THE EFFICACY OF THE DYE BINDING TEOIINIOUE FOR THE DETERMINATION OE THE PROTEIN CONTENT OF SAUSAGE EMULSIONS Thesis for the Degree of M. S. ' MICHIGAN STATE UNIVERSITY ‘ GEORGE I. SERERIOII . 1972 .. ... . _... ,5: :3; .iifiiii ,_.._,..~..m§ . E .. a h m ABSTRACT THE EFFICACY OF THE DYE BINDING TECHNIQUE FOR THE DETERMINATION OF THE PROTEIN CONTENT OF SAUSAGE EMULSIONS by George J. Seperich With the advent of new federal legislation regulating processed meat products, greater control of the product composition by the manu- facturer has become necessary. The high speed methods for preparing processed meats necessitate the use of rapid forms of analysis for the individual components. Therefore, the dye binding procedure was tested for its ability to determine the protein content of a processed meat product. This study involved the use of four dyes initially, Acid Orange II (AC-7), Brilliant Orange (AC-12), Orange G (AD-10) and Alizarin Cyanine Green (AG-25) but upon examination of a few basic dye parameters only two dyes were carried through the entire study. These dyes were Brilliant Orange (AG-12) and Orange G (AO-lO). These dyes were exposed to the major individual protein fractions of a muscle, the myofibrillar, sarcoplasmic and stroma proteins. They were also tested in a model emulsion system using the myofibillar and sarcoplasmic proteins with a commercial soybean and cottonseed oil (Wesson). And finally their ability to bind to the proteins present in a sausage emulsion, pro- cessed (smoked, cooked and chilled) and raw, was ascertained. The results of exposing Brilliant Orange (AG-12) and Orange G (AO-lO) to the individual protein fractions indicate that AO-12 is capable of binding in all three fractions. The dye AO-lO bound to these proteins but to a lesser extent. A greater amount of AO-12 was bound by the myofibrillar proteins than by the sarcoplasmic or stroma proteins. George J. Seperich This relationship was not apparent for AO-lO. It bound the myofibrillar and sarcoplasmic proteins in similar quantities but in a greater amount than to stroma proteins. The model system demonstrated that for both dyes the dye binding capacity, (DBC) was greater in the emulsion than in aqueous preparation of the myofibrillar and sarcoplasmic proteins. However, further tests disclosed that this increase in BBC was not attributable to the oil itself, since the DEC remained constant for emulsions of constant protein but increasing amounts of oil. Another observation derived from this model system was that as the protein concentration increased the DEC appeared to decrease and that this relationship was not linear but curvilinear. A similar observation was made in the sausage emulsion. This curvilinear relationship was found for both dyes, but was more apparent for AO-12 than for AO-lO. In the sausage emulsion, it was found that the unprocessed or raw emulsion bound more dye than the processed or cooked emulsion. However, the processed emulsion invariably had a higher protein content than the unprocessed emulsion. When individual batches of sausage emulsion were compared among themselves, a nearly linear relationship existed between the BBC and the protein concentration for both dyes with both raw and cooked emulsions. However, if the entire range of the protein concen- tration was considered, it was apparent that the curvilinear relationship encountered in the model systemwasIalso present in the sausage emulsion. This relationship existed for both dyes. The unprocessed emulsion occupies the upper portions of the curve and the processed emulsions are in the smlsi to con lat S) Show I George J. Seperich in the lower portion of the curve. This was true for all three sausage emulsions, 24%, 30% and 36% fat, repeated twice. From the above described curvilinear relationship, it was possible to compute a regression equation describing the dye binding in a particu- lar system for each dye. Previous investigators generally attempted to show the linearity of the relationship of dye binding parameters and concentration of protein, but these equations illustrate that the rela- tionship is more complex or curvilinear. THE EFFICACY OF THE DYE BINDING TECHNIQUE FOR THE DETERMINATION OF THE PROTEIN CONTENT OF SAUSAGE EMULSIONS By George J. Seperich 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 1972 n; bud! es appreci jective Beatric §arati: the 0t} aC'filow" ACKNOWLEDGEMENTS The author wishes to express his appreciation to Dr. James F. Price