14 - d'ld 7. 2% B -EEfl 9;. 22 20 _1 COOKED A5 92 E6 4 18:2 18:3 18:1 of unsatura- Figure 23. Effect of cooking on the proportion ted fatty acids in the phospholipid fraction. significantly not are 1Fatty acids bearing the same letter different (P<0.05) t standard errors ex; coc sic con 1i] The enc com the dat stc the 182 the M1 der Prc 101 The decrease in the di- and polyenoic fatty acids is explained by the rapid autoxidation of these acids after cooking. Sato and Hegarty (1971) came to a similar conclu- sion and suggested that cooking exposes the labile lipid components to catalytic agents. It might also be due to the liberation of heme iron by cooking (Igene et al., 1981). The saturated as well as the monoenoic fatty acids are fairly stable, thus they usually either stay at the same level or increase in proportion to the decrease in poly- enoic acids, when measured as percentage of the total composition. However, the results of this work revealed that with prolonged frozen storage the monoenoic acid oxi- dation was initiated. Changes in the proportions of fatty acids during frozen storage~of the cooked breast meat are similar to those of the raw breast meat. The high susceptibility of cooked meat to autoxidation might be due to the denaturation of the protein which facilitates exposure of iron to unsatu- rated fatty acids. Eriksson (1975) showed that protein denaturation increased the ability of the heme-containing proteins, peroxidase and catalase to promote lipid oxida- tion. 102 Changes in Fatty Acid Composition Percent in Neutral Lipids of Turkey Breast During Frozen Storage and Cooking The fatty acid composition from raw and cooked turkey breast muscle was calculated as percentage of the total fatty acids in the neutral lipid fraction. Results are presented in Tables 19 and 20. Twenty different fatty acids were identified and quantified from the neutral lipid frac- tion. The most prevalent fatty acids had 16 carbon atoms (palmitic and palmitoleic) and 18 carbon atoms (stearic, oleic and linoleic acids). These C:16 and C:18 fatty acids composed 92.5 percent of the total fatty acids in the neutral lipids. Palmitic acid accounted for a major por- tion of the saturated fatty acids (about 2/3 of that in the fresh raw and the fresh cooked breast muscled. Oleic and linoleic acids accounted for a large percentage of the unsaturated fatty acids (more than 4/5 of that in the fresh raw and in the fresh cooked breast muscle). This distribu- tion of fatty acids in neutral lipids is typical and has been previously reported by Issacks et a1. (1964) in triglyceride fraction of adipose tissue of the laying hens and Katz et al. (1966) in neutral lipids from chicken tissues. Since oxidation depends on the presence of unsaturated fatty acids, an analysis of individual fatty acids was made 103 Table 19. Changes in fatty acid composition as percent of the neutral lipids of raw turkey breast muscle of the indicated frozen storage times Months of storage at -25 C Fatty acid2 0 1 3 6 12 12:0 0.003 0.00 0.05 0.06 0.07 14:0 1.37 1.41 1.36 1.34 1.56 14:1 0.23 0.05 0.19 0.29 0.22 15:0 0.14 0.04 0.09 0.18 0.16 16:DMA 0.35 0.19 0.44 0.18 0.15 16:0 21.64 21.45 22.42 21.94 22.57 16:1 3.84 3.88 3.89 3.82 3.97 18:DMA 0.51 0.24 0.59 0.55 0.57 16:2 0.41 0.19 0.38 0.46 0.43 18:0 8.70 8.76 9.05 8.86 9.26 18:1 32.70 32.74 34.33 34.16 34.59 18:2 23.19 22.27 22.41 22.62 23.33 18:3 1.48 1.57 1.60 1.59 1.24 20:2 0.38 0.35 0.29 0.47 0.19 20:3 0.38 0.35 0.22 0.42 0.14 20:4 2.52 1.03 0.93 1.07 0.81 20:5 0.48 0.10 0.27 0.24 0.31 22:3 0.11 0.04 0.10 0.08 0.0 22:4 0.64 0.21 0.48 0.56 0.19 22:5W6 0.33 0.15 0.21 0.28 0.02 22:5w3 0.34 4.74 0.46 0.42 0.12 22:6 0.24 0.20 0.26 0.44 0.09 Sat. 31.88 31.70 32.96 32.38 33.63 Mono 36.93 36.67 38.40 38.27 38.79 Dienoic 23.97 22.81 23.08 23.54 23.95 Polyenoic 6.52 8.39 4.51 5.08 2.92 Unsat. 67.42 67.87 66.00 66.89 65.65 1Results represent means of 10 determinations 2 Number of carbon : number of double bonds 3Percent of total fatty acids 16:DMA = Dimethyl acetal of hexadecanal 18:DMA c Dimethyl acetal of octadecanal W = Omega 104 Table 20. Changes in fatty acid composition as percent of the neutral lipids of cooked turkey breast muscle for the indicated frozen storage times Months of storage at -25 C Fatty acid2 0 1 3 6 12 12:0 0.043 0.00 0.00 0.00 0.00 14:0 1.52 1.41 1.47 11.44 1.54 14:1 0.15 0.08 0.02 0.16 0.21 15:0 0.09 0.06 0.00 0.04 0.18 16:DMA 0.38 0.03 0.00 0.04 0.08 16:0 22.48 22.30 22.87 22.47 23.28 16:1 3.89 4.16 4.18 4.07 4.19 18:DMA 0.57 0.20 0.16 0.26 0.40 16:2 0.42 0.18 0.10 0.22 0.30 18:0 9.10 8.77 8.91 8.84 9.21 18:1 33.70 34.81 35.02 34.55 35.14 18:2 21.89 22.22 22.03 22.92 22.76 18:3 1.46 1.49 1.36 1.47 1.13 20:2 0.35 0.45 0.24 0.27 0.20 20:3 0.58 0.41 0.22 0.21 0.12 20:4 1.43 1.43 1.66 1.49 0.86 20:5 0.57 0.32 0.28 0.28 0.05 22:3 0.16 0.28 0.14 0.14 0.01 22:4 0.49 0.37 0.49 0.59 0.18 22:5W6 0.25 0.21 0.46 0.19 0.01 22:5w3 0.27 0.55 0.22 0.17 0.07 22:6' 0.22 0.28 0.17 0.17 0.03 Sat. 33.23 32.54 33.25 32.79 34.21 Mono 37.74 39.05 39.23 38.79 39.54 Dienoic 22.65 22.85 22.37 23.41 23.26 Polyenoic 5.43 5.33 4.99 4.71 2.41 Unsat. 