' -—-— q ' ‘ ' . "‘ V‘p'h '. 1. 1": "1’ N. _.» 3 .. - ‘3. '1 1-}1x.-G;,;;:A” .‘Sfifi 1 .1. .1 ‘. V l . - A ..< k On I <. K. .1 .\ 1 {1,3} . $.13 '7 1 15. I.) r- 3‘. 1 _ 0' , . . I 'rfi -F 24:47 h. ‘hétg‘; . 1'19. 1:»; "j 3' . 1. . H.“ 'In“, *6: A31 ‘ 1 ' .. f. .57 - \n V$m133%%2' ‘ . .3 f: “sh “Ia; 1:115. "r “I. 1;. ”‘1. d. '1 145‘ 311.1211 ' '1 K“ Wu?“ 'v .1111 :I'o'__l "1‘6; R 5. “19‘? -. . < 1 ‘ I. I. '0‘. ' #4‘ V I I“ ' '-‘ j ' l 0 ”kn :- . . 11.. 1.1111 1-1;...1W .97 ‘M 1:1111"”1.”‘ - $3141”? “'1 . ‘5 L"; nl 1<\»I;I 11%») E151. 1.11:1: 11‘" ' " : IN I . 1 .|. ’1 ' '1 31-1 Hggbfl 1' L41 .1; . ‘ . 'I 1 ‘ . ‘\»{v \ 5‘ 11:13.3). max?“ “5 - *1 11 m 3‘ ain’t-T" I“ S; "I \éét‘ " - . \‘n ‘0”':Q1f--y1- o 1 ‘ .11 “1 «5+ MM: . \ 'éifi'mwz '3". ' . Viv}? ai>lffip 4% 0f 5, ‘ 1 ‘ I - ,3' I‘ll}. 1%; “J: . as- 3,1" .i_ . ‘ r‘l . ‘ 11R). 1.: i I I ‘5‘.- l: H J‘l rut. k ‘17.},le 1, "M ‘. I“ "5 ' 4'0 "’4 . _ . > ‘ n'-" '- '1'. 1 13 u ‘1‘ 1 “r‘h’ QB I f . . I 3"“ I - g‘ "4 .1 ’ I V... I"‘Ih‘. 'I‘ ‘ 1 r I: I- ' , .1 - Vs ._ - 3:- _-: Er‘ - - ‘ '5' .57 ‘ ., x [I] VT. .%:"”"',-“ ‘III‘A; 3. {if‘. 39"?“ 1'57" gfiéjri .y'. "1. 3i: szfldtimfl g 1.3“ #2:; . 7' ~ 1' 211.111 1 :T . “'32.':{;§;'t")'r( ' '1' JV.” l I 21 ' U .1 \ “.40“ '1 “g '1 ‘| 1“» ‘ .. Sufi)”: - ‘- "'1:».'"11£‘«: 2. I I "3‘2“;th 11’1”“3'3 . . ~11“ ‘”W”Vfifii. I ' 1. :‘:i" ' - $4}: . "P I: N I J. 1:11.119“ ~ I, 'I u 6 _‘V, v, ”,1. 1‘.‘ I: :3} l " ‘1‘ " “f!" '-' II I. '3], ‘. "Izg'r‘ " ‘ln‘¢f""*f~‘ p111; wfl " 1' 1» :1;';-. 1.1+ « ~.:;:.'- . ,. . . 1 ‘_ 1... :1 .; .1' o-‘JCIJJIP‘. ”‘11 1, I41 .' 1 1 _.' ‘ 1‘ , 1- . 1 J: . F‘ . u .. ”I. a s l. 1 um ”4 TI“: '.-{5.‘%;'WY-‘.1;\‘. . If“ ‘ 1 .2. I ' 1 ‘ 9 I . ${ ' "1 ""1I“?*f‘;311‘-11:..1.§7§,J.1‘91 I; )131515Kéfi1 5‘ Idulfl'fi‘l’ A; .I I ‘ ' ”"3; 4%. . “‘|‘ ' 1 I: :L' " '| W" ”1:11;. ”5.1....111111“ '1 :“_- '- "’l I Y’ “'9 '. 11:». 1111:." :‘ll'o’ ‘1 | . ‘ tll‘l I I E_...“ fl, l::'{“” I...“ "11' 1’?!‘ (I .. ,f 5‘5; :1.) " l'1 . "‘k‘ I H! “11.1! ”1"” -- 4;. 3511,1151. ' ':I 131' 'JIIF" 11";1311 1,11" ,k 1‘! , .' ‘ 4,151- ;'1- 51,, 1‘ 4' ‘11." I . 1011;151:4411- I 1'17. 1'. . ”ct-11"...” _ 1' -.‘.ff.l'1:1'I-\.-' 91”.. 1.10. |1"' ‘E‘Uccw This is to certify that the thesis entitled EFFECT OF HYDROLYZED VEGETABLE PROTEIN, AUTOLYZED YEAST PROTEIN AND BUTYLATED HYDROXYANISOLE ON THE STABILITY OF PALM OIL 1%résebgqlgglbySYSTEM Mark E . Ukhun has been accepted towards fulfillment of the requirements for Masters - Food Science & Human n degree 1 Nutrition May 18, 1978 I)ate 07639 EFFECT OF HYDROLYZED VEGETABLE PROTEIN, AUTOLYZED YEAST PROTEIN AND BUTYLATED HYDROXYANISOLE ON THE STABILITY OF PALM OIL IN A MODEL SYSTEM By Mark E. Ukhun A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Food Science and Human Nutrition 1978 ABSTRACT EFFECT OF HYDROLYZED VEGETABLE PROTEIN, AUTOLYZED YEAST PROTEIN AND BUTYLATED HYDROXYANISOLE ON THE STABILITY OF PALM OIL IN A MODEL SYSTEM By Mark E. Ukhun The objective in this work was to demonstrate the antioxidant properties of hydrolyzed vegetable protein (HVP) and autolyzed yeast protein (AYP) in a model system composed of varying levels of either the hydrolyzed vege- table protein or autolyzed yeast protein, the antioxidant butylated hydroxyanisole (BHA), carboxymethyl cellulose (CMC), palm oil and distilled water. The system was freeze- dried and allowed to oxidize under controlled conditions. The rate of oxidation at 600C was monitored by measuring weight gains in the model system and by deter- mining absorbance attributable to diene conjugation at 233 nm in the extracted oil every two days. Under the conditions of this experiment, butylated hydroxyanisole, hydrolyzed vegetable protein and autolyzed yeast protein were found to improve the stability of palm oil in the model system. The degree of stabilization con- ferred by these antioxidants was dependent on their concen- tration in the model system. Antioxidant effectiveness was Mark E. Ukhun in the order BHA > AYP > HVP. Combinations of these antioxidant compounds led to greater stabilization of the palm oil but only negative synergism could be demonstrated with BHA and HVP combina- tions. The greater stabilization was also observed to be concentration dependent. Good correlation between the weight gain and absor- bance due to diene conjugation at 233 nm was obtained when both methods were used to monitor the rate of oxidation of the palm oil in the model system. Dedicated to the only God of the universe ii ACKNOWLEDGEMENTS The author greatly appreciates the skillful commitment of his major advisor, Professor L.R. Dugan, Jr. to his overall academic growth in Michigan State University. For their time and effort in serving on my committee and for critically reading this manuscript, no less appre- ciation goes to Professor P. Markakis, Professor L.E. Dawson and Professor T.R. Pierson. Finally, the moral support of my colleagues in the Food Science Lipid Laboratory of Michigan State University is appreciated. iii TABLE OF CONTENTS LIST OF TABLES. LIST OF FIGURES INTRODUCTION. LITERATURE REVIEW Palm Oil. . . Mechanism of Lipid Oxidation. . Lipid Oxidation in Food Systems Lipid- Protein Interaction Role of Antioxidant . . Mechanism of Antioxidant Action . Proteins and Amino Acids as Antioxidants. Role of Water in Fat Stability. . . Evaluation of the State of Oxidation. EXPERIMENTAL. Materials and Equipment Methods . Determination of Iodine Number of. Palm Oil. . . Determination of Free Fatty Acid . Preparation of Methyl Esters for Gas- Liquid Chromatography. Gas-Liquid Chromatography. Moisture Determination Percentage Recovery. Purification of 150- octane . . . Preparation of the Model Systems Weight Gain Studies. . . . . . . . . . . Extraction of Palm Oil from Freeze—dried Materials. . . Measurement of Absorbance Changes. RESULTS AND DISCUSSION. Moisture Content Percentage Recovery. iv Page vi —J NM—l—J—J—l—J commune—401p #- N KO LON OLD woo QC) wwwwwww MWWNN—JO 0000 014:- (.20 CW coco 0000 Page Effect of BHA on the Stability of Palm Oil. . . . 40 Effect of HVP on the Stability of Palm Oil. . . . 45 Effect of Autolyzed Yeast Protein .(AYP) on the Stability of Palm Oil . . 52 Effect of the Combination of HVP or AYP and BHA on the Stability of Palm Oil . 55 Comparison of the Antioxidant Effects of BHA, HVP and AYP in Palm Oil . . . 74 Comparison of Weight Gain Measurements to Absorbance Values . . . . . . . . . 76 SUMMARY AND CONCLUSIONS . . . . . . . . . . . . .. 79 BIBLIOGRAPHY.................... 89 Table 10 LIST OF TABLES Fatty Acid Composition of the Palm Oil. Characteristics of the Palm Oil Percentage Recovery of Palm Oil from the Freeze-dried Systems Using Chloroform Alone and a Chloroforszethanol Mixture (50:50 v/v) Weight Gain, Increase in Absorbance and Induction Periods of Palm Oil for Various Levels of Butylated Hydroxyanisole in the Model System. . Weight Gain, Increase in Absorbance and Induction Periods of Palm Oil for Various Levels of HVP in the Model System Weight Gain, Increase in Absorbance and Induction Periods of Palm Oil for Various Levels of AYP in the Model System Induction Periods for Combinations of Hydro- lyzed Vegetable Protein and Butylated Hydrox- yanisole in the Model System. . . . Induction Periods for Various Combinations of Autolyzed Yeast Protein and Butylated Hydroxyanisole in the Model Systems Synergism in Model Systems of HVP, BHA, Palm Oil and CMC, Stored at 600C for 27 days Synergism in Model Systems of AYP, BHA, Palm Oil and CMC, Stored at 600C for 43 days vi Page 37 37 39 42 47 51 56 57 59 6O Figure l0 Weight changes in model systems of BHA, CMC and palm oil, Weight gain of CMC, HVP, 60°C for 27 Weight gain of CMC, HVP, 60°C for 27 Weight gain of CMC, HVP, 600C for 27 Weight gain of CMC, AYP, 60°C for 43 Weight gain ofO CMC, AYP, 600 C for 43 Weight gain of CMC, AYP, 60°C for 43 Absorbance LIST OF FIGURES stored at 600C for 27 days. in freeze- dried model systems BHA and palm oil stored at days. . . . . in freeze- dried model Systems BHA and palm oil stored at days. . . . . . . . . in freeze-dried model systems BHA and palm oil stored at days. . . . . . . . . in freeze- dried model systems BHA and palm oil stored at days. . in freeze- dried model systems BHA and palm oil stored at days. . . . . . . . . in freeze- dried model systems BHA and palm oil stored at days. . . . . . . . . increase in model systems of CMC, HVP, BHA and palm 011 stored at 60°C for 27 days Absorbance increase in model systems ofO CMC, HVP, BHA and palm oil stored at 600 C for 27 days Absorbance increase in model systems of CMC, HVP, BHA and palm 011 stored at 60°C for 27 days vii Page 43 62 63 64 65 66 67 68 69 70 Figure Page ll Absorbance increase in model systems of CMC, AYP, BHA and 001 011 stored at 60°C for 43 days . . . . . . . . . . . . . . . 7] l2 Absorbance increase in model systems ofO CMC, AYP, BHA and palm oil stored at 60 C for 43 days . . . . . . . . . . . . . . . . 72 l3 Absorbance increase in model systems ofO CMC, AYP, BHA and pal oil stored at 60 C for 43 days . . . . . 