for hiSTpatience, encouragement and guidance throughout this study. The thorough efforts of the examining committee, Drs. Charles E. Cress, Charles M. Stine, Robert A. Merkel and Lawrence E. Dawson, were also appreciated. Mrs. Mildred E. Spooner is thanked for her encouragement and ob- jective comments on this study. The author is very grateful to Mrs. Beatrice Eichelberger for her efforts in the typing and physical pre- paration of this manuscript. The support and unsolicited opinions of the other Meat Laboratory graduate students concerning this project are acknowledged and appreciated. To my wife a special thank you for her encouragement and always solicited assistance in preparing this manuscript. Finally, the author udshes to express his appreciation to his parents and sister whose per- sistent, "When are you going to finish?" has served and will continue to serve as an inspiration in my academic endeavors. ii TABLE OF CONTENTS INTRODUCTION 0 O O O O O I O O O O O O O O O O O O O O O O I O O O 1 LITERATURE REVIEW . . . . . . . . . . . . . . . . . . . . . . . . 3 General Theory of Dye Binding . . . . . . . . . . . . . . . . 3 Application of Dye Binding Technique: Dairy Chemistry . Application of Dye Binding Technique: Cereal Chemistry Application of Dye Binding Technique: Meat Products . . 1 WOO‘I Nature of Sausage or Meat Emulsion Products . . . . . . . . . l6 PartiCIe Size 0 O O O O O O O O O O O O O O O O O O O 0 l7 Membrane Formation . . . . . . . . . . . . . . . . . . . l7 Membrane Proteins . . . . . . . . . . . . . . . . . . . 18 Basic Amino Acids Available in Protein Membranes . . . . 19 Effects of Non-Protein Components Upon Membrane Formation 20 Source of Protein Membrane . . . . . . . . . . . . . . . 21 METHODS AND MATERIALS 0 O O O O O O O O O O O O O O O O O O O O O 24 Meat Preparation for Sausage Emulsions . . . . . . . . . . . 24 Sausage Preparation . . . . . . . . . . . . . . . . . . . . . 25 Preparation of a Model System Emulsion . . . . . . . . . . . 26 Protein Extraction Procedure . . . . . . . . . . . . . . . . 27 Dye Binding Procedures . . . . . . . . . . . . . . . . . . . 28 K1 eldahl Protein Ar‘alys18 I I O O O O O O O O O O I O O O O I 29 Moisture Determination . . . . . . . . . . . . . . . . . . . 30 Fat Determination . . . . . . . . . . . . . . . . . . . . . . 31 Disc Gel Electrophoresis . . . . . . . . . . . . . . . . . . 31 Statistical Analysis . . . . . . . . . . . . . . . . . . . . 32 iii q—flvn I ALL... 5" Page RESULTS AND DISCUSS ION O O O O O O O O C O O O O O O O O O O I O O 3 3 Dye Parameters O O O O O O O O O O O O O O O O O O C O 0 O 0 33 pH 0 C C O O C O C O O C C O I 0 O O I O C O O O O O O O 34 Absorbancy Index (Extinction Coefficient) . . . . . . . 37 Retention by Filter Paper . . . . . . . . . . . . . . . 38 Dye Binding and Protein Systems . . . . . . . . . . . . . . . 39 Hammerstein's Casein . . . . . . . . . . . . . . . . . . 39 Bov1ne serm Albumin (BSA) 0 O O O O O O O C O O O O C O 41 Meat-Lipid system I O O O O O O 0 O 0 O O O I O O O O O O O O 46 Meat and Protein Component System . . . . . . . . . . . . . . 49 Protein Component System . . . . . . . . . . . . . . . . 49 Effect of Anion Upon Dye Binding . . . . . . . . . . . 52 Effect of Dye Binding on Disc Gel Electrophoresis . . . 52 Model Emulsion System . . . . . . . . . . . . . . . . . . . . 54 Effect of Lipids on Dye Binding . . . . . . . . . . . . 56 MYOfibrillar Proteins . O O O O O O O I O O O O O O O O 58 Sarcoplasmic Proteins . . . . . . . . . . . . . . . . . 6O Sausage Emulsions and Dye Binding . . . . . . . . . . . . . . 63 Dilution of the Sample . . . . . . . . . . . . . . . . . 63 DBC and Sausage Emulsions . . . . . . . . . . . . . . . 65 Regression Equations . . . . . . . . . . . . . . . . . . 69 SWY MD CONCLUSIONS 0 O O O O O O I O O O I O O O C O O O O O 75 BIBLIOGMIIIY O O O O O O O O O O O O O O O O O O O O I O O O O O 0 79 APPENDIX 0 O O O O O I O O O O O O O O O O O O O O O O O O O O O O 87 iv LIST OF TABLES Table Page 1 A comparison of absorbancy index values (extinction coefficients) determined in separate studies by three investigators I I I I I I I I I I I I I I I I I I I I I I I 37 2 Percent retention by filter paper (Whatman No. 40) of azo and anthroquinone dyes at pH 1.9 and 3.0 . . . . . . . . . 38 3 The effect of muscle protein type upon the DBC (Hg dye bound/mg protein) of AO-lO and AO-12 . . . . . . . . . . . 