65.82 67.23 66.59 66.91 65.21 1Results represent means of 10 determinations 2Number of carbon : number of double bonds 3Percent of total fatty acids 16:DMA = Dimethyl acetal of hexadecanal 18:DMA - Dimethyl acetal of octadecanal W = Omega 105 for neutral lipids in raw and cooked, and fresh and frozen turkey breast muscle. Although neutral lipids have a much less important role in autoxidation of poultry meat because of the lower content of polyunsaturated fatty acids (6.52 and 5.43% in the raw and cooked meat compared to 23.91 and 23.66% in phospholipids of raw and cooked meat) the propor- tion of polyenoic fatty acids in them consistently dec- reased during frozen storage with a loss of up to half their proportional amount by 12 months of storage. The greatest difference between the percentage of polyenoic acids in neutral and phospholipids is due to the higher content of arachidonic acid (20:4) in the phospholipid fraction. These results agree with the findings reported by Miller et al. (1962) and Katz et a1. (1966). The effect of cooking on the proportion of fatty acids in neutral lipids is presented in Figures 24 and 25. These graphs show that fatty acids in neutral lipids of turkey breast meat are not greatly altered during cooking. The levels of total unsaturated and saturated fatty acids were not significantly (P<0.05) affected by cooking. However as a result of cooking, a slightly greater proportion of monoenoic acids was observed (mostly oleate). The reverse was the case for dienoic acids which amount to about 99% linoleate (18:2). 106 80 E: RAW a COOKED 70 A ‘Alz FATTY ACID (7:) ‘6 3 8 8 h) C) 10 O Unaot Mono Dienoic Trianoic Tetroanoic Sat Figure 24. Effect of cooking on the proportion of fatty acids in neutral lipid fraction. 1Fatty acid groups bearing the same letter are not significantly different (P<0.05) 2 Bars represent standard errors. 107 FATIY ACID (7:) 18:1 18:2 18:3 20:4 Figure 25. Effect of cooking on the proportion of unsatura— ted fatty acids in neutral lipid fraction. 1Fatty acids bearing the same letter are not significantly different (P<0.05) Bars represent standard errors. 108 The results indicate that only minor changes occurred in fatty acid profiles of neutral lipids during frozen storage and cooking of turkey breast meat (Table 19 and 20, Figures 24 and 25). These results are in agreement with the Chang and Watts (1952) and Igene et al. (1981) which showed only slight changes in the fatty acid composition of neutral lipids during storage and cooking. The effects of frozen storage on the proportion of saturated fatty acids regardless of cooking are illustrated in Figure 26 for both neutral lipids and phospholipids. Significant changes in the unsaturated fatty acids in phos- pholipids were found during frozen storage of the meat (unlike the neutral lipids). The changes are largely pro— duced by time long term storage (12 mo.) effect. Other changes in unsaturated fatty acids (as shown earlier in Tables 17 and 18) which occurred during the storage inter- vals of 1, 3, and 6 months are not valuable as indicators of oxidative rancidity. Fatty Acid Composition of Fresh Raw Turkey Thigh Lipids Fatty acid composition of phospholipids and neutral lipid fractions of fresh raw turkey thigh muscle are shown in Table 21. In general the fatty acid profile of thigh meat is similar to that of the breast meat except for some minor differences. The mono- and dienoic acids are higher 109 70 r f ' I f r ‘ I ' I ' I r r ' I f I q &’_’.\ '1 654 ML. UNSAT _ 50- u a J S? 55- . v o ‘ _. 1W. UNSKT 2 50-1 - E ‘ ‘ E ‘54 .4 40- 1 i '1 351 ML. SAT 7 ‘ e——:: PL.SKT ‘ 3° ‘1 I ' I " I ' r I’ I ' 7* I 7’ 7’ -113 5 7 91'113151'7 STORAGE TIME (MONTHS) Figure 26. Effect of storage on the composition of satura— ted and unsaturated fatty acids in neutral and phospholipids. neutral lipid (NL); Phospholipid (PL); saturated (SAT); unsaturated (UNSAT) 1 Bars represent standard errors. Sat Hon Bier. Poly Unsa: ”(The 2V3] 48. Percen 451mm“ . Omega 110 Table 21. Fatty acid profile of fresh turkey thigh lipids Phospholipids Neutral lipids Fatty acid1 Mean STD4 Mean STD 12:0 0.02 0.005 0.10 0.007 14:0 1.95 0.637 1.22 0.047 14:1 1.22 0.686 0.25 0.031 15:0 0.98 0.222 0.25 0.035 16:DMA 4.23 1.321 0.00 0.000 16:0 12.22 0.497 19.82 0.788 16:1 0.93 0.156 3.92 0.902 18:DMA 1.18 0.511 0.42 0.105 16:2 0.47 0.214 0.47 0.142 18:0 16.90 0.308 7.36 0.336 18:1 15.09 0.914 34.15 1.229 18:2 23.92 0.200 26.10 0.970 18:3 0.57 0.083 2.25 0.190 20:2 0.22 0.041 0.47 0.084 20:3 0.76 0.192 0.47 0.381 20:4 13.77 0.715 0.81 0.335 20:5 0.46 0.248 0.76 0.583 22:4 2.14 0.203 0.51 0.365 22:5W6 0.98 0.154 0.36 0.227 22:5w3 1.08 0.090 0.24 0.069 22:6 0.89 0.078 2.65 5.175 Sat. 32.08 0.534 28.74 0.801 Mono 17.34 1.043 38.32 1.875 Dienoic 24.61 0.364 27.04 0.996 Polyenoic 20.65 1.668 5.48 2.331 Unsat. 62.50 1.368 70.84 0.710 1Number of carbon : number of double bonds 2Values represent means of 5 determinations 3Percent of total fatty acids 4Standard deviation 16:DMA 8 Dimethyl acetal of hexadecanal 18:DMA s Dimethyl acetal of octadecanal w = Omega ac not res firm and MIT Him 1913. PhOSphI phOSPhc breast 1 Dim: 111 in the thigh phospholipids 17.34 and 24.61% vs. 12.74 and 20.87% in the breast phospholipids. This same trend exists in neutral lipids. The mono- and dienoic acids are 38.32 and 2L04%, respectively in thigh meat while the correspon- ding values for breast meat are 36.93 and 23.97%. Identification of Dimethyl Acetal of Hexa- and Octadecanal Chromatographic results of fatty acids in phospholipids (Tables 17 and 18, Pages 96 and 97) indicated the presence of two unknown compounds. One of them was eluted preceding the methyl ester of hexadecanoic acid and the second was eluted preceding the octadecanoic acid. Methylated fatty acid standards were found to have retention times which did not correspond to those of the unknowns. From the above results and mass spectrophotometric analyses it was con- firmed that the compounds were not fatty acid methyl esters and it was suspected that they were plasmalogen aldehyde derivatives or oxidation products. However, the ident- ification and quantification of dimethyl acetals (DMA) of hexa- and octadecanal has not been reported in turkey meat phospholipids. Their identification and proportion in the phospholipid fraction is reported for raw and cooked turkey breast meat throughout 12 months frozen storage. Dimethyl acetals are formed from methylation of the aldehydes during esterification of the phospholipids. The hem- and om (111 913811310 )1 01 381113) \hhl‘, )eng 'hese two cancic aci The . 