73 viii INTRODUCTION Today's grocery shops are filled with a formidable number of foods prepared as convenience foods while others are for institutions. Whether it is at the shop level or at the home level,including institutions, it is important that such foods retain their qualities if they are to remain acceptable over a considerable and reasonable period of time. To achieve this, many of them, which contain fats or lipids, are stabilized with antioxidants, of which only a few are approved by law. Butylated hydroxytoluene (BHT), butylated hydroxyani- sole (BHA), propyl gallate (PG) and tert-butylhydroquinone (TBHQ), all of which are phenolic, are among others which are used in food stabilization. Given the fact that many of the antioxidants in use are phenolic, synthetic, and that consumers are beginning to point accusing fingers at them, it is desirable, if not imperative, that alternative sources of antioxidants such as natural and non-phenolic antioxidant substances be searched for. There are reports in the literature of certain pro— teins and amino acids interacting with known antioxidants to impart stability to fats or oils in model systems. Working with a soup and gravy base, Bishov t al. (I967) reported that hydrolyzed vegetable protein had antioxidant proper- ties. In a study by Koch _t al, (l97l) linoleate salts of lysine and arginine were shown to exhibit resistance to autoxidation. A potential exists for interaction between proteins and lipids in a food system with intermediates such as hydroperoxides, peroxy- and oxy-radicals, and carbonyl com- pounds entering into a variety of interactions to possibly account for the changes observed in such a system. However, because of the complex nature of the interactions involved and because of a number of factors such as pH, ionic strength, temperature, light, etc. which can influence results, it has not yet been possible to apply such results in a standard way to produce formulations. The present study was undertaken to (i) determine if two proteins, hydrolyzed vegetable protein (HVP) and autolyzed yeast protein (AYP), have any antioxidant activity in a model system, the lipid component of which is palm oil. (ii) find out if synergism exists between these proteins and the phenolic antioxidant BHA. (iii) find out the effect of varying the concentrations of the protein and the phenolic antioxidant, individually and when used in combination. Palm oil was chosen, despite its known relative sta- bility, because of its increasing importance in industries, especially in tropical Africa, in food processing and in other processes involving the incorporation of oil. More- over, room exists for improving its stability. LITERATURE REVIEW Palm Oil Useful applications of this oil have ranged from a hand cream and medicine in l8th century Britain to a lubri- cant in the modern tin plate manufacturing industry in the United States. Most palm oil today is, however, consumed as edible oil, especially in shortening and margarine, while advances in fractionation methods have expanded its use in the manufacture of salad oils. From the chemical standpoint, refining is one of the most important steps in the processing of palm oil and it includes: (l) Neutralization: This involves treating crude oil with an alkali such as caustic soda, to effect a nearly com- plete removal of the free fatty acids. These acids (and a small amount of neutral oil) are saponified - converted into soap - by the process. (2) Bleaching: This requires either heating the oil or treating it with bleaching earth to remove carotenoid pig- ments, which impart the red color characteristic of crude palm oil. The carotenoid pigment converts to vitamin A in man and other animals; hence removal of this pigment to meet consumer preferences for clear cooking oil or white shor- tening also reduces nutritive value of palm oil. 4 (3) Deodorization: This is accomplished by steam heating the relatively non-volatile crude oil to remove relatively volatile taste and odor substances, while main— taining low pressures to prevent atmospheric oxidation of the oil. (4) Winterization: The major triglycerides, or fat molecules, in palm oil are oleo-dipalmitin - melting temper- ature of 35°C (96°F) — and palmitodiolein - melting tempera- ture of l9OC (66°F). Therefore, at room temperature the latter group of triglycerides normally will melt while the former remains solid, separating the oil into liquid and solid fractions. Palm oil thus can also be fractionated through the winterization process. Chilled crude-oil runs onto a conveyor belt, which suspends the solid portion (stearin) while allowing the liquid portion (olein) to separate. Stearin serves as a pastry shortening and cocoa butter substitute, while olein is more suitable as a cooking and salad oil. Stability of Lipids Dugan (l976) has described stability as the capability of a fat, oil, or fatty food to maintain a fresh taste and odor during storage and use and it is related to composition of the lipid moiety, the nature and degree of stress on the system, the presence or absence of pro-oxidants and antioxi- dants, and the effectiveness of packaging. He added that fats with substantial unsaturation in the fatty acids are usually unstable or moderately unstable and foods containing them reflect this instability; but that vegetable oils usually tend to be more stable than some of the animal fats, such as lard, even though the total unsaturation of the vegetable oils may be greater because natural antioxidants are usually present in the vegetable oils. Smith and Dunkley (l962) and Waters (l97l) have also referred to the promoting effects of light, especially of short wave length, heat and metal catalysts, especially iron and copper, in the autoxidation of the lipid components of foods. Because of its comparatively high level of palmitic and oleic acids, palm oil is regarded as stable when com- pared to many other oils of plant origin. The stability of the lipid or fat component of a food is, of course, related and relatable to the level of deterioration of that food. Lipolysis is a term which refers to the hydrolysis of ester linkages of lipids resul- ting from enzymes, from thermal stress, or from chemical action. A major cause of food deterioration is also attributable to fat rancidity as a result of oxidation - ”oxidative rancidity”. Lipolytic rancidity is thought to pose less of a flavor problem than oxidative rancidity because the former develops off-flavor only in those fats which contain short-chain fatty acids (less than C14). Mechanisms of Lipid Oxidation The reaction of oxygen with unsaturated fatty acids in lipids constitutes the major means by which lipids or containing foods deteriorate. Oxidation of fat is fre- quantly alluded to as autoxidation because the rate of oxidation increases as the reaction proceeds. Although this process, which involves the uptake of atmospheric oxygen, is common to most foods containing plant or animal tissue, the mechanisms involved may vary (Tappel, l953, l955; Lea, l962). Dugan (l976) has described the oxidation process as proceeding through a free-radical chain reaction mechanism involving three stages: (l) initiation, formation of free radicals; (2) propagation, free-radical chain reac- tion, and (3) termination, formation of non—radical products. In the initiation stage an unsaturated hydrocarbon loses a hydrogen to form a radical: RH + R- + H- and oxygen may add at the double bond to form a diradical: H H I l R - C C - R' + 02 + R - R' H l C 0—0— I l _ 0. It is also possible for oxygen in the singlet state to interpose between a labile hydrogen to form a hydroperoxide directly (RH + 02 + ROOH). During propagation, the chain reaction is continued by R- + 02 + R00- and R00- + RH + ROOH + R' to form peroxy- radicals, hydroperoxides, and new hydrocarbon radicals. The new radical formed then contributes to the chain by reacting with another oxygen molecule. When two radicals interact, termination occurs: R' + R' + RR ROO~ + ROO- + ROOR + 02 R0- + R~ + ROR ROO' + R- + ROOR 2R0° + 2R00- + ZROOR + 02 Because of the high energy necessary for the rupture of a CH bond (about 80 Kcal) and for diradical formation at the double bond, it is thought that energy reducing pheno- mena such as by metal activation, enzyme catalysis or photo- oxidation might be involved (Farmer and Sutton, l943; Bolland and Koch, l945). Oxidation of Monoenoic Acids: When a monoene is oxi- dized, hydrogen is abstracted, during the initiation stage, from either of the carbons 0’to the double bond, so that two radicals are possible, each of which can assume two forms through resonance. When oxygen and hydrogen are added at each radical site, hydroperoxides are formed. Oxidation of Polyenoic Acids: Oxidation of polyenes is initiated more readily and proceeds more rapidly than oxidation of monoenes. Most natural polyunsaturated fatty acids have l,4-pentadiene structures: 1 l 1 I R - C = C - C - C = C - R l Because the radical formed by the abstraction of hydrogen from the methylene group having contiguous double bonds on both sides is unstable, the electrons redistribute to form a conjugated double bond system. Thus a radical is formed at either the l or the 5 carbon of the original pentadiene system. Oxidation of Saturated Acids: Severe conditions are necessary for the oxidation of saturated acids and the hydroperoxide group generally forms on the carbon 8 to the carboxyl or ester group. The hydroperoxide then changes to a keto group, giving rise to a B-keto acid. Hydroperoxides are readily decomposed by high-energy radiation, thermal energy, metal catalysis, or enzyme acti- vity, with the means depending on the system in which they exist. The oxidation then becomes more complex after the development of a quantity of ROOH with the formation of more free radicals which add to the chain process (Dugan, l96l). Farmer and Sutton (l943) have also made mention of the role of thermal instability in the decomposition of hyderperox— ides. The decomposition of a hydroperoxide can take place by a monomolecular as well as by a bimolecular mechanism (Mabrouk and Dugan, l960: Lundberg, l962) as Shown below: Monomolecular ROOH -—-9 R0- + -OH Bimolecular 2ROOH ——-> 2120- + H20 Mabrouk and Dugan (I960) reported a lower activation energy for the bimolecular than for the monomolecular mecha- nism for hydroperoxide breakdown while lower hydroperoxide 10 concentration and higher temperatures were said to promote monomolecular decomposition. The proliferation of radicals causes acceleration of oxidation without requiring new initiation events. In his own contribution to the explanation of the autoxidation process, Baddings (l960) has stated that: "(l) The rate of autoxidation is dependent on the energy required for the rupture of an 0—CH bond. (2) The reaction can be accelerated by light, trace metals, biological catalysts and radical-forming products such as benzoyl peroxide. (3) Autoxidation is inhibited by compounds which react with free radicals to form non-radical products. Conse- quently, the free radical chain reaction may be interrupted and the formation of chain initiating free radicals can be prevented. A number of compounds, such as certain phenols, inhibit oxidation when present in small amounts. (4) When a hydroperoxide decomposes to form RO- radi— cals, these in turn are capable of a series of reactions leading to several products which can be isolated from oxi- dizing lipid system; and the following reactions are apparently involved. Some of these products are radicals and they are capable of continuing in the chain propagation process." Others, such as the hydroxy acids, keto acids, and aldehydes, are commonly found in oxidizing lipid systems. .The aldehydes, many of which are short chain, and the short- chain acids derived by further oxidation of those aldehydes, ll _ C = l H H 0 0 + + R'OH R'- are largely responsible for the off-flavor and odors charac- teristic