51 4 The effect of varying the lipid concentration upon the DBC of myofibrillar beef proteins for two azo dyes at two con- centrations, pH 5.5 in a 9% protein solution . . . . . . . 57 5 The effect of sample dilution upon the DBC of proteins in a sausage emulsion (unprocessed) . . . . . . . . . . . . . 64 6 Comparison of correlation coefficients for four dye parameters of AO-12 on the protein concentration of three different protein systems . . . . . . . . . . . . . . . . . 72 7 Comparison of correlation coefficients for four dye parameters of A0-12 on the protein concentration of three different protein systems . . . . . . . . . . . . . . . . . 72 Figure \J 10 ll 12 13 Figure 10 ll 12 13 LIST OF FIGURES Molecular structure of azo and anthroquinone dyes selected for protein determination study . . . . . . . . . . . . . Adherence of dyes to Beer's law at respective wave lengths, pH 1 o 9 ’ Citrate-phosphate bUffer o o o o o o o o o o o o o Adherence of dyes to Beer's law at respective wave lengths, pH 3.0, citrate-phosphate buffer . . . . . . . . . . . . . The effect of two different dye concentrations upon the free dye concentration of four dyes in Hammerstein's casein I I I I I I I I I I I I I I I I I I I I I I I I I I Comparison of the number of dye molecules bound per protein molecule (BSA), pH 3.0 for all four dyes . . . . . . . . . Log-log plot of protein concentration vs 5, average number of dye molecules bound/protein molecule . . . . . . . . . Plot of the absorbance at respective wavelengths vs the protein concentration of a meat-lipid system . . . . . . . The dye binding ability of three different protein classes and the dyes A0_10 and AO—lz o o o o o o o o o o o o o o a The effect of anion type used in myofibrillar protein extraction upon the DBC of the azo dyes A0-12 and AO-lO . The effect of A0-12 upon the disc gel electrophoretic properties of myofibrillar, sarc0plasmic and stroma Proteins I I I I I I I I I I I I I I I I I I I I I I I I I Comparison of the dye binding ability of beef and pork myofibrillar proteins in a model emulsion system with AO—lz (103 tug/ml) and Ao-lo (0.04 mg/ml) o o o o o o o o 0 Comparison of the dye binding ability of beef and pork sarcoplasmic proteins in a model emulsion system with A0-12 (1.3 mg/ml) and AO-lo (0.04 mg/ml) o o o o o o o o o The effect of three different lipid levels and two emulsion states upon the DBC of proteins in a sausage emulsion for AO-lO (0.04 mg/ml) . . . . . . . . . . . . . . . . . . . . vi Page 33 35 36 40 43 45 48 50 53 55 59 51 66 figure Figure Page 14 The effect of three different lipid levels and two emulsion states upon the DBC of proteins in a sausage emulsion for Ao-lz (1.3 mg/ml) o o o o o o o o o o o o o o o o o I o o 67 vii ‘n‘itJ greater . necessar praduets Rapid me have bee tein con method. The tein det cream ar PIOtein from Udj for the Beat pr fresh m We Stu inveSti teSt th binding tein de F0] The? Wer- INTRODUCTION With the advent of legislation regulating processed meat products, greater control of product composition by the manufacturer has become necessary. The high speed processing utilized to manufacture the meat products necessitates a rapid form of analysis of product composition. Rapid means of analyzing for the fat and water level in a sausage product have been found and are in use today, but the determination of the pro- tein content continues to involve a time consuming operation, the Kjeldahl method. The recent success of the dye binding method in automating the pro— tein determination in such diverse products as cereal grains, milk, ice cream and ground meat meals, provided the possible means for a rapid protein determination in a processed meat product. Using adaptations from Udy (1971) and Ashworth (1970), two dyes AO-lO and AO-lZ were tested for their applicability in determining the protein content of a processed :meat product. Previous work by many investigators involved the use of fresh meat, ground meat, meat meals or frankfurters. However, except for two studies (Torten and Whitaker, 1964; and Bunyan, 1964) all of the other investigators included meat only as another protein source upon which to test the method. Therefore, it was felt that further study of the dye 'binding method was necessary to ascertain its value as a method for pro- tein determination. For this study, two dyes were eventually chosen, A0-12 and AO-lo. ‘They were chosen for their different binding abilities. These two dyes were exposed to a processed meat product, frankfurter in two states, completely processed and raw emulsion. The