111/12% I'm 0 16.61‘ cm and {I 112 hexa- and octadecanals are long-chain aldehydes released from plasmalogens. Plasmalogens have been found in a varie- ty of animal tissues (Webster, 1960, Rapport and Norton, 1962; Peng and Dugan, 1965 and Neudoerffer and Lea, 1967) These two aldehydes correspond to hexadecanoic and octade— canoic acids and are found as enol ethers. The gas chromatographic analysis using 10% SP-2330 on 100/120 Supelcoport showed a Significantly lower concentra- tion of the 16:DMA in the alkali transesterified sample (6.6%) than in the acid esterified sample (12.2%), but the corresponding values of 18:DMA for the two methods were 2.7 and 2.0%, respectively. However when the saponified por- tion of phospholipids was extracted and analyzed after being esterified, the result obtained was similar to that reported when the boron trifluoride-methanol method of esterification was used directly. Gas chromatography-mass spectrophotometric analysis indicated that there were two compounds that are present in substantial amounts in turkey phospholipids and have high molecular weights of 286 and 314 (Figures 3 and 4 Appendix C). These results indicate that the compounds could not be oxidation products. The unknowns produced large mass peaks at 75 m/e and mass units produced were two mass units higher than fragmentation peaks produced by fatty acid methyl esters. hy Ia ac. th dir rte to The wet cor 4.81 391. (197 bree whij 1.52 leve but tUrkE incre 113 Friedel and Sharkey (1956) indicated that oxygenated hydrocarbons must contain two oxygen atoms and no unsatu— rated bonds in order to produce such peaks. The dimethyl acetals produce such peaks. Also the molecular weight of these compounds are the same as those of C:16 and C:18 dimethyl acetals. Gardner et al. (1972) using infrared spectroscopy repo- rted the possible presence of hexadecanal and octadecanal in chicken meat. In this work, DMA of hexa- and octadecanal were found to be present in both raw and cooked turkey breast lipids. The quantity of hexa- and octadimethyl acetals in the raw were 8.45 and 2.67%, respectively (Table 17, Page 96). The corresponding values in the freshly cooked breast meat were 4.08 and 1.14% (Table 18, Page 97). These values are in agreement with the results reported by Gardner et al. (1972). They indicated that phospholipids from raw broiler breast contained 8.54% hexadecanal and 2.68% octadecanal, while the cooked muscle contained 8.41% hexadecanal and 1.52% octadecanal. Moerck and Ball (1973) reported a lower level of hexadecanal (2.36%) in bone marrow phospholipids but the octadecanal level was 1.6%. During 12 months of frozen storage of raw and cooked turkey breast meat the percentage of hexa- and octadecanal increased the hexadecanal by about 21 and 56% in the raw 114 and cooked samples, respectively. And the octadecanal increased by about 12 and 52% in the raw and cooked samples, respectively. As the aldehyde of these compounds are associated with the membranous structures of the tis- sues, the prolonged frozen storage and / or cooking might help release them completely because of the deteriorative changes which occur in the meat tissues. The fresh raw turkey thigh muscle contained lower pro- portion of both hexa- and octadecanal (4.23 and 1.18%, respectively) compared to fresh raw turkey breast. Gardner et al. (1972) reported in raw'chicken thightmeat were 6.00 and 2.38%. The neutral lipids contained traces of these compounds. Aerobic Plate Count (APC) In meat systems microbial growth limits the shelf life and results in important flavor changes in the meat product (Dawson et al., 1975). Although certain lipolytic or fat— splitting microorganisms, such as some species of Pseudomonas, can hydrolyze the fat causing hydrolytic ran- cidity, the main cause of changes in meat is oxidative rancidity produced by chemical combination of oxygen in the air with the unsaturated fatty acid component of the fat. Figure 27 demonstrates that the logarithmic growth phase of aerobic organisms for both refrigerated raw breast 115 .umoe aoxusu amuouomfiummu 3mm you unsoo mumfim ofinouoo coommuoum pcm mucoEumOuu mo uoowum.- muswwm ngavuaflgohm 2 3 «e 2 aI~.~ I- tP I I b'! I D «. hm: 10.1... «I: mas/moon am: am OI OO'I figs 2.: ”.6365 r 8.0.....—.v.—.«.— o.