of stale or rancid foods. Lipid Oxidation in Food Systems: It is known that in a refined fat or oil, oxidation usually proceeds through autocatalytic processes, reaction rate increasing with time as a result of product formation which catalyzes the reac- tion; but as Dugan (l96l) and Lundberg (l962) pointed out, in foods the processes are more complex and factors such as proteins, carbohydrates, moisture, salt and enzymes influence the rate and course of the oxidation process. In consonance with this, Lea (l962) also stated that minor food components such as the natural aroma, flavor and coloring substances which make food attractive, as well as nutritionally impor- tant vitamins can also increase the susceptibility of a food to autoxidative deterioration. Lipid-Protein Interactions: The interaction between lipids and protein and/or amino acids in food systems is 12 increasingly being investigated especially in model systems. The exact mechanism has not yet been fully established. Possibly, the binding of lipids to proteins or vice-versa may depend on such factors as configuration, multiple attachment sites, and/or the matching of polarity between opposing groups in much the same manner as for the combina-4 tion of enzyme and substrate. Chapman (l969) referred to metal ion, electrostatic and hydrophobic binding as being important in the interac- tion between lipids and proteins. In consonance with this, Gurd (l960) suggested that the interactions might be between similar types of functional groups in the two classes of compounds, such as between non-polar hydrophobic residues of the fatty acid moieties of lipids and the similar resi- dues of certain of the Side chain groups of proteins while other interactions may involve polar or charged groups. Primary covalent linkages, such as ester bonds were regarded as being of little importance. The consequences for the food system of interaction between proteins and lipids can be classified into those affecting the protein component of the food and those affecting the lipid component of the food. The effects of lipid oxidation during freeze-drying are well known: changes in rehydration properties, texture, off-flavor development and therefore the limitation of the shelf-life of the food are some of these. These observations prompted Koch (l962) to speculate that interaction between l3 the products of lipid oxidation and proteins does take place during the processing and storage of dried foods. In their studies on a gelatin-linoleate system in the dry state, Zirlin and Karel (l960) observed that gelatin- lipid reactions lead to the scission of protein coupled with cross-linking. Other workers have shown that oxidation of lipid- protein systems often leads to browning and copolymerization of oxidized lipids and proteins as depicted by losses in the chemical and physical properties of protein. Narayan and Kummerow (l958) have referred to complex formation in the presence of oxidized lipids with the reaction being environ- ment dependent and different proteins responding differently. Complex formation was attributed to secondary bonds such as hydrogen bonding. The isolation of fluorescent compounds with -C = N - functional groups from an oxidizing system consisting of methyl linoleate and coho salmon myosin and the observation that the e-amino groups of myosin were destroyed led Brad- dock and Dugan (l973) to speculate a cross-linking reaction between amino groups and the products of fatty-acid oxida- tion. Lysine, methionine and histidine were found to be the most susceptible to such destruction. Solubility changes and the loss of protein-amino—acids were reported by Roubal and Tappel (l966a) as being the result of enzyme and protein damage from peroxidizing lipids with polymerization of proteins with polyunsaturated lipids l4 occurring as oxidation progresses; while radical attack, rather than aldehyde attack, on protein is mainly respon- sible for protein damage in a lipid-protein system (Roubal and Tappel, l966b). Working with carbonyl compounds, Schwenke (l975) stated that blocking of a-e-amino groups alters the iso- electric point, electrophoretic properties, solubility and precipitation characteristics of proteins. This led to the postulation that radicals from oxidizing lipids play a role in protein destruction with the possibility that carbonyl- amine reactions, as in Maillard browning, form Schiff bases and secondary products. The observations by Jarenback and Liljemark (l975) that linoleic acid hydroperoxides were much more effective than linoleic acid in reducing the amount of protein in a KCl-extract led them to suggest more extensive binding between the linoleic acid hydroperoxides and protein than between the linoleic acid and protein. The other side of the picture of a protein-lipid system is the stabilizing effect certain proteins have on the lipid component of a food system. Most of the work in this area has been done using freeze-dried model systems. Role of Antioxidant: Dugan (l976) has described an antioxidant as a ”substance that is added to fats or fat- containing foods to retard oxidation and thereby prolong their wholesomeness and palatability.” Antioxidants interfere with or delay the onset of the oxidative breakdown TB of fats and fatty foods (Blanck, l955). Ideally, an antioxidant should (i) have no harmful physiological effects; (ii) not contribute an objectionable flavor, odor, or color to the fat or food in which it is used; (iii) be effective in low concentrations; (iv) be fat soluble; (v) persist following processing to provide effective pro- tection to food in which it exists; (vi) be readily available; and (vii) be economical. Butylated hydroxyanisole (BHA) (a mixture of 2-t- butyl-4 hydroxyanisole and its isomer 3-t—butyl-4-hydroxyani- sole), butylated hydroxytoluene (BHT) which is 2,6-di-t- butyl-4-methylphenol, esters of gallic acid, and di-tert- butyl hydroquinone are commonly added to food since they are effective and they comply with safety regulations of the Federal Food and Drug Administration of the United States. These antioxidants are illustrated below: OH A CH3 l l CH3 and l O'CH3 (BHA) l6 CH3 OH CH3 OH CH3 -C-"" *C --CH3 HO—' \‘—‘0H CH3 CH3 ,/ H3 OOCHZCHZCH3 (BHT) (PG) H3C 0H CH H3C CH3 H (TBHQ) Many naturally occurring substances such as the tocopherols (a, 8, v, 0), lecithin or mixtures of phosphatides, flavones, sterols, and sulfhydryl compounds have some antioxidant properties. BHA is synergistic with BHT and propylgallate, and with several antioxidants. This means that when used in combination these antioxidants are more effective than when used alone. However, the use of BHT with propyl gallate results in negative synergism, in which the keeping quality of a fat is less than expected from the sum of the effects of each used alone. Dugan and Kraybill (l956) also showed that the addition of BHA and the tocopherols beyond optimum levels resulted in negative synergism. l7 Mechanism of Antioxidant Action Cort (l974) has indicated that the phenolic antioxi- dants act as electron or hydrogen donors thereby quenching electron mobility and interrupting the free-radical chain reactions. In food systems, the most effective antioxidants function by interrupting the free-radical chain mechanism (Shelton, l959) while those used in gasoline, lubricants, rubber, and other applications may function as peroxide decomposers (Dugan, l976). Antioxidants, such as ascorbic acid, function by being preferentially oxidized and they afford relatively poor protection. An antioxidant AH reacts with radicals produced during autoxidation according to the scheme postulated by Uri (l96l): R' + AH + RH + A- R0; + AH + ROH + A: ROO- + AH + ROOH + A. R- + A- + RA R0- + A- + ROA This scheme shows that antioxidants interfere with the free- radical, chain oxidation process and that reaction products of antioxidant molecules and oxidized lipid molecules may appear among the final products. Relatively little evidence exists in support of the latter statement. It is also pos- sible that the antioxidant is oxidized directly by oxygen. Tappel (l962) has reported that tocopherol is partly l8 oxidized to tocoquinone. Some antioxidants are effective in prolonging the keeping time of both fats and oils and the foods containing them. Such antioxidants are known as ”carry through anti— oxidants” since they ”carry through” or survive the thermal stresses, steam distillation, and pH effects of processing to give longer shelf-life to the finished food (Dugan, l962). Proteins and Amino Acids as Antioxidants In some protein-lipid systems it would appear that the destructive effects to proteins is matched by a stabilizing effect to the lipid component of the system. Bishov and Henick (l972) reported an antioxidant effect of AYP and HVP in a freeze—dried model system of lipid, carboxymethyl cel- lulose, the protein, and distilled water. Earlier work by Bishov _t‘al. (l967) had indicated improved fat stability in dehydrated proteinaceous food mixes of chicken flavored soup and gravy mix; while the universality of synergism between protein hydrolyzate and phenolic antioxidants was shown by Bishov and Henick (l975). The ability of protein to impart stability in such a system was also found to vary with the type of protein, method of preparation, pH, and experimental condition. For example, in their work as reported above, differences in the stabilizing effect between the AYP and HVP were attri- buted to the milder conditions of autolysis in comparison to those of hydrolysis, which gave better stabilizing poten- tial to the former. HVP is obtained by hydrolyzing soy l9 protein with an acid followed by alkaline neutralization. The salt which may be thus formed is also thought to lower its antioxidant effectiveness. AYP, on the other hand, is obtained with minimum heat treatment Since the yeast cells are self-digesting. In consonance with the above, Hayes _t al. (l977) has hinted, with respect to soy flour antioxidant, that there is no clear cut, generalized relationship between concen- tration of added soy flour and the degree of stability obtained. He added that the nature of other product ingre- dients, the character of the fat or oil involved and whether the soy flour is full fat or defatted all probably compli- cate the outcome. The antioxidant activity of soy flour has been ascribed to its natural component; such as isoflavones which Walz (l93l) isolated. Amino acids and peptides are