product composition was varied to simulate a broad range in composition parameters with the median product representing the federal guidelines. Further study was done to determine the applicability of model system work to the actual product and to ascertain basic facts concerning the dyes and the proteins to which they bind. LITERATURE REVIEW General Theory of Dye Binding The use of dyes in conjunction with protein structure elucidation or with the determination of protein content has been reported extensively in the literature. Chapman 32 a1. (1927) working with gelatin, deaminized gelatin and casein reported that the amount of dye which combines with a particular protein bears a definite relationship to the content of the basic amino acids in the protein molecule. He further stated that the amount of dye which is combined with the protein molecule can be correlated with the free basic groups of arginine, lysine and histidine. Putnam and Neurath (1944) working with anionic detergents reported similar attachment to the cationic groups of proteins. Fraenkel-Conrat and Cooper (1944) concluded from their work that Orange G combined stoichio- metrically with basic protein groups, specifically the guanidyl, imidazole and amine (a and 3) groups, in a buffer of pH 2.2. It was also shown that Safranine 0 at pH 11.5 binds with the carboxyl, phenyl and thiol groups present on the protein. They worked with Egg Albumin, B-Lactoglobulin, Casein, Fibrin, Gelatin, Gliadin, Insulin, Lysozyme and Zein. Klotz gt ‘al. (1946) proved the validity of statistical considerations with regard to the binding of the anionic dye Orange G (AC-10) to the protein, crystal- line bovine serum albumin (BSA), and further concluded that the binding of one of these anions does not affect the binding of a second anion, except in so far as the first anion reduces the number of spaces available far t fact sin; protc refer to tt is tt in rt than hydro leavi dOne - thoti aers t he fel and Or the dy 01 the “lar c for the second anion by one. Further note was made with reference to the fact that the maximum number of bound anions, 22, corresponded roughly to the number of arginine residues in the bovine serum albumin molecule. Schroeder and Boyd (1957) studying the acid dyes, Quinoline yellow, Fast Red E and Anthroquinone Green C, and their reactions with basic polymers concluded that the basic attraction is ascribable to ionic forces, i.e., simple coulombic forces result in the affixation of the dye anion on the protonated fiber cation. These findings are applicable to the class referred to as "acid dyes". Klotz (1957) sought to collate all of the information on dye binding and relate it to protein structure. This led to the hypothesis that an acid dye's ability to bind at»a particular site is the result of a low hydroxyl amino acid and basic amino acid content in the protein, particularly if the number of hydroxyl groups is less than the carboxylic and basic groups. In this state, he noted, the hydroxyl groups tend toward hydrogen bonding with the carboxylic groups leaving the basic groups relatively free to react with the dyes. WOrk done with human serum albumin (BSA) and acid dyes appears to support this hypothesis. Osipow (1962) attempted to measure the surface area of poly- mers via the configurational requirements of the binding dyes, however, he felt further work was necessary. working with crystal violet chloride and Orange II (AD—7) on samples of anatase and rutile he concluded that the dyes adsorbed to form a bimolecular film by lying flat on the surface of the polymer. Thus the possibility exists that any change in the mole- cular configuration of the dye away from planarity could result in steric residUe hindrance to adsorption. Rosen and Klotz (1957) described the reaction between dye and protein in terms of a series of multiple equilibria, since different binding sites on the protein have different affinities for the dye molecule. Laiken and Nemethy (1971) in describing a model for binding flexible ligands support the hypothesis for multiple equili- bria in binding systems but add that this model appears reasonable only for small ions and other rigid monofunctional molecules. Povlovskaya and Pasynski (1963) found that dye binding declined steadily with increasing atmospheric pressure (0-760 mm) and was always below that found under comparable conditions in gaggg, Low oxygen pressure showed greater ten- dency to displace the dye than did the high pressures. It