—. a . m I a .m o ' ._. cupric ions > rock salt > fine flake 125 salt > control. However these treatments were found to have no detectable effect on lipid oxidation as monitored by changes in fatty acid profile, per se. while the propo- rtion of phosphatidyl ethanolamine (PE) and phosphatidyl choline (PC) was significantly'(P<0.0l) decreased by the effect of iron and copper. The percent of nonheme iron of the total iron in turkey meat was found to be much higher than that reported in other species. This might be due to the lower myoglobin content of poultry meat and it could be one of the major factors which contribute to the lower stability of turkey meat to lipid oxidation, especially thigh meat which has about 3 times more total iron than the breast meat. Also copper content in thigh meat was found to be 3 times that in the breast. Rock salt had a signif- icantlylarger prooxidant effect (P<0.0l) than pure salts (P<0.05) which might be due to the metal impurities. Tenox 6 was a more effective antioxidant in controlling the prooxidant effect of both iron and copper (P<0.01). The other antioxidant which contained vitamin E, ascorbyl palmitate and citric acid as the active ingredients prod- uced no significant control on the prooxidant activity of iron and copper. The proportion of phospholipids in total fat was higher in breast than in the thigh meat, but the total amount of phospholipids was higher in the thigh meat. There was a 126 decrease in the proportion of phospholipids with frozen storage and cooking. This decrease corresponds to the variation in the level of malonaldehyde as measured by TBA test. Also, the proportion of the phospholipid classes PE and PC was found to decrease after cooking and during frozen storage (P<0.01). However the PE was found to to be more susceptible to autoxidative degradation than PC, this is because the former is much more unsaturated. The loss in the unsaturation could be accounted for by the changes in 18:2, 20:2, 20:3, 20:4, 22:3 and 22:6 fatty acids. The proportion of these acids showed a relative decrease after cooking and after 12 months of frozen storage. But there were no significant changes for these acids throughout the shorter storage periods. These results lead to the conclu- sion that fatty acid profile changes are not ideal indica- tors for monitoring lipid oxidation for the stored poultry products. However, only minor changes have occurred in fatty acid profile of neutral lipids during frozen storage and cooking bf turkey breast meat. Dimethyl acetal of hexa- and octadecanal were identi- fied and quantified in phospholipid fraction in raw and cooked turkey breast throughout 12 months frozen storage as well as in the fresh raw thigh meat. The neutral lipids contained traces of these compounds. 127 By aerobic plate count it was found that spoilage was detected in raw turkey breast meat refrigerated at 4 C after 21 days. The number of viable microorganisms was found to decrease with frozen storage (P<0.01). The results of this study clearly Show the sensitivity of the very small quantity of phospholipids in turkey muscle to oxidation and concomitant flavor change. The severe effects of free iron and copper ions along with the presence of salt was also demonstrated. APPENDICES APPENDI CBS Appendix A Table 1. Chemical analysis of calcium magnesium free salt (CMF) and fine flake salt1 Chemical CMF Salt % Fine Flake Salt % NaCl 2 99.910 99.960 NBZSO4 0.091 none CaSO4 0.000 0.038 (261.2 0.000 0.001 kgclz 0.000 0.003 ca PPM 3.000 115.000 Mg PPM 2.000 8.000 Cu PPM 0.090 0.080 Fe PPM 0.250 0.300 ,— 1Analyzed by Diamond Crystal Co. St. Clair, Michigan. 20“ moisture free basis. NO’C determined, but typical values are as follows: Ni, 0.08 PPF1; Zn, 0.05 PPM: Pb none detected or less known 0.05 PPM. 128 129 Table 2. Particle Size analysis of calcium magnesium free salt (CMF) and fine flake saltl On USS CMF Salt Fine Flake Salt 20 tr. 0.0 30 10.0 tr. 40 45.0 0.5 50 38.0 16.5 70 5.5 39.0 80 0.0 15.0 100 0.0 14.5 Pan 1.5 14.5 \ énalyzed by Diamond Crystal Salt Co. St. Clair, Michigan. Tab 1 e 3. Chemical analysis of rock salts for iron and copper content1 r Sal t: type Iron (PPM) Copper (PPM) Northern rock salt (NRS) 39-62 1-19 SOVIthern rock salt (SRS) 35.23 1.27 l . . Ahalyzed in Department of Animal Sc1ence, Michigan State U“ iversity. Appendix B Table 1. Triangle test questionnaire for cooked turkey meat. QUESTIONNAIRE FOR TRIANGLE TEST (Taste for Flavor only) NAME : DATE ' PRODUCT: Turkey Meat Two of these three samples are identical, the third is different. 1. Please write the code number in the space indicated and check the odd sample. Code No. Check Odd Sample Indicate the degree of difference between the duplicate samples and the odd sample.. Slight Moderate Much Extreme 3. Acceptability: Odd sample more acceptable Duplicates more acceptable ‘4. Comments: 130 131 Table 2. Cooked chicken rating score form. CHICKEN WARMED-OVER FLAVOR Name Date Directions: Taste the reference sample first to familiarize your self with fresh chicken flavor. Then taste the coded samples in the order listed below. Mark the degree of warmed-over flavor by plac- ing a vertical line on the horizontal line corresponding to the three digit code. Evaluate sample for warmed-over flavor only. Expectorate after tasting in the cup provided and rinse mouth with water between samples. c098 NO WARMED-OVER VFRx STRONG WARMED- FLAVOR OVER FLAVOR —--— — - I ------------------- I ------------------- I --— —— - 1 ------------------- I ------------------- I ——- -- - I ------------------- I ------------------- I —-—- ~ - 1 ------------------- I ------------------- I -.._ ~- 1 ------------------- 1 ------------------- I Appendix C . 111:0 18:1 2034 F 1682 Figure 1. Gas chromatographic spectrum of turkey phospho- lipid fatty acids. 13% SP-2330 on 100/120 Supelcoport; 30 m x 0.24 mm i.d. capillary column; 0.2 film thickness; helium velocity = 35 >-»pyL.pP-.p>pr «r Can“: b. . i 0.8 aAynlaJ r. .. .8. Mgsuaiu! 