others. Soybeans normally contain small quantities of peptides and amino acids which are present as a result of incomplete protein synthesis or possibly because of some protein degra- dation (Smith and Circle, l972a). Another class of compounds associated with the antioxidant activity of soy flour has been the phospholipids and Smith and Circle (l972a) have indicated that only part of the phospholipids are removed in commercial extraction of soybean flakes. In their studies with methyl linoleate in freeze- dried models, Karel gt al. (l966) found that certain amino acids including histidine, B-amino-butyric acid, lysine and 20 cysteine had substantial antioxidant activity and that the nature of this activity was different from that observed with propyl gallate since the main effect of the amino com- pounds was to prolong the induction period and to affect the initial rate of oxidation. No effect was present in the more rapid, bimolecular phase of oxidation, whereas propyl- gallate had an inhibiting effect in the latter stage. As a follow-up to this, Tjho and Karel (l969) reported that histidine in low concentration has an antioxidant effect in the very early stages of oxidation followed by a slight pro-oxidant effect in the later stages. Working with herring oil, Marcuse (l960) observed that while histidine had an antioxidant effect, cysteine was pro-oxidant. Marcuse (l96l) also reported on a study of the antioxidant effects of several amino acids added to aqueous solutions of linoleate at pH 7.5 and concluded that: (l) Histidine, alanine, methionine and lysine reduced oxygen absorption by linoleate by as much as 50-80%. (2) Each amino acid had an optimum concentration for the antioxidant activity and at high concentrations showed an activity inversion; becoming pro-oxidant rather than antioxidant in its action. (3) The antioxidant activity depended on pH, the pre- sence of other antioxidants or synergists, and the state of oxidation of linoleate. (4) Since antioxidant properties were shown in the absence of tocopherol and other antioxidants, amino acids 2T may act as primary antioxidants. Then in l962, Marcuse observed that the antioxidant effect is enhanced, and the pro-oxidant effect is lowered by an addition of phosphate. It has been indicated here earlier that carbonyl—amine reactions might be important in the lipid-protein interac- tions which, more often than not, lead to protein disinte- gration while promoting the stability of the lipid component. There is, however, no doubt that many possibilities exist for explaining the mechanism behind the stabilizing effect of protein and/or amino acids in lipid protein systems. Working on the stability of amine salts of linoleic acid, Dorlores and Lopieks (l97l) suggested that the carboxylic acid group of the fatty acid may form a salt with one of the amine groups of the basic amino acids, leaving the other amine group available to affect the double bond area of the fatty acid. If the latter interaction is a charge-transfer complex, it is a very weak one involving only partial dona- tion of electrons and little bond distortion. It would appear that the protective effect is specifically associated with the solid state. The rigid crystalline lattice of lysinium linoleate could act as a barrier to the diffusion of oxygen and oxidi- zer intermediates, and to the propagation of chain reaction. On the other hand, studies of the oxidation in solution by Koch gt al. (l97l) suggest that the salt complex also pro- longs the induction period and retards the rate of 22 autoxidation. Therefore, one might speculate that initially the solid complex physically prevents oxygen from entering the reaction. After the oxygen succeeds in attacking the organic substrate, the oxidation would be terminated by a mechanism in which the basic amino acid acts as an oxygen scavenger or a free-radical chain terminator (Dorlores and Lopieks, l97l). Role of Water in Fat Stability Water is known to retard lipid oxidation in many dehydrated and low moisture food products (Stephens and Thompson, l948). On the other hand, removing water from a food by freeze-dehydration results in a product that has a porous, sponge-like matrix. The porosity of the dehydrated food permits ready access by oxygen to the components of the food, thereby facilitating oxidative changes of food lipids. Several hypotheses have been advanced to explain the protective effect of water in retarding lipid oxidation. For example, while Halton and Fisher (l973) suggested that water has a protective effect due to retardation of oxygen diffusion, Uri (l956) postulated that water lowers the effectiveness of metal catalysts such as copper and iron. Salwin (1959) mentioned that water is attached to sites on the surface, thereby excluding oxygen from these sites. Lee (l958) indicated that water had protective effects because it promotes nonenzymatic browning, and browning can result in the formation of antioxidant compounds. Finally, 23 the formation of hydrogen bonds between water and hydro- peroxides and the consequent retardation of hydroperoxide decomposition was also thought by Lee (l958) to at least partially account for the stabilizing effect of water on lipids. Evaluation of the State of Oxidation Measurement of the state of oxidation of a lipid or a lipid-containing food is very complicated and uncertain. Measurement of peroxide value is useful to the stage at which extensive decomposition of hydroperoxides begins. Measurement of carbonyl compounds is useful provided secon- dary reactions and volatilization have not occurred to a significant extent. The pattern of oxidation may be changed by the reaction of aldehydes with amino groups from proteins, amino acids, and amino-lipids, or by the nature of the stress on the system. This, in turn, causes different carbonyl species to form. Measurement of acid-type carbonyls is not revealing if used alone. A composite of peroxide, TBA, carbonyl type, and acid determination provides perhaps the best indication of the state of oxidation; yet the data are so variable that no oxidized system has been very well characterized or defined (Dugan, l976). Link and Formo (l96l) have also indicated that although various physical methods, such as polarography and absorption spectroscopy are highly useful in elucidating reaction mechanisms and identifying products, these must normally be augmented by methods of degradation, separation, and 24 chemical analysis. Ultraviolet (UV) Absorption Method. The ultraviolet Spectrum is particularly useful in composition studies on fats and in the analysis and identification of fatty materi- als. Spectral examination has become indispensable in studying and following chemical reactions of fatty materials, especially cis-trans and position isomerization, oxidation and polymerization. The spectra of fats can indicate whether they have been mishandled, since oxidation and poly- merization products have characteristically different spectra from those of fresh materials (Swern, l964). He has also stated that unsaturated fatty materials absorb light in the ultraviolet region of the spectrum. This region is between 200-400 mu. In the case of non-conjugated and satu- rated materials, the absorption is weak and general and cannot be used for analytical purposes. In describing the application of UV absorption to oxidation studies, Mehlenbacher (l960) also indicated that the oxidation of polyunsaturated fatty acids produces perox- ides and the position of the double bond shifts to a conju- gated form. Conjugated linkages give rise to characteris- tic and intense absorption bands within the spectral range of 200-400 nm, while the absorption of isolated bouble bonds within the same region is very weak. This, he added, is the basis for the ultraviolet absorption method for determining the state of oxidation. 25 However, many workers including Rusoff _t _l. (l945), Pitt and Morton (l957) have reported ultraviolet spectral data on numerous non-conjugated compounds. In addition, when ultraviolet absorption methods indi- cate the presence of small quantities of conjugated compounds in natural fats, the results must be carefully interpreted because non-conjugated polyunsaturated components may have undergone conjugation as a result of autoxidation or other mishandling (Swern, l964). He had earlier reported (l96l) that the oxidation of polyunsaturated fatty acid is accom- panied by increased ultraviolet absorption and that the magnitude of change is not easily related to the degree of oxidation because the effects upon the various unsaturated fatty acids are different in quality and magnitude as depicted in the analysis of a sample containing dienoic, trienoic, tetraenoic and pentaenoic groups where the total diene content was due not only to linoleic acid, but also to each of the more unsaturated acids. Therefore, according to Swern, the spectral change for a given substance should be used as a relative measure of oxidation, rather than its measurement. Early workers such as Dann and Moore (l933) and Moore (l937), showed that linoleic and linolenic acids are readily isomerized to conjugated dienoic and trienoic acids respec- tively by heating them with an alcohol or ethylene glycol solution of potassium hydroxide at elevated temperatures. Mitchell, Kraybill and Szcheile (l943) applied this 26 isomerization procedure to the direct quantitative determi- nation of the linoleic and linolenic acid content of fats. In his study on the Shelf-life stability of peanut butters during long and short-term storage, Angelo _t al. (l975) found good correlation between the spectrophotometric determination of conjugated diene hydroperoxides and the peroxide value determinations, over four and twelve week periods of storage. The spectrophotometric estimation of conjugated hydroperoxides was stated to require smaller samples, to be quicker, more accurate and simpler than the method of peroxide value determination. The non-requirement of additional reagents and non-dependence upon a chemical reaction or color development were also stated as obvious advantages over the method of peroxide value determination. Vitamin A, which has a characteristic maximum at 325 mp can be determined in fats and in vitamin A concentrates by absorption in the ultraviolet region of the electromagnetic spectrum. 