was suggested that this effect could be the result of competitive displacement of dyes. Aizawa (1969) utilizing equilibrium dialysis determined the binding capacities of Orange G (AO-lO) and Brilliant Orange (AD-12) as 2 and 5 molecules bound dye per molecule of protein, respectively. Moriguchi gt 31, (1971) opposed the idea of covalent or electrostatic bonding to explain dye binding with bovine serum albumin. The results of their work with dyes in a low polarity environment appears to indicate hydrophobic bonding. Jonas and Weber (1971) utilizing 1-anilinonaphthaLene-8—sulfonate and bovine serum albumin lend support to the hydrophobic bonding hypothesis. They suggest that the strong anionubinding sites on serum albumin include cationic and hydrOphobic residues with indications that tryptophan residues are near the binding sites. Further evidence is presented that arginine residues are closely associated with the strong anion binding sites of dyes f aPplic areas ¢ 5in mi “y (15 PIOtein fitters Cluded Sites 0 stated serum albumin. Terada (1971) reaffirmed the ionic bonding concept between the anionic dye and the cationic residues. Support of his thesis was pro- vided by the use of Ponceau 3R dye and BSA, which suggested that the binding sites of BSA to Ponceau 3Rwere related to cationic residues of BSA, such as cramino, gramino and guandine groups. The basic hypothesis concerning the binding of dye molecules to sites on a protein molecule involved the formation of bonds with the cationic residues on the protein. However whether the bonding involved only the availability of the site for binding (Klotz Qfinfila: 1946; and Terada gt .31., 1971) or whether selectivity is also involved, i.e., one site being more favorable than another (Rosenrand Klotz, 1957; and Laiken and Nemethy, 1971) has yet to be resolved adequately. Application of Dye Binding Technique: Dairy Chemistry With the above described data accumulated on the applicability of dyes for use in protein chemistry a number of workers have sought practical applications for the dye binding process in analytical procedures. The areas of most notable success were Dairy and Cereal Chemistry. The dye binding method readily availed itself to rapid analysis and even automation. Udy (1954, 1956) used Orange G (AD-10) to determine quantitatively the protein fractions of wheat flour. Ashworth and Seals (1957) reported on factors which affected dye binding in milk. Ashworth £5 31. (1960) con- cluded from further work with Orange G and milk that all of the binding sites on milk protein dofinot have the same affinity for dye. Dolby (1961) stated that under given conditions the dye binding of any one protein is quantitative, but that proteins vary so much in this property that any change in the proportions of proteins in a natural product may considerably alter the amount of dye taken up per gram of protein, i.e., the dye binding capacity (DBC). Indeed this same note of caution was echoed by Colen— brander and Martin (1971) who noted a difference in DBC in sows milk be- tween species of sows which they thought was likely due to the variation in the relative distribution of protein or nitrogen constituents or both. Ashworth (1966) adapted the dye binding method to a wide range of dairy products. He found the method applicable to protein determination in fresh milk, evaporated milk, sherbets and whole milk, chocolate milk and cheddar cheese. The tests were performed with Orange G (AO-lO) and Brilliant Orange (AD-12). While much of the early work with the dye binding process in milk was done with Acid Orange 10 (Orange G), Tsugo g£_§l, (1966), determined the applicability of twenty three (23) dyes to protein determination. Among the twenty three dyes were Orange G (AC-10) and Amido Black 103 (AB-l), but they served as standards of comparison. The dyes were compared for their solubility in a citrate buffer (0.1M citrate buffer at pH 1.9), the stability of the solution, absorbance index (K), and the binding capacity with protein in milk. Four dyes proved adequate in three of these categories and possessed excellent binding capacity in milk, the fourth criterion. They were Acid Orange II (AC-7), Resorcine Brown (AD-24), Xylene Acid Milling Orange (AD-65) and Diacid Light Green GS (AG-25). The authors stated that these three dyes excepting Resorcine Brown could be 3?; intere 0: age in; an utiliz sis ta the Ha Co. milk as "in 8011 stated 1 ratio, c She method (: detemim‘ dry milk, be