136 CH2--O--CH=CH-Rl (ma unsaturated ether) I CH -O-CO-R2 I CHZ-O-ggO-X Plasmalogen Oxidation U CH3(CH2)n-EH (long chain aldehyde) Methylation {y OCH3 CH3(CH2)n$H OCH3 Dimethyl acetal (DMA) Rl= Fatty aldehyde residue R2= Fatty acid residue X = Choline, ethanolamine, or serine residue (ethanolamine is the most common) n = 14 in DMA of hexadecanal and 16 in DMA of octadecanal. Figure 5. Formation of dimethyl acetals from plasmalogens. B I BL I OGRAPHY BIBLIOGRAPHY A.O.A.C. 1975. "Official Methods of Analysis", 12th. Ed. Association of Official Agricultural Chemists. Washin— gton, D. C. A.O.C.S. 1974. American Oil Chemists. Official Method Cd 12-25 in "A.O.C.S. Official and Tentative Methods". Champaign, IL. Acosta, S. 0., Marion, W. W. and Forsythe, R. H. 1966. Total lipids and phospholipids in turkey tissue. Pou- ltry Sci. 45:169. Allen, C. E. and Foegeding, E. A. 1981. Some lipid chara- cteristics and interactions in muscle foods. A rev- iew. Food Technol. 35:253. Awad, A., PowRie, W. D. and Fennema, O. 1968. Chemical deterioration of frozen bovine muscle at - 4 C. J. Food Sci. 33:227. Barron, E. S. G. and Lyman, C. M. 1938. Studies on biological oxidations. X. The oxidation of unsaturated fatty acids with blood hematins and hemochromogens as catalysts. J. Biol. Chem. 123:229. Bidlack, W. R. and Tappel, A. L. 1972. A proposal mecha- nism for the TPNH enzymatic lipid peroxidizing system of rat liver microsomes. Lipids 7:564. Body, D. R. and Shorland, F. B. 1974. The fatty acid composition of the main phospholipid fractions of the rumen and abomasum tissues of foetal and adult sheep. J. Sci. Fd. Agric. 25:197 Bowers, J. A. 1984. Personal communication. Kansas State University, Department of food and nutrition. Manhattan, Kansas. Braddock, R. J. and Dugan, L. R., Jr. 1972. Phospholipid changes in muscle from frozen stored Lake Michigan Coho salmon. J. Food Sci. 37:426. 137 138 Brown, W. D., Harris, L. S. and Olcott, H. S. 1963. Cata- lysis of unsaturated lipid oxidation by iron protopor— phyrin derivatives. Arch. Biochem. Bio-Phys. 101:14. Buttkus, H. A. 1967. The reaction of myosin with malonal- dehyde. J. Food sci. 32:432. Campbell, A. D. and Turkki, P. R. 1967. Lipids of raw and cooked beef and pork. J. Food Sci. 32:143. Castell, C. H., Maclean, J. and Moore, B. 1965. Rancidity in lean fish muscle. 1V. Effect of sodium chloride and other salts. J. Fish. Res. BD. CANADA, 22:929. Castell, C. H. and Spears, D. M. 1968. Heavy metal ions and the development of rancidity in blended fish mus- cle. J. Fish. Res. BD. CANADA, 25:639 Chang, I. and Watts, B. M. 1952. The fatty acid content of meat and poultry before and after cooking. J. Am. Oil Chem. Soc. 29:334. Chen, C. C., Pearson, A. M., Gray, J. 1., Fooladi, M. H. and Ku, P. K. 1984. Some factors influencing the nonheme iron content and its implications. J. Food Sci. 49:581. Choudhury, R. B. R. and Arnold, L. K. 1960. The determination of the neutral oil content of crude vege- table oils. J. Am. Oil Chem. Soc. 37:87. Corliss, G. A. and Dugan, L. R., Jr. 1970. Phospholipid oxidation in emulsions. Lipids 5:486. Cross, C. K. and Ziegler, P. 1965. A comparison of the volatile fractions from cured and uncured meat. Food Res. 30:610. Dahle, L. K., Hill, E. G. and Holman, R. T. 1962. The thiobarbituric acid reaction and autoxidation of poly- unsaturated fatty acid methyl esters. Arch. Biochem. Biophys. 98:253. Davidkova, E. and Khan, A. W. 1967. Changes in lipid composition of chicken muscle during frozen storage. J. Food Sci. 32:35. Dawson, L. E. and Gartner, R. 1983. Lipid oxidation in mechanically deboned poultry, a review. Food Technol. 37:112. 139 Dawson, L. E., Stevenson, K. E. and Gertonson, E. 1975. Flavor, bacterial and TBA changes in ground turkey patties treated with antioxidants. Poultry Sci. 54:1134. Deibel. R. H. and Lindquist, J. A. 1981. General food micro- biology, laboratory manual. Burgess Publishing Comp- any, Minneapolis, Minnesota. Dugan, L. R., Jr. 1960. Antioxidants that protect your fats and foods made with your fats. Circ. 63, Amer. Meat Inst. Foundation, Chicago, Illinois. Dugan, L. R., Jr. 1976. Lipids. In "Principles of food science, Part 1. Food chemistry". Edited by Fennema, O. R., P. 182. Marcel Dekker, INC. New York and Basel. Dyer, W. J. 1953. Main problems of fish protein denatura- tion. Proceedings of symposium on cured and frozen fish technology. Swedish Institute for food preserva- tion research, Goteborg, Sweeden. November 1953. Ecosoft, 1984. Microstat. Version 4.0. Ecosoft, Inc. Indianapolis, Indiana. Eisensmith, S. P. 1985. Plotit, interactive graphics and statistics. Version 1.0. Michigan State University. El-Gharbawi, M. I. 1964. Studies on some chemical and physicochemical characteristics of freeze-dried raw beef during storage under modified atmospheres. Ph.D. dissertation, Michigan State Uni., E. Lansing. El-Gharbawi, M. I. and Dugan, L. R., Jr. 1965. Stability of nitrogenous compounds and lipids during storage of freeze-dried raw beef. J. Food Sci. 30:817. Eriksson, C. 1975. Aroma compounds derived from oxidized lipids. Some biochemical and analytical aspects. J. Agric. Food chem. 23:126. Fishwick, M. J. 1968. Changes in the lipids of turkey muscle during storage at chilling and freezing temperatures. J. Sci. Fd. Agric. 