0,8-unsaturated carbonyl compounds can be deter- mined in autoxidized fats. Ultraviolet absorption has been used fairly extensively in studying the autoxidation of drying oils since the conjugation of polyunsaturated compo- nents parallels oxygen absorption (Swern, l964). Oxygen Uptake. From a practical standpoint, atmos- pheric oxidation of fats may be more or less sharply dif- ferentiated into that occurring in highly unsaturated oils, which is accompanied by polymerization, and is generally useful in the preparation of protective coatings (Privett, 27 l959); and that occurring in less unsaturated materials, which leads to the development of rancidity and is the source of most of the spoilage of edible fats and oils (Riemen- schneider, l955). Both types of oxidation, however, follow from the same types of reaction between oxygen and the unsaturated constituents of fats (Swern, l964). Oxygen normally exists in a triplet state but it has been established that electr0philic singlet oxygen reacts directly with olefinic molecules. Since RH and ROOH are in singlet states, singlet oxygen could react without change of spin and with conservation of energy. Conversion of oxygen to the singlet state can be accomplished by photosensitiza- tion in the presence of suitable sensitizers, such as chlorophyll, or possibly by the heme pigments myoglobin or by their derivatives. Thus, trace amounts of pigments or other sensitizers may be responsible for the initial forma- tion of hydroperoxides in some systems (Dugan, l976). The rate of oxygen uptake by an oxidizing system may be monitored by the use of the oxygen bomb test, which in- volves placement of a sample in a bomb-like device, the addition of a fixed amount of oxygen, and the monitoring of pressure at constant temperature. The time at which oxygen uptake accelerates as determined by a decrease in pressure, corresponds to the stability of the sample (Dugan, l976). Another method which involves the uptake of oxygen is the use of manometric respirometers such as the Barcroft- Warburg apparatus, and Gilson differential respirometer 28 which have automatic recording devices. The sample is added to reaction vessels and along with reference flasks are connected to manometers and agitated periodically with the temperature held constant (Gilson, l963). The progress in oxidation is monitored by pressure changes in the chamber which contains pure oxygen or air (Lundberg, l962). Cor- relations have been made between the development of perox- ides and the first presence of odors typifying rancidity (Dugan, l976). Working with beef, Tappel (l962) observed that only 50% of the oxygen absorbed during the initial oxidative deterioration was attributable to lipid peroxidation. Pos- sible oxidation of the -SH groups of the protein was sug- gested to account for the other 50% and thereby introducing errors into the estimations of the extent of oxygen uptake. The great care needed in oxygen uptake studies and its relative usefulness when large numbers of samples are involved for a relatively short period of time under closely controlled conditions have been referred to by Stuckey (l968). Literature also exists which reports measurement of oxygen uptake by weight gain studies in which weight changes are directly correlated with the uptake of oxygen. For example, Fukuzumi at al (l976) used the method to study the antioxidant effect of phenothiazine derivatives on methyl linoleate. EXPERIMENTAL Materials and Equipment Hydrolyzed Vegetable Protein: Hydrolyzed vegetable protein (HVP) was obtained from A.E. Staley Mfg. Co., Illinois. Autolyzed Yeast Protein: Autolyzed yeast protein (AYP) was also obtained from A.E. Staley Mfg. Co., Illinois. Carboxymethyl Methyl Cellulose (CMC): Carboxymethyl Cellulose (CMC) was supplied by Polyscience, Inc., Warring- ton, Pennsylvania. Palm Oil: This was obtained from A.E. Staley Mfg. Co., Decatur, Illinois. Butylated Hydroxyanisole (BHA): This antioxidant was obtained from Eastman Chemical Products, Inc., Tennessee. Chemicals: The chemicals used in this work were of analytical grade. Solvents: All the solvents with the exception of iso- octane were also of analytical grade. Spectrophotometer: A Beckman DU-24OO spectrophotometer from Beckman Instruments, Inc., Fullerton, California, was used in this work to monitor absorbance changes. Vacuum Oven: A vacuum oven, Model 29, Precision Sci- entific Instruments, Chicago, Illinois, equipped with a Cenco Hyvac-4 vacuum pump, Cenco Instrument Corporation, 29 30 Chicago, Illinois, was used for the determination of the moisture content of the freeze-dried samples. Electrical Oven: An ordinary electrical oven without any vacuum was used in the Schaal oven-test. Methods Determination of Iodine Number of Palm Oil The Hanus Method as outlined in the Official Methods of Analysis of the AOCS (l974) was employed in the iodine number determination of palm oil. Determination of Free Fatty Acid The official AOCS (l974) method was also used in the determination of the free fatty acid content of the palm oil. Preparation of Methyl Esters for Gas-Liquid Chromatography Metcalfe at al. (l966) has described a method for rapidly preparing methyl esters from lipids. 4 ml of 0.5 N methanolic NaOH was added to approxi- mately l50 mg of palm oil and the mixture heated on a steam bath until the fat globules were in solution. This was then followed by boiling the soaps with 5 ml of BF3-methanol for 2 minutes to give a quantitative conversion of the fatty acids to methyl esters. Enough saturated NaCl was added to the methyl esters, which were then readily withdrawn with a syringe. 3l Gas-Liquid Chromatography Gas liquid chromatography of the palm oil methyl esters obtained as described above was run with a F and M Scientific (Hewlett Packard) dual column, temperature programmed gas chromatograph equipped with a flame ionization detector. The columns were packed with l0% SP 2340 on Supelcoport (lOO- l20 mesh) as a solid support. The column temperature was lOOOC at 4OC/min.; detector temperature was 250°C. Inlet temperature - 200°C, chart speed was O.25”/min.; nitrogen pressure 60 psi and air, hydrogen pressure was 25 psi. Esters were identified by comparing their retention times with those of standard mixtures of known fatty acid methyl esters while peak areas were calculated by multiplying peak height by the width at half-height from which the rela- tive percentage composition of the fatty acids could be determined as shown below: % composition of fatty acid Ax = Ax + Ay + Az X ‘00 (where Ax, Ay, A2 are individual peak areas). Peroxide Value Peroxide value (PV) of the palm oil was determined by the AOCS official method, Cd 8-53, 3rd Ed. (l974). 32 Moisture Determination The moisture contents of the model systems after the freeze-drying process were determined by the procedure enu- merated in the A.O.C.S., l3.003, 9th ed. (l960). Percentage Recovery The percentage recovery of the palm oil component of the freeze-dried samples were monitored by dissolving l.6 g of the freeze-dried sample in 50 ml respectively of chloro- form and chloroformzmethanol 50:50 mixture for 24 hours. The whole mixture was then separated into a solvent frac- tion containing the dissolved palm oil and a residual fraction containing the carboxymethyl cellulose and any other undissolved component, by filtration through a Whatman filter paper number l. This extraction was made twice for each l.6 9 sample. The solvent containing the dissolved palm oil was then slowly evaporated on a steam bath under a hood leaving the palm oil. I The percentage recovery was calculated as L x lOO, y where x = weight of palm oil obtained after extraction with either chloroform alone or with chloroformzmethanol mixture (50:50). y = calculated weight of palm oil in l.6 g of freeze-dried sample. 33 Purification of ISO-octane To obtain accurate results from the absorbance measure- ments, the iso-octane used in this work had to be purified before use, using the AOCS (l974) method. Preparation of the Model Systems Each of the model systems was prepared by adding in succession to a blender cup 800 ml of distilled water, the antioxidant material (HVP, AYP, BHA), carboxymethyl-cellu- lose (CMC), and l0 9 of palm oil. l0, 20, 30, and 40% (oil weight basis) each of either AYP or HVP were used in this study while the BHA was used at levels of 0.005, 0.0l, and 0.02% (oil weight basis). In each of the model systems, the amount of CMC used was such that the CMS:palm oil ratio was lzl. A control model system in which any antioxidant material was excluded was also prepared. The entire mixture was then blended for 2 min. and thereafter transferred to a freeze-dry tray and frozen thus for 24 hours after which it was freeze-dried for another 48 hours. Weight Gain Studies For the purpose of monitoring weight gain by the Schaal oven test, l.6 g of the freeze-dried material was placed in a petri-dish (5.5” in diameter and 3/4" high) and kept in an oven at a temperature of 60°C. This was done in duplicate for each model system used. 34 Since Eubank and Gould (l942) have hinted that the arrangement of samples inside the oven is a factor affecting results, all the freeze-dried samples were kept at the same positions in the oven throughout the period of this study. Weights were taken every two days for a 27-day period in the case of the model systems containing HVP as the antioxidant material and for 43 days in the case of those containing AYP as the antioxidant material. The period for the BHA was either 27 or 43 depending on whether it was used along with HVP or AYP. Extraction of Palm Oil from Freeze-dried Materials To be able to take absorbance values for the system undergoing oxidation in the oven at 60°C, it was necessary to extract the palm oil from the system. Folch gg gl. (l957) described a method for the extrac- tion of lipid from a tissue which involves the use of chloroform-methanol (2:l, v/v). However, because no great differences as indicated by weight gain and absorbance measurements were observed when chloroform was used alone in the extraction procedure, the latter was used in subse- quent extractions. Therefore, to achieve extraction in this work, l.6 g of the freeze-dried sample undergoing oXidation in the oven was disintegrated into 50 ml of chloroform and left for 24 hours to achieve sufficient extraction. This was followed by separation of the chloroform containing the extracted 35 lipid and the undissolved residual matrix through a Whatman filter paper number l. To further separate the chloroform from the palm oil, slow evaporation under a hood was done until