applied with safety to the bulk protein testing of milk. It is interesting to note that they thought that Amido Black 103 (AB-1) and Orange G (AO-lO) appeared to be relatively inferior, both in protein bind- ing and in the desired properties of a dye. Yet Hammond g£_§1, (1966) utilized Orange G in a binding method to determine the amount of proteoly- sis taking place in a milk sample. The method was not as accurate as the Hall and Rosen tests but it did not depend upon complete proteolysis and the release of amino acids before being useful. As a further exten- sion of this technique, May (1970) proposed the use of Remazol Brilliant Blue R (AB-21) to detect the loss of available lysine through the pro- cessing of cheddar cheese. His method entailed the use of heat to acti- vate the dye and a coarse Sephadex G25 column eluted with(125M Borate buffer at pH 8.0. As the dye-protein complex moves down the column, the unchanged dye follows it. By this method he was able to report no loss in available lysine through processing. Cole (1969) in a review of methods of protein determination cited milk as the ideal substrate for Orange G binding, since the proteins are "in solution" ready for complex formation and precipitation. He further stated that this method proved capable of detecting the decline in the ratio, casein/total protein which occurs in milk from cows with mastitis. Sherbon (1970) presented the recommendation that this dye binding method (the Udy Dye Binding Method) be adopted as an official method for determining the protein contents of fluid milk, half and half, non fat dry milk, ice cream mix, chocolate drink and butter milk. The dye he reumende laborator' protein. oiiieial techniquu in Europ Arid-o Bl Danish c Ho proved the met several binding t0 Show range techniq the San total 3 equiltio flour d Protein Starch. recommended was Acid Orange 12 (Brilliant Orange). He reported an inter- laboratory coefficient of variation of 1% or less, i.e., around 0.03% protein. The same measurement for the Kjeldahl method is 3%. Thus an official application of the dye binding technique was established. The technique has already been automated and introduced to the dairy industry in Europe. Thomasow gt_al, (1971) reported that a six month trial using Amido Black 10 B (AB-l) for routine protein determinations in a commercial Danish dairy yielded results with a mean error of i 0.005%. Application of Dye Binding Technique: Cereal Chemistry However, the application of this technique to a non-fluid system proved to be more challenging. Udy (1954) demonstrated the potential of the method in determining the protein content of wheat. He showed that several of-the wheat fractions exhibited considerable constancy in dye binding powers and it should be emphasized that different varieties fail to show any significant variation within any given protein. He used Orange G (AD-10) and Safranin O (BR-2). However, in working with this technique, he found that the particle size of the sample could influence the sensitivity of the test, since it would affect solubility and the total surface presented to the dye. Udy (1956) noted that a regression equation with a correlation coefficient (r) of 0.997 for straight grade flour did not hold for other grades, and that the amount of dye bound by protein is the difference between the total dye bound and that bound by starch. However, Cole (1969) added that since the starch content of flour averaged 67%, 10% fluctuations in the protein value by the DBC method netted in {1361) c :oztent the tect abuilt Bur :owards soybean azmg t tandar felt, c were ex noted 1 10 method would still be within the errors of other methods. Hart g£_§l, (1961) criticized the use of the Udy analyzer in determining the protein content of wheat because of lack of reproducibility. But their criticism dealt mainly with the mechanical apparatus and not with the efficacy of the technique. They suggested a recalibration of the apparatus to nullify a built in bias resulting from preparation of samples for analysis. Bunyan (1959) attempted to apply the Orange G dye binding technique towards determining the protein content of various meals, e.g., whalemeal, soybeanmeal and groundnutmeal. He encountered a great deal of variation among the samples but the soybean meal and groundnut meal had the lowest standard deviation of all of the samples, 11.5-1.7%. This variation, he felt, could be tolerated but he expressed concern that if an atypical meal were