19:440. Folch, J., Lees, M. and Stanley, G. H. S. 1957. A simple method for isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226:497 140 Fooladi, M. H. 1977. The role of nitrite in preventing development of warmed-over flavor in cooked meat from different species of animals. M. S. Thesis, Michigan State University, E. Lansing. Frankel, E. N. 1985. Chemistry of free radical and sing— let oxidation of lipids. Prog. Lipid Res. 23:197. Frankel, E. N., Neff, W. E. and Selke, E. 1984. Analysis of autoxidized fats by gas chromatography-mass spectro- metry. IX. Homolytic vs. heterolytic cleavage of pri- mary and secondary oxidation products. Lipids 19:790. Friedel, R. A. and Sharkey, A. 6., Jr. 1956. Mass spectra of acetal type compounds. Anal. Chem. 28:940. Gaddis, A. M., Ellis, R. and Currie, G. T. 1961. Carbonyls in oxidizing fat. V. The composition of neutral volatile monocarbonyl compounds from autoxidized oleate, linoleate, linolenate esters, and fats. J. Am. Oil Chem. Soc. 38:371. Gardner, H. K., Huber, C. S., Bourland, C. T. and Smith, M. C., Jr. 1972. Identification and quantitation of hexadecanal and octadecanal in broiler muscle phospholipids. Poultry Sci. 51:1056. Govindarajan, S. 1973. Fresh meat color. CRC Crit. Rev. Food Sci. Nutri. 4:117. Govindarajan, S., Hultin, H. O. and Kotula, A. W. 1977. Myoglobin oxidation in ground beef. Mechanistic studies. J. Food Sci. 42:571. Gray, J. I. 1978. Measurement of lipid oxidation. review. J. Am. Oil Chem. Soc. 55:539. Green, B. E. and Price, L. G. 1975. Oxidation—induced color and flavor changes in meat. J. Agric. Food Chem. 23:164. Herz, K. O. and Chang, S. S. 1970. Meat flavor. In "Advances in food Res". 18:1. Academic Press, New York. Hills, G. L., and Conochie, J. 1946. The oxidant effect of commercial salt in fats and oils. J. Council of Sci. Ind. Res. (Australia), 18:355. A 141 Hirano, Y. and Olcott, H. S. 1971. Effect of heme compounds on lipid oxidation. J. Am. Oil Chem. Soc. 48:523. Hornstein, I., Crowe, P. F. and Heimberg, M. J. 1961. Fatty acid composition of meat tissue lipids. J. Food Sci. 26:581. Hornstein, I., Crowe, P. F. and Hiner, R. 1967. Composition of lipids of some beef muscles. J. Food Sci. 32:650. Horvat, R. J., McFadden, W. H., Ng, H., Lane, W. C., Lee, A., Lundin, R. E., Schere, J.R. and Shepherd, A. D. 1969. Identification of some acids from autoxidation of methyl linoleate. J. Am. Oil Chem. Soc. 46:94. I.C.M.S.F. (International Commission on Microbiological Specifications for Foods). 1980a. Temperature. II. Effect of chilling (cooling, refrigeration). In "Microbial ecology of foods" Vol. 1. Factors affecting life and death of microorganisms. P. 6. Academic press. New York, London, Toronto, Sydney, San Francisco. I.C.M.S.F. (International Commission on Microbiological Specifications for Foods). 1980b. Poultry and poultry meat products. 1V. Frozen raw poultry and poultry meat products. In "Microbial ecology of foods" Vol. 2. Food Commodities P. 447. Academic Press. New York, London, Toronto, Sydney, San Francisco. I.C.M.S.F. (International Commission on Microbiological Specifications for Foods). 19800. Poultry and poultry meat products. V. Heat-processed poultry meat pro- ducts. In "Microbial ecology of foods" Vol. 2. Food Commodities P. 452. Academic Press. New York, London, Toronto, Sydney, San Francisco. Igene, J. O. and Pearson, A. M. 1979. Role of phospholi- pids and triglycerides in warmed-over flavor develop- ment in meat model systems. J. Food Sci. 44:1285. Igene, J. 0., King, J. A., Pearson, A. M. and Gray, J. I. 1979. Influence of heme pigments, nitrite and non-heme iron on development of warmed-over flavor (WOF) in cooked meat. J. Agric. Food Chem. 27:838. 142 Igene, J. 0., Pearson, A. M. and Gray, J. I. 1981. Effect of length of frozen storage, cooking and holding tempe- rature upon component phospholipids and the fatty acid composition of meat triglycerides and phospholipids. Food Chem. 7:289. Igene, J. 0., Yamauchi, K., Pearson, A. M. and Gray, J. I. 1985. Mechanisms by which nitrite inhibits the develo- pment of warmed-over flavor in cured meat. Food Chem. 18:1 Ingold, K. U. 1962. Metal catalysis. In "Symposium on foods: lipids and their oxidation". Edited by Schultz, H. W., Day, E. A. and Sinnhuber, R. 0., P. 93. The AVI Publishing Co., Westport, CT. Issacks, R. E., Davies, R. E., Ferguson, T. M., Reister, R. and Couch, J. R. 1964. Studies on the avian fat composition. 1. Effect of dietary fat on the fatty acids of the triglyceride and phospholipid fraction of the blood plasma and adipose tissue lipids of the laying hen. Poultry Sci. 43:105. Jacobson, M. and Koehler, H. H. 1970. Development of rancidity during short time storage of cooked poultry meat. J. Agr. Food Chem. 18:1069. Kakuda, Y., Stanley, D. W. and Van de Voort, F. R. 1981. Determination of TBA number by high performance liquid chromatography. J. Am. Oil Chem. Soc. 58:773. Kanner, J. and Harel, S. 1985. Initiation of membranal lipid peroxidation by activated metmyoglobin and methe- moglobin. Archives of Biochemistry and Biophysics 237:314. Katz, M. A., Dugan, L. R., Jr. and Dawson, L. E. 1966. Fatty acid in neutral lipids and phospholipids from chicken tissues. J. Food Sci. 31:717 Keller, J. D. and Kinsella, J. E. 1973. Phospholipid changes and lipid oxidation during cooking and frozen storage of raw ground beef. J. Food Sci. 38:1200. Kendrick, J. and Watts, B. M. 1969. Acceleration and inhibition of lipid oxidation by heme compounds. Lipids 4:454. 143 Keskinel, A., Ayres, J. C. and Snyder, H. E. 1964. Deter- mination of oxidative changes in raw meats by the 2- Thiobarbituric acid method. Food Technol. 