the characteristic pungent smell of chloroform was no longer detectable. The residue on the filter paper was again treated with 50 ml chloroform and refiltered into the original filtrate to give higher recovery of the palm oil. Thus, two washings were done for each l.6 gm of freeze-dried sample. Measurement of Absorbance Changes Having extracted the palm oil from l.6 g each of the oxidizing system as described above, the absorbance due to diene conjugation as oxidation progressed in the oven was monitored every two days by placing l0 mg of the oil, accurately weighed, into 30 ml test tube into which l0 ml of purified iso-octane (2,2,4-trimethyl pentane) was poured and mixed thoroughly in a Fisher mini shaker. The mixture was then filtered through a Whatman No. l filter paper and absorbance values taken in duplicate on a Beckman DU spec- trophotometer at 233 nm using purified iso-octane as blank. Absorbance measurements were obtained for a period of 27 and 43 days which were respectively used for the systems containing HVP and AYP as antioxidants while that for BHA was either for 27 or 43 days depending on whether it was used along with HVP or with AYP. RESULTS AND DISCUSSION The peroxide value of the palm oil used in this Study was found to be 2. This value is low enough to enable us to assume that the sample had not undergone extensive deteri- oration before use. Reservations about the use of peroxide values in measuring the state of oxidation in a lipid system exist as indicated by Labuza (l969) who maintained that in lipid-protein systems, possible interactions between the protein and lipid hydroperoxide during autoxidation can occur and thereby diminish the quantity of measured hydro- peroxide. Similarly Dugan (l976) has stated that measurement of peroxide value is useful to the stage at which extensive decomposition of hydroperoxides begins. The fresh palm oil contained no protein material and had not yet attained the stage at which extensive decomposi- tion of hydroperoxides occurs. The peroxide value of 2 also apparently reflects the well known stability of palm oil and perhaps, the processing efficiency of the supplying company, as indicated by the high oleic and palmitic acids shown in Table l, and by the low free fatty acid content and iodine value shown in Table 2. To keep the sample fresh and to maintain the low peroxide value, it was stored throughout the period of this 36 37 Table l. Fatty Acid Composition of the Palm Oil Fatty Acid Percentage Myristic acid °l4:0 Trace Palmitic acid °l6:0 47.50 Stearic acid °l8:0 4.82 Oleic acid °l8zl 37.26 Linoleic acid °l8:2 8.39 Hexadecenoic acid °l6 l 2.50 Table 2. Characteristics of the Palm Oil Value Peroxide Number 2 Free Fatty Acid 0.05% Iodine Number 43 38 study in a refrigerator at 0°C. Moisture Content: The moisture content of the freeze- dried model system was determined and found to be 3.5%. The moisture content is important because this work involved a freeze-dried model system. To be a true dry model system, the moisture content had to be as low as possible. In addition, it was reasonable to attempt to eliminate Substances or compounds extraneous to the anti- oxidants or the palm oil in the model system which might obscure an assessment of the effectiveness of the antioxi- dants used in this study. For example, as if water were an antioxidant in the conventional sense,it retards lipid oxi- dation in many dehydrated and low moisture food products according to Stephens and Thompson (l948). Finally, removing as much water as possible from a food system by freeze-drying results in a product that has a porous, sponge-like matrix, the porosity of which permits oxygen ready access to the components of the food, thereby facilitating oxidative changes. Percentage Recovery Table 3 shows the percentage lipid recovered using chloroform alone and chloroformzmethanol 50/50 v/v mixture in extracting the palm oil from the autoxidizing freeze- dried model system. The results indicate that no great differences in percentage recovery occurred between the two solvents. On the basis of this, the chloroform solvent 39 Table 3. Percentage Recovery of Palm Oil from the Freeze- Dried System Using Chloroform Alone and a Chloro- forszethanol Mixture (50:50 v/v) Time % Recovery Absor- Weight % Recovery Absor- Weight (days) of oil with bance (gm) of oil with bance (gm) CHCL3 alone (HOFSOCV 8H 1 95.821 0.035 3.521 96.103 0.034 3.521 3 96.014 0.036 3.521 96.012 0.035 3.521 5 95.498 0.034 3.521 06.001 0.036 3.522 7 95.782 0.035 3.521 95.869 0.034 3.521 9 96.116 0.035 3.521 95.921 0.034 3.521 11 96.019 0.072 3.550 96.013 0.058 3.548 13 95.994 0.084 3.598 96.011 0.080 3.596 15 96.210 0.194 3.596 95.921 0.195 3.599 17 96.022 0.233 3.589 95.764 0.212 3.599 Values = means of duplicate samples differing not more than 0.5%. 40 alone was used in the extraction of palm oil from the model system throughout the period of this study. Folch g3 g1. (1957) described a method for the extrac- tion of lipid from tissues which involved the use of chloro- form-methanol (2:1 v/v), apparently because this gave a higher percentage recovery than the use of chloroform alone. This contrasts with the findings of this study. The difference might be accounted for by the high porosity of our freeze-dried samples and their low thickness (about 1") which permitted extensive and intensive soaking of the sam- ples and allowing greater dissolution of the palm oil in the chloroform. In addition, it is possible that under our experimental conditions, interactions between the lipid and protein component of the model system and/or other compo- nents were such that the solubility of the oil in normal lipid solvent such as chloroform was not seriously altered. The 5-6% of the unrecovered palm oil may have been lost through such interactions. The same Table 3 shows that absorbance and weight gain changes were not much different when either solvent was used in extraction. The probability therefore that subsequent absorbance and weight gain changes recorded throughout this study may have been different if a chloroformzmethanol mixture had been used was ruled out. Effect of BHA on the Stability of Palm Oil One way by which the stability of palm oil in the model system was followed was by determination of weight 41 changes. Weight changes reflect changes in the rate of oxygen absorption which itself reflects the susceptibility of the system to autoxidation, other factors being constant. Table 4 Shows the percentage weight gain observed with a freeze-dried model system which contained palm oil, carboxymethyl cellulose (CMC), and distilled water, without any antioxidant and others containing various levels of the antioxidant BHA in the system. The results are in consider- able agreement with the conventional theory regarding autox- idation: an initial lag phase during which there is no remarkable gain in weight attributable to autoxidation followed by a log phase in which increased oxygen absorption leads to logarithmic increases in the rate of autoxidation attributable to a free radical mechanism. Finally there is a termination phase during which non-radical and non-reac- tive species are formed. Dugan (1976) refers to the three phases as initiation, propagation and termination, and these are depicted in Figure 1. Of interest too is the fact that even after 9 days of storage at 60°C, the palm oil in the control model system had virtually no increase in weight. Just what this means in terms of stability is shown when this is compared to the induction period of only 8 hours which Chahine g3 g1. (1974) obtained in studies with crude whale oil, stored at 40° C. Cort gg g1. (1975) obtained induction periods of 12, 15, 6 and 6 for corn, peanut, sunflower and safflower oils, respectively, when they were stored at 45°C. The 42 Table 4. Weight Gain, Increase in Absorbance and Induction Periods of Palm Oil for Various Levels of Buty- lated Hydroxyanisole in the Model Systems System System System Control with with with Model 0.005% 0.01% 0.02% System BHA BHA BHA Weight Gain in 11.13 8.13 5.87 0.75 27 days (%) Induction Period 11 17 21 - by Weight Gain (DayS) Increase in Absor- 0.252 0.133 0.094 0.008 bance in 27 Days Induction Period 11 17 21 - by Increase in Absorbance (Days) - No detectable end of induction period within the 27-day test period. Values = means of duplicate samples differing not more than 0.5% 43 .mxau AN Lot oooo pa nmeoum .F_o EFma c:m.o:u .<:m we mamumxm Faves cw mmmcmgu pnmwmz .F mezmwd om pm o_ P 0 .III .Iaul III: < I .H...HIJ'IIIA.HII ;HI.IIH,.L1H1..W1|.1.1. O Qlul 0.111 s « C‘L al.}, O < 0 . oo.op . l T : .uzu mo mewpmzm Fence cowe616~mmce cw cwmm agave: .N mezmvu r1_ 0 a I , r1 r1 H I M 3 r1 .m. a>z fioq + z xoq + Foeucou u H r1 ..m Mu u r1 m. a>z gem + z aom + Poeocou n w z, a>z New + z aom + _0e0e08 n m m m a): No_ + «rm amoo.o + _0epeou u a fir _ L a>= NOP + Foeucou 0 u u 3 op (rm Nmoo.c + Poepeou u m < Poeucou n < 63 .mzmn mm to» uoom um umcoum Fwo EPII new : .uzu we Isoumxm Pwnos coweuumNmmem cw :wmm “Iowa: .m mesme I o I. 1I1 w E I..— All H. II III + III ISI + 2:28 1 I E l M. I>I III + _0IIIOI u I I I ..I m. III III + III Ipo.o + _0III0I u I Ir; mm I>I III + _0III0I u I I mm III III + III I_o.o + Poepcou u I fiL Q>I &0N + Fogpcou N m m III II_ + III I_o.o + Peepeoo u I I>I III + _0Ipeou n I . rI ..I_ III Iro.o + _0II:0I u I I Focucou u < I I 64 .mIII II III IoII In vmgoum Iwo EIII new I .uzo Io msmpmzm Imuoe cmwgcumNmmgw :_ IIIm Icm_m3 .I mezme If. rim III H1. F1_ I I I I M III III + III III + 2:28 u I I w... III III + Ioeueou u I I .FI MW III III + III III.I + FoepeoI u I m. (%) I>I Nom + Pogpcou n o I>x ION + : .uzu Io IEIuIII Pmuoe cw mmmIqucI mucIIIomII .m IIIIII : c I I ”a. Im. .0 I q 9 ,U .3 f 3 rl I I IL . u III III + III IIII.I + IIIIIII " I I :III.I.I .aa IL 9 III III + IIIIIII u I I rL .m III III + III IIII.I + IIIIIII u I rm I h: H_m I III III + IIIIIII u I m” It III III + III IIII.I + IIIIIII u I .u Fl w III III + IIIIIII u I I III III + III IIII.I + IIIIIII " I IIII.I III III + IIIIIII u I III IIII.I + IIIIIII u I Foggcou u < r1 I 69 .IIII II III IIII II IIIIII III EIII III III .I>: .uzu Io msmumzm PIIoE IIIII-IIIIII II IIIIIIIII IIIIIIIIII .m IIIIII III - I - III.I LI: I I . .L I L III III + III III.I + IIIIIII u I I I III.I III III + IIIIIII u I III III + III III.I + IIIIIII u I m; .III III III IIIIIII - I III III + III III.I IIIIIII III III IIIIIII III.I III III + III III.I IIIIIII III III IIIIIII III III.I IIIIIII IoIpIou mu £53 19 aSPBJDUI aoueqxosqv 7O .IIII II III IIII II IIIIII III EIII III III .I>I .uzu Io mamamxm Imvoe cmmgu-mImmII :I mmmImIUII mocmogomn< .oI IIIIII , I III.I IL! E E I I a I C E H wu sea 19 aseaxaul aouequosqv III III + III III.I + IIIIIII u I I LIII.I III III + IIIIIII u I r1, III III + III III.I + IIIIIII n I I III III + IIIIIII u I III III + III III.I + IIIIIII u I III III + IIIIIII u I I bom.o L I>I No— + I Nor + Iogpcou u u < Iom I>I Iom III Iop + + .IIII II III IIII II IIIIII III EIII III III .III .III II msmumxm Imnoe IIIII-IIIIII II IIIIIIUII moIIIIoIII .op IIIIII III Imoo.o I>I IOI III Imoo.o a>< Ion III Imoo.o I>< ION III Imoo.o I>< Iofi III Imoo.o :: IIIpcou n IIIucou u Iqucou n IIIucou u IIIpcou u IIIIIII u IIIucou u IoIpcou u IIIucou u Iqucou u CD C) ooo.o voo_.o .com.o mu £92 19 aSESJDUI aaueqxosqv 72 EIII III III .III .uzu Io mEmpmam PmIoE ImIIIsINIIII cw mmmImIIII mucmnIomII I>I I>I I>I I>I IOI Iom ION III 4. .IIII II III IIII II IIIIII III .II IIIIII III Ipo.o I>I IOI III III.I I>I Iom III Ipo.o I>< ION III Ipo.o I>< IoI III Ipo.o C: IIIIIII u IIIIIII u IIIIIII u Iqucou n IIIIIII u IoIpIou u IoIpcoo n IIIIIII u Iqucou u IIIucou u ooo.o LtOOFIO :oom.o mu £33 19 aseaJoul BOUQQJOSQV 73 I>I I>I I>I I>I III III ION Iop .mzmv II Low uooI In umgoum pwo spam Ucm III .I>I .Izu Io msmpmzm quoe Imwgn-w~mmLI cw mmmmmgucw mucmngomnI .IP mgzme rWu F— JH. a!“ me III III I - ‘ I + I+:I.Imo;o Fogucou 1.oop.o I>I III Pogucou + III .Imo.o Fogucou I>I Iom Fogucou + IIMI III.I Pogucou I>I III Fogucou + III .Imo.o Fogucou .roow.o I>I Io? Fogucou III III.I Fogucou Pogpcou mu £93 19 SSPBJDUI SDUPQJOSQV 74 Comparison of the Antioxidant Effects of BHA, HVP and APY in Palm Oil A look at all the results recorded in this work and already referred to in the discussion clearly indicate that BHA was more effective in providing improved stability to the palm oil components in the model system than AYP which in turn was more effective than HVP. There are numerous reports which indicate that given the same experimental con- ditions all antioxidants are not equally effective. BHT or BHA were reported by Cort _t__l, (l975) to be inferior to TBHQ in improving the stability of a lipid system. Sherwin and Luckadoo (l970) made a similar observation. As a more striking contrast, Marcuse (l960) working with herring oil observed that while the amino acid histidine had an anti- oxidant effect, cysteine was a pro-oxidant under the same conditions of activity. The observation in this study indicating differential activity among HVP, AYP and BHA is therefore in agreement with other works. Since the working conditions in these studies were maintained the same as much as possible throughout the period of experimentation, any differences in activity among the antioxidant BHA, AYP and HPV can only be accounted for by the intrinsic differences among them and/or dif- ferences in their preparations. First BHA is a synthetic antioxidant specifically prepared to provide improved stability in lipid systems prone to autoxidation while AYP and HVP are regarded as natural products or products of 75 natural products which, in any case, are beginning to receive more serious attention as antioxidants. The greater effectiveness of AYP over HVP has been postulated by Bishov and Henick (l972) to be due to the milder conditions of autolysis of AYP compared to the hydrolytic conditions involved in the preparation of HVP - conditions which are regarded as more severe and therefore more destructive to the functionality of the HVP. The salt which is produced by the neutralization of the HVP during its preparation may also be pro-oxidative and therefore lower the antioxidant effectiveness of the HVP. Even with similar conditions of preparation one would reasonably expect differences in effectiveness as antioxidants attri- butable to differences in their composition such as in amino acids, which have been reported by many workers to have antioxidant effects (Marcuse, l960, 1962, Karel _£__l,, l966, Smith and Circle, l972, Scarborough and Watts, l949) and in peptide profile (Bishov and Henick, l972, l975). According to Dugan (l962) some antioxidants are effective in prolonging the keeping time of both fats and oils and the foods containing them and he described such antioxidants as "carry through antioxidants" since they "carry through" or survive the thermal stresses, steam distillation and pH effects of processing to give longer shelf-life to the finished food. The procedure which was used in this study to prepare the model system which in- volved mixing the CMC, HVP or AYP, palm oil and BHA in 76 distilled water by mechanical agitation in a blender, fol— lowed by freezing for 24 hours and by freeze-drying for another 48 hours and finally holding at 60°C for periods of 27 and 43 days could be assumed to amount to food pro- cessing. In view of this, the improved stability observed with the BHA and HVP and AYP after all these stresses is tantamount to "carry through" properties for these com- pounds. It might also be stated here that differences in effectiveness reported in this study among the above three antioxidants may have been due, among other things, to the differences in extent to which the stress conditions of processing employed in the study altered their functionali- ties. Comparison of Weight Gain Measurements to Absorbance Values In measuring the extent of autoxidation in a lipid or in estimating its state of autoxidation it is the usual practice to use more than one method for the purpose of comparison which can inspire more confidence in the results obtained. This is because the whole concept of oxidation or autoxidation is such a complex thing. In this study, therefore, the absorbance due to diene conjugation and weight gain studies attributable to oxygen uptake were used to measure oxidation. Labuza _£.il- (l969) reported that after 24 hours peroxide value determinations became lO-ZO times less than the expected values and therefore suggested that the value is not a useful method for the measurement of oxidation in 77 foods containing protein, especially under accelerated conditions. Angelo _t _l. (l975) noted that the mechanism which causes the peroxidation of polyunsaturated fatty acids pro- duces conjugated diene hydroperoxides and that both peroxide value and diene conjugation measure the same products of oxidation which are hydroperoxides. On account of this, the measurement of diene conjugation by spectrophotometric measurement of absorbance at 233 nm as was done in this study did eliminate the need to take peroxide value measure- ment. The tables and figures already referred to in this discussion so far indicate that, in general, the weight gain values were in considerable agreement with the absor- bance values which indicate diene conjugation. Although Swern (l96l) has noted that the diene conjugation as measured by absorption of ultraviolet radiation at 233 nm was in agreement with oxygen uptake (weight gain measure- ment) only in the initial stages of oxidation, in oils containing linoleate or more highly unsaturated systems, the results here show agreement at all stages of oxidation, to the extent involved in these studies. The suggestion of Corliss (l968) that although UV measurements of conjugated dienes do not indicate the precise stage of lipid oxidation and that these can be used in combination with quantitative determinations such as weight gain measurements hold correct in these studies. 78 Any specific differences which are observable from the results such as in the specific lengths of the induction periods, the maximum weight gain values and absorbance values and in the final weight and absorbance, might be accounted for by differences in the precision of both methods and certain physical phenomena one of which could have been oxygenation. Oxygenation makes it possible to have physical absorption of oxygen by a system and therefore increase in weight without a corresponding increase in UV absorption to parallel increased oxygen uptake as one would expect from oxidation (Campbell gt al., l974). It is, however, likely that the phenomenon was too subtle to measure for this study. SUMMARY AND CONCLUSION This work was done with the objective of probing the antioxidant effects of Butylated Hydroxyanisole (BHA), Hydrolyzed Vegetable Protein (HVP) and Autolyzed Yeast Protein (AYP) on palm oil in a model syste, composed of either HVP or AYP, BHA, CMC and palm oil. Results indicated a) that BHA, HVP and AYP had antioxidative effects on palm oil and therefore improved the stability of the palm oil in the model system; that such improved stability was directly related to the concentration of each of these antioxidant compounds in the system; that the order of effectiveness was BHA > AYP > HVP; that combinations of the antioxidants gave greater stability but combinations of BHA and HVP gave only negative synergism which was also found to be concentration dependent; good correlation between the weight gain measure- ments and absorbance due to diene conjugation in monitoring oxidation of palm oil in the model system. 79 BIBLIOGRAPHY Angelo, A.S., Ory, R.L. and Brown, L.E. l975. Comparison of methods for determining peroxides in processed whole peanut products. J. Am. Oil Chem. Soc. ggz34. A.O.C.S. l974. Official Method of the American Oil Chemists r Society, Chicago, Illinois. Austudillo, M.D., Funes, E. and Sanz, F. 1968. Effects of radiation on olive oil. An. Quin. ggz787. Baddings, H.T. 1960. Principles of autoxidation processes in lipids with special regard to the development of autoxidation off—flavors. Neth. Milk and Dairy J. L l4:2l6. Bishov, S.J., Masouka, Y. and Henick, A.S. l967. Fat quality and stability in dehydrated proteinaceous food mixes. Food Tech. glzl48. Bishov, S.J., Henick, A.S., Giffee, J.w., Nii, I.T. and Press, P.A. l97l. Quality and stability of some freeze-dried foods in ”zero" oxygen headspace. J. Fd. Sci. ;§:532. Bishov, S.J. and Henick, S.A. l972. Antioxidant effect of protein hydrolyzates in a freeze-dried model system. J. Fd. Sci. glz873. Bishov, S.J. and Henick, A.S. l975. Antioxidant effect of protein hydrolyzates in freeze-dried model systems. Synergistic action with a series of phenolic anti- oxidants. J. Fd. Sci. 51:345-347. Blanck, F.C. l955. In Handbook of Food and Agriculture. Reinhold Publ. Co., N.Y. Bolland, A.C. and Koch, S.D. l945. Rearrangement of double bonds during autoxidation. J. Chem. Soc. 1§:54-547. Braddock, R.J. and Dugan, L.R. l973. Reaction of autoxi- dizing linoleate with coho salmon myosin. J. Am. Oil Chem. Soc. ng343. 