18:101. Kraybill, H. R., Dugan, L. R., Jr., Vibrans, F. C., Swartz, V. and Bondi, V. 1949. Studies on antioxidants. II. AMIF-72 as an antioxidant for animal fats. Am. Meat Inst. Found. Bull. No. 4. 6PP. Kunsman, J. E., Field, R. A. and Kazantzis, D. 1978. Lipid oxidation in mechanically deboned red meat. J. Food Sci. 43:1375. Labuza, T. P. 1971. Kinetics of lipid oxidation in foods. CRC Crit. Rev. Food Technol. 2:355. Lea, C. H. 1953. Recent development in the study of oxidative deterioration of lipids. Chem. and Ind. 1303. Lea, C. H. 1957. Deteriorative reactions involving phos- pholipids and lipoproteins. J. Sci. Fd. Agr. 8:1. Lea, C. H. 1962. The oxidative deterioration of food lip- ids. In "Symposium on foods: lipids and their oxi- dation". Edited by Schultz, H. W., Day, E. A. and Sinnhuber, R. O., P. 3. AVI Pub. Co., Westport, CT. Liu, H. P., and Watts, B. M. 1970. Catalysts of lipid oxidation in meats. 3. Catalysts of oxidative ranci- dity in meats. J. Food Sci. 35:596. Love, J. D. and Pearson, A. M. 1971. Lipid oxidation in meat and meat products. A review. J. Am. Oil Chem. Soc. 48:547. Love, J. D. and Pearson, A. M. 1974. Metmyoglobin and nonheme iron as pro-oxidants in cooked meat. J. Agr. Food Chem. 22:1032. Love, J. D. and Pearson, A. M. 1976. Metmyoglobin and non—heme iron as pro-oxidants in egg-yolk phospholipid dispersions and cooked meat. J. Agric. Food Chem. 24:494. Love, R. M. 1966. Protein denaturation in frozen fish. X1. The proportion of tissue water converted to ice. J. Sci. Fd. Agric. 17:465. 144 Marion, W. W. and Forsythe, R. H. 1964. Autoxidation of turkey lipids. J. Food Sci. 29:530. Marion, J. E. and Miller, W. O. 1968. Phospholipids and component fatty acids of chicken tissues. Poultry Sci. 47:1453. Maxey, R. B., Froning, G. W. and Hartung, T. E. 1973. Microbial quality of ground poultry meat. Poultry Sci. 52:486. Mecchi, E. P., Pool, M. F. and Klose, A. A. 1953. The role of tocopherol content in the stability of chicken and turkey fats. Poultry Sci. 32:915. Miller, E. C., Menge, H. and Denton, C. A. 1962. A compa- rison of fatty acid content of skin of chickens and turkeys. Poultry Sci. 41:1667. (Abstract). Moerck, K. E. and Ball, H. R., Jr. 1973. Lipids and fatty acids of chicken bone marrow. J. Food Sci. 38:978. Moerck, K. E. and Ball, H. R., Jr. 1979. Influence of microorganisms on the carbonyl compounds of chicken tissue. J. Agric. Food Chem. 27:854 Morris, S. G., Myer, J. S., Kip, M. L. and Riemenschneider, R. W. 1950. Metal deactivation in lard. J. Am. Oil Chem. Soc. 27:105. Morrison, W. R. and Smith, L. M. 1964. Preparation of fatty acid methyl esters and methyl acetals from lipids with boron fluoride methanol. J. Lipid Res. 5:600. MSTAT, 1985. A microcomputer program for the design, manag- ement and analysis of agronomic research experiments. Version 3.0. MSTAT development team, Michigan State University and Agricultural University of Norway. E. Lansing, Michigan. Nair, V. and Turner, C. A. 1984. The thiobarbituric acid test for lipid peroxidation: structure of the adduct with malonaldehyde. Lipids 19:804. Neudoerffer, T. S., and Lea, C. H. 1967. Effects of dietary polyunsaturated fatty acids on the composition of the individual lipids of turkey breast and leg muscle. Br. J. Nutr. 21:691. 145 O’Brien, P. J. 1969. Intracellular mechanisms for the decomposition of a lipid peroxide. 1. Decomposition of a lipid peroxide by metal ions, heme compounds and nucleophiles. Can. J. Biochem. 47:485. O’Keefe, P. W., Wellington, G. H., Mattick, L. R. and Stouffer, J. R. 1968. Composition of bovine muscle lipids at various carcass locations. J. Food Sci. 33:188. Pearson, A. M. and Gray, J. I. 1983. Mechanism responsi- ble for warmed—over flavor in cooked meat. "The Mail- lard reaction in foods and nutrition". Edited by Waller, G. R. and Feather, M. 8., ACS symposium series 215: 287. Pearson, A. M., Gray, J. 1., Wolzak, A. M. and Horenstein, N. A. 1983. Safety implications of oxidized lipids in muscle foods. Food Technol. 37:121. Pearson, A. M., Love, J. D. and Shorland, P. B. 1977. Warmed-over flavor in meat, poultry and fish. Adv. Food Res. 23:1 Peng, C. Y. and Dugan, L. R., Jr. 1965. Composition and structure of phospholipids in chicken muscle tissues. J. Am. Oil Chem. Soc. 42:533. Pikul, J., Laszczynski, D. E., Bechtel, P. J. and Kummerow, F. A. 1984. Effect of frozen storage and cooking on lipid oxidation in chicken meat. J. Food Sci. 49:838. Pikul, J., Leszczynski, D. E. and Kummerow, F. A. 1983. Elimination of sample autoxidation by butylated hydro- xytoluene addition before thiobarbituric acid assay for malonaldehyde in fat from chicken meat. J. Agric. Food Chem. 31:1338. Pryor, W. A., Stanley, J. P. and Blair, E. 1976. Autoxi- dation of polyunsaturated fatty acids. 11. A suggested mechanism for the formation of TBA-reactive materials from prostaglandin-like endoperoxides. Lipids 11:370. Rapport, M. M. and Norton, W. T. 1962. The chemistry of lipids. Ann. Rev. Biochem. 31:103. Rhee, K. S. 1978. Minimization of further lipid peroxida- tion in the distillation 2-thiobarbituric acid test of fish and meat. J. Food Sci. 43:1776. 146 Robinson, M. E. 1924. Myoglobin and metmyoglobin as oxi- dative catalyst. Biochem. J. 18:255. Rouser, G., Galli, C. and Lieber, E. 1964. Analytical fractionation of complex lipid mixtures: DEAE cellulose column chromatography combined with quantitative thin layer chromatography. J. Am. Oil Chem. Soc. 41:836. Ruenger, E. L., Reineccius, G. A. and Thompson, D. R. 1978. Flavor compounds related to the warmed-over flavor of turkey. J. Food Sci. 43:1199. Salih, A. M., Smith, D. M., Price, J. F. and Dawson, L. E. 1986. An improved extraction method for determining 2- thiobarbituric acid values of poultry meat. Poultry Sci. 65:118. SAS, 1985. Statistical Analysis Systems. Version 5.08. SAS Institute Inc., Cary, North Carolina. Used at Michigan State University, Computer laboratory. Sato, K. and Hegarty, G. R. 1971. Warmed-over flavor in cooked meats. J. Food Sci. 36:1098. Sato, K. and Herring, H. K. 1973. The chemistry of warmed- over flavor in cooked meats. In proc. of 26th. ann. reciprocal meat conference of the American Meat Scie- nce Assoc. P. 64. Schricker, B. R. and Miller, D. D. 1983. Effect of coo- king and chemical treatment on heme and nonheme iron in meat. J. Food Sci. 48:1340. Schricker, B. R., Miller, D. D. and Stouffer, J. R. 1982. Measurement and content of nonheme and total iron in muscle. J. Food Sci. 47:740. Sinnhuber, R. O. and Yu, T. C. 1977. The 2-thiobarbit— uric acid reaction. An objective measure of the oxida- tive deterioration occurring in fats and oils. J. Japanese Soc. Fisheries 26:259. Sinnhuber, R. O. and Yu, T. C. 1958. 2-thiobarbituric acid method for the measurement of rancidity in fishery products. 2. The quantitative determination of malona- ldehyde. Food Technol. 12:9. 147 Siu, G. M. and Draper, H. H. 1978. A survey of the malon— aldehyde content of retail meats and fish. J. Food Sci. 43:1147. Smith, G. J. and Dunkley, W. L. 1962. Initiation of lipid peroxidation by reduced metal ions. Arch. Biochem. Biophys. 98:46. SPSS, 1984. Statistical Package for the Social Sciences. Version 9.0. Michigan State University. Computer laboratory. Steel, G. D. and Torrie, J. H. 1960. "Principles and procedures of statistics". McGraw—Hill Book Co., New York. Swern, D. 1961. Primary products of olefinic autoxid- ations. In "Autoxidation and antioxidants". Vol. 1. Edited by Lundberg, W. O. Interscience, New York. Tappel, A. L. 1952. Linoleate oxidation catalyzed by big muscle and adipose tissue extract. Food Res. 17:550. Tappel, A. L. 1955. Unsaturated lipids oxidation cata— lyzed by hematin compounds. J. Biol. Chem. 217:721. Tappel, A. L. 1962. Hematin compounds and lipoxidase as biocatalysts. In "Symposium on foods: lipids and their oxidation". Edited by Schultz, H. W., P. 122. AVI Pub. Co., Inc., Westport, CT. Tarladgis, B. G. 1961. A hypothesis for the mechanism of the heme catalyzed lipid oxidation in animal tissues. J. Am. Oil Chem. Soc. 38:479. Tarladgis, B. G. and Pearson, A. M. and Dugan, L. R., Jr. 1962. The chemistry of the 2-thiobarbituric acid test for the determination of oxidative rancidity in foods. 1. Some important side reactions. J. Am. Oil. Chem. Soc. 39:34. Tarladgis, B. G., Pearson, A. M. and Dugan, L. R., Jr. 1964. Chemistry of the 2-thiobarbituric acid test for determination of oxidative rancidity in foods. 2. Formation of the TBA-malonaldehyde complex without acid-heat treatment. J. Sci. Food Agric. 15:602. 148 Tarladgis, B. G., Watts, B. M., Younathan, M. T. and Dugan, L. R., Jr. 1960. Distillation method for the quanti- tative determination of malonaldehyde in rancid foods. J. Am. Oil Chem. Soc. 37:403. Tichivangana, J. 2. and Morrissey, P. A. 1985. Metmyoglo- bin and inorganic metals as pro-oxidants in raw and cooked muscle systems. Meat Sci. 15:107. Tsai, Lee-Shin and Smith, L. M. 1971. Role of the bases of phosphonyl bases of phospholipids in the autoxida- tion of methyl linoleate emulsions. Lipidszl96. Wangen, R. M., Marion, W. W. and Hotchkiss, D. K. 1971. Influence of age on total lipids and phospholipids of turkey muscle. J. Food Sci. 36:560. Waters, W. A. 1971. The kinetics and mechanisms of metal catalyzed autoxidation. In "Symposium: metal catalyzed lipid oxidation". Presented at the ISF-AOCS world Congress, Chicago, IL. Watts, B. M. 1954. Oxidative rancidity and discoloration in meat. Adv. Food Res. 5:1. Webster, G. R. 1960. Studies on the plasmalogens of nervous tissue. Biochem. Biophys. Acta. 44:109. Wills, E. D. 1965. Mechanisms of lipid peroxide formation in tissues. Role of metals and hematins in the cataly— sis of the oxidation of unsaturated fatty acids. Bio- chem. Biophys. Acta. 98:238. Wills, E. D. 1966. Mechanisms of lipid peroxide formation in animal tissues. Biochem. J. 99:667. Wilson, B. R. 1974. Effect of turkey lipids and phospho- lipids on warmed-over flavor measured by TBA analysis in muscle from several species. M. S. Thesis, Michigan State University, East Lansing. Wilson, B. R., Pearson, A. M., and Shorland, F. B. 1976. Effect of total lipids and phospholipids on warmed—over flavor in red and white muscle from several species as measured by TBA analysis. J. Agric. Food Chem. 24:7. Witte, V. C., Krause, G. F. and Bailey, M. E. 1970. A new extraction method for determining 2—thiobarbituric acid values of pork and beef during storage. J. Food Sci. 35:582. 149 Yamauchi, K. 1972. Effect of heat treatment on the deve- lopment of oxidative rancidity in meat and its isolated tissue fraction. Bull. Fac. Agric., Miyazaki Univ. 19:147. Younathan, M. T. and Watts, B. M. 1959. Relationship of meat pigments to lipid oxidation. J. Food Sci. 24:737. Younathan, M. T. and Watts, B. M. 1960. Oxidation of tissue lipids in cooked pork. Food Res. 25:538. Zipser, M. W., Kwon, T. W. and Watts, B. M. 1964. Oxida- tive changes in cured and uncured frozen cooked pork. J. Agric. Food Chem. 12:105. MICHIGAN STATE UNIV. LIBRARIES mllWI"HIWIWHIWI"IIWI‘IIIUIIIUWWWII 31293006606291