80 8l Campbell, I M., Caton, R.B. and Crozier, D.N. l974. Com- plex formation and reversible oxygenation of free fatty acids. Lipids 9:9l6. Chahine, M.H. and MacNeill, R.F. l974. Effect of stabili- zation of crude whale oil with TBHQ and other antioxi- dants upon keeping quality of resultant deodorized oil. A feasibility study. J. Am. Oil Chem. Soc. 61: 37. . Chapman, D. l969. Physical studies of lipid-lipid and lipid-protein interactions. Lipids 3:25l. Corliss, G.A. l968. Phospholipid oxidation in emulsions. Ph.D. Thesis, Mich. State Univ., East Lansing, MI. Cort, w.M. l974. Antioxidant activity of tocopherols, ascorbyl palmitate and ascorbic acid. J. Am. Oil Chem. Soc. glz32l. Cort, w.M. l974. Hemoglobin peroxidation test screen antioxidants. Food Tech. Oct; 36:60. Cort, W.M., Scott, H.J., Araufo, M. and Mergens, w.J. l975. Antioxidant activity and stability of 6-hydroxy- 2,5,7,8-tetramethyl chroman-Z-carboxylic acid. J. Am. Oil Chem. Soc. gzl74. Dahle, C.E. and Nelson, D.H. l94l. Antioxigenic fractions of oat and soybean flour. J. Dairy Sci. 25:29. Dann, J.N. and Moore, T. l933. The absorption spectra of the mixed fatty acids from cod-liver oil. Biochem. J.‘21:ll66. Dorlores, V. and Lopieks, S.D. l97l. Basis of stability of amine salts of linoleic acid. 2. Structure-property correlation on lysinium linoleate. J. Fd. Sci. 36: 482. Dugan, L.R. and Kraybill. l954. AMIF Bulletin #l8. Dugan, L.R. and Kraybill. l956. Tocopherols as carry- through antioxidants. J. Am. Oil Chem. Soc. 31:525. Dugan, L.R. l96l. Development and inhibition of oxidative rancidity of foods. Food Tech. lile. Dugan, L.R. 1962. In Encyclopedia of Chem. Tech. Ed. Kirk, R.E. and Othmer, D.F. Vol. 2. Intersc. Ency. N.Y. 82 Dugan, L.R. l976. In Principles of Food Science. Part I. Food Chemistry. Ed. Fennema, O.R. Marcel Dekker Inc., N.Y. Eubank, F.C. and Gould, I.A. l942. The oven and aeration methods as means of accelerating fat oxidation. Oil and Soap. 13:205. Farmer, E.H. and Sutton, D.A. l943. The course of autoxi- dation reactions in polyisoprenes and allied comd pounds. VII. Rearrangement of double bond during autoxidation. J. Chem. SOC-.llii547- Folch, J., Lees, M. and Sloane-Stanley, G.H. l957. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226:497. Fukuzumi, K., Ikeda, N. and Egawa, M. l976. Phenothiazine derivatives as new antioxidants for the autoxidation of methyl linoleate and their reaction. J. Am. Oil Chem. Soc. II:623. Gilson, E.W. l963. Differential respirometer of simplified and improved design. Sc. l4l:53l. Gurd, R.N. 1960. Association of lipids with proteins. In Lipid Chemistry. Ed. Hanahan, D.J. John Wiley and Sons, Inc., N.Y. Halton, P. and Fisher, E.A. l973. Studies on the storage of wheaten flour: II. The absorption of oxygen by flour when stored under various conditions. Cereal Chem. II:267. Hayes, R.E., Bookwalter, G.N. and Bagley, E.B. l977. Anti- oxidant activity of soybean flour and derivatives - A Review. J. Fd. Sci II:1527. Jarenback, L. and Liljemark, A. 1975. Effects of linoleic and linoleic acid hydroperoxides on myofibrillar pro- tein. J. Food Tech. Igz437. Karel, M., Tannenbaum, S.R., Wallace, D.H. and Maloney, H. l966. Autoxidation of methyl linoleate in freeze- dried model systems. Effects of added amino acids. J. Fd. Sci. IIz892. Karel, M., Schaich, K., Roy, R.B. 1975. Interaction of methyl linoleate with some proteins and amino acids. J. Agr. Food Chem. I;:l59. 83 Kimoto, W. and Gaddis, A.M. l970. Precursors of alk-2,4- dienals in autoxidized lard. J. Am. Oil Chem. Soc. IQ403. Koch, R.B. l962. Dehydrated foods and mOdel systems. In Symposium on Food Lipids and their Oxidation. Ed. Schultz, H.W., Day, E.A. and Sinnhuber, R.O. AVI Publ. Co., Inc., Westport, CT. Koch, S.D., Hyatt, A.A. and Lopieks, D.V. l97l. Basis of Stability of amine salts of linoleic acid. I. Gen- erality of the oxidation. Protection and effect of physical state. J. Fd. Sci. £91479- Labuza, T.P., Tsuyuki, H. and Karel, M. l969. Kinetics of linoleate oxidation in model systems. J. Am. Oil Chem. Soc. nnz409. Lea, C.H. l962. The oxidative deterioration of food lipids In Lipids and their Oxidation. Ed. Schultz. AVI Publ. Co., Inc., Westport, CT. Link, W.E. and Formo, M.W. l961. Analysis of autoxidative mixtures In Autoxidation and Antioxidants. Vol. l Ed. Lundberg, W.O. Intersc. Publ., N.Y. Lips, H.J., Marshall, J.B., White, W.H. and Grant, G.A. l949. Ration biscuits. 5. Effects of kind and concen- yration of various constituents on keeping quality. Can. J. Research 274:4l9. Lundberg, W.O. l962. Oxidative rancidity in food fats and its prevention. In Autoxidation and Antioxidants. Vol. II. Ed. Lundberg, W.0. Intersc. Publ., N.Y. Lundberg, W.O. l962. MechaniSms. In: Symposium on Food Lipids - their oxidation. Ed. Schultz, H.W., Day, E.A. and Sinnhuber, R.O. AVI Publ. Co., Westport, CT. Mabrouk, A.F. and Dugan, L.R. 1960. A kinetic study of methyl linoleate and linoleic acid emulsions in the presence of sodium chloride. J. Am. Oil Chem. Soc. Igz486. Mahon, J.H., and Chapman, R.A. 1953. The relative rates of destruction of propyl gallate and butylated hydroxy- anisole in oxidizing lard. J. Am. Oil Chem. Soc. In: 34. Marcuse, R. 1960. Antioxidative effects of some amino acids. Nature l86:886. 84 Marcuse, R. l962. The effect of some amino acids on oxi- ’ dation of linoleic acid and its methyl esters. J. Am. Oil Chem. Soc. Igz97. Mehlenbacher, V.C. 1960. The analysis of fats and oils. The Garrard Press Publ., Illinois. Metcalfe, L.D., Schmitz, A.A. and Petka, J.R. l966. Rapid preparation of fatty acid esters for gas chromatogra- phic analysis. Analytical Chem ;§:5l4. Mitchell, H.J., Kraybill, H.R. and Zscheile, F.P. l943. Quantitative spectral analysis of fats. Ind. Eng. Chem. (Anal. edition) Igzl-3. Moore, T. l937. Changes in the absorption spectra of mixed fatty acids from cod-liver oil. Biochem. J. II:ll66. Narayan, K.A. and Kummerow, F.A. l958. Oxidized fatty acid protein complexes. J. Am. Oil Chem. Soc. §§:52. Neil, J. and Page, L. l956. The effects of antioxidants on frozen ground pork. Food Tech. 19:315. Olcott, H.S. and Mattill, H.A. l936. Antioxidants and the oxidation of fats. 4. Lecithin as an antioxidant. Oil and Soap (Chicago) Igz98. Overman, A. l947. Antioxidant effect of soybean flour in frozen pastry. Food Res. Igz365. Pitt, G.A.J. and Morton, R.A. l957. In Progress in the Chem. of Fats and other Lipids. Vol. 4., Chapt. 9, Pergamon, N.Y. Privett, 0.5. l959. Autoxidation and Autoxidation poly- merization. J. Am. Oil Chem. Soc. Inz507-512. Riemenschneider, R.W. l955. I Handbook of Food and —-— Agric. Chapt. 8, Reinhold, N.Y. Roubal, W.T. and Tappel, A.L. l966a. Damage to proteins, enzymes and amino acids by peroxidizing lipids. Arch. Biochem. Biophys. ll3:5. Roubal, W.T. and Tappel, A.L. l966b. Polymerization of proteins induced by free radical lipid peroxidation. Arch. Biochem. Biophys. ll3:l50. Rusoff, I.I , Klevers, J.R. and Holman, H.B. l945. Extreme UV absorption spectra of the fatty acids. J. Am. Chem. Soc. glz673-678. 85 Salwin, H. l959. Defining minimum moisture content for dehydrated foods. J. Fd. Sci. n§:ll6. Scarborough, D.A. and Watts, B.M. l949. The pro-oxidant effect of ascorbic acid and cysteine in aqueous fat systems. Food Tech. IgzlSZ. Schwenke, K.D. l975. Modifications of proteins by reaction with carbonyl compounds. Nahrung Igz92l. Shelton, J.R. l959. Mechanism of antioxidant action in the stabilization of hydrocarbon systems. J. Appl. Poultry Sc: Iz345. Sherwin, E.R. and Luckadoo, B.M. l970. Studies on anti- oxidant treatments of crude vegetable oils. J. Am. Oil Chem. Soc. II:l9. Smith, J.S. and Dunkley, N.L. l962. Inhibition of lipid peroxidation by a reduced metal ion. Arch. Biochem. Biophys. g§:46. Smith, A.K. and Circle, S.J. l972a. Protein products as food ingredients. In Soybeans: Chemistry and Tech- nology. Ed: Smith, A.K. and Circle, S.J. p. 339, AVI Publ. Co., Westport, CT. Stephens, H.H. and Thompson, J.B. l948. The effect of shortening stability on commercially produced army ration biscuits. II. Development of oxidation during storage. J. Am. Oil Chem. Soc. g§:389. Swern, D. l96l. Primary products of olefinic autoxidation. In Autoxidation and Antioxidants. Vol. l. Ed: Lundberg W.O. Intersc. Publ., N.Y. Swern, D. l964. In Baileys Industrial Oil and Fat Products. 3rd Ed. p. l 2. Sylvester, N.D., Lampitt, L.H. and Ainsworth, A.N. l942. Determination of the stability of oils and fats. J. Soc. Chem. Ind. (London) QI:l65. Tappel, A.L. l953. Oxidative fat rancidity in food products. I. Linoleic oxidation catalyzed by hemoglobin and cytochrome c. Food Res. I§:560. Tappel, A.L. l955. Studies of the mechanism of vitamin E action. 111. In vitro copolymerization of oxidized fats with proteins. Arch. Biochem. Biophys. II:266. Tappel, A.L. l955. Unsaturated lipid oxidation catalyzed by hematin compounds. J. Biol. Chem. 2l7:72l. Tappel, A.L. l962. .L. Lipids and their Oxidation. p. 426. Ed. Schultz, H.W., Day, E.A. and Sinnhuber, R.O. AVI Publ. Co., Inc., Westport, CT. Tarladgis, B.G., Pearson, A.M. and Dugan, L.R. l962. The chemistry of the 2-thiobarbituric acid test for the determination of oxidative rancidity in foods. J. Am. Oil Chem. Soc. I9:34. Tjho, K.H. and Karel, M. l969. Autoxidation of methyl linoleate in freeze-dried model systems. 4. Effects of metal and of histidine in the absence of water. J. Fd. Sci. §£3540- Togashi, H.J., Henick, A.S. and Koch, R.B. l96l. The oxi- dation of lipids in thin films. J. Fd. Sc. g§:l86. Tumerman, L. and Webb, B.H. l965. Coagulation of milk and protein denaturation. In Fundamentals of Dairy Chemistry. Ed. Webb, B.H. and Johnson, A.W. AVI Publ. Co., Inc., Westport, CT. Uri, N. l956. Metal ion catalysis and polarity of environ- ment in the aerobic oxidation of unsaturated fatty acids. Nature l77:ll77. Uri, N. l96l. Metal catalysis. In Autoxidation and Anti- oxidants. Ed. Lundberg, W.O. Intersc. Publ., Wiley and Sons, Inc., N.Y. Vol. l, p. 33. Walz, E. l93l. Isoflavone and saponin glucosides in soya hispida. Ann. 489:ll8 (in Chem. Abstract, l93l, 24 (2l):4574). Waters, W.A. l97l. The kinetics and mechanism of metal catalyzed autoxidation. In Symposium: "Metal Cata- lyzed lipid oxidation" presented at ISF-AOCS World Converence, Chicago, Illinois. Zirlin, H. and Karel, M. l969. Oxidation effects in freeze- dried gelatin methyl linoleate model system. J. Fd. Sci. IgzlGO.