encountered the dye binding results would be entirely misleading. He noted that for bound dye vs protein (%) none of the regression lines passed through the zero point. This, he concluded, would mean that protein was not the only constituent binding the dye. Moran gt El- (1963) also working with soybean meal observed that the capacity of soybean meal to bind Orange G decreased with.heat treatment. The "quality" of the meal did not change significantly until the soybean meal was autoclaved (121°C) for more than 45 minutes; a second decrease occurred at 90 minutes. This index of "qual- ity" was established by using a ratio of the dye binding capacity of a processed soybean meal to that of a rawxmeal, a change in the percentage of protein as determined by a regression equation. These workers concluded that the Orange G method should be considered to be more specifically related to the nutritional value of the soybean meal than protein content t:e 1:1 I. I an. to COX] the DUI 11 since it reacts with amino acids of critical nutritional importance. These results were corroborated by Pomeranz (1965) who demonstrated via the dye binding procedure and the biuret method that protein values were lower in rather advanced stages of soybean meal heating. These lower protein values in soybean meal were attributed to heating which initiated a series of changes including a decrease in protein solubility, loss of specific activity of enzymes and hormones and changes in the availability of certain amino acids to enzymatic digestion. He concluded that both methods would suffice to detect toasted or even oven heated soybean meal. Mossberg (1965) listed the problems of applying the dye binding and Kjeldahl techniques to protein determination in cereals. Of relevance to our study would be the variation in binding of dye due to variation in protein content and heating. The variation in protein content was due to the genetic make up of the cereals and extraction procedures. The cause of variability by heating was attributed to denaturation. Also cited as causes for variation were milling characteristics, interfering carbohy- drates, variation in nitrate content, variation in amino acid and peptide content, variation in the proportions among different proteins and in the method of handling of samples for analysis which would also concern our study. He also detected a decrease in DBC with heating in wheat samples. Mossberg (1966) noted a decrease in barley DBC upon heating. Outen 35 a1. (1966) presented data relating the DBC to protein con— tent of herbages. Several points should be noted. Although herbages contained soluble coloring materials, these did not contribute significantly in: I in- nu» 12 to the absorption @ 470Tmaof Orange G (AO-lO). The overall relationship between the total nitrogen content of the grass and the uptake of dye was linear, but the scatter of points was much greater than had been found in cereals and milk. This he attributed to soluble protein present in the dye protein solution which would not precipitate upon binding the dye. Kastner (1965) achieved excellent results with Orange G and a variety of cereal products. Szeverengi and Hazko (1966) also obtained acceptable results in agreement with semi micro Kjeldahl results. They worked with malt, wheat and yeast samples in Orange G. Volodin 35 31. (1968) utilized the Orange G method to determine the protein fraction in pea pod seeds and achieved results which were closely correlated with Kjeldahl results. Mossberg (1968) found that in all of the cereals (excluding corn) there was a correlation between nitrogen content and the DBC, and between basic amino acids (BAA) and the DBC. The correlation coefficient (r) between nitrogen and DBC was r = 0.767. For the basic amino acids and DBC, r was 0.940 and for the total material and the DBC, r was 0.819. Kaul g£_§1, (1970a) utilizing a rapid semi micro dye binding technique developed for rice found the method readily adaptable to maize, wheat, mung beans, red gram, sorghum and millets. Kaul g£_§l, (1970b) further adapted the method using Orange G to bengal gram, green gram and jower. Lawrence g£_al, (1970) utilized Amido Black 10B (AB-l) with polyacrylamide gel electrOphoresis, and through the use of a densitometer, he was able to determine the DBC and dispersion of various protein fractions in wheat flour. hzclicatior With 4 birding tee lhe work 11 on :eat an. products. Bunya seals and the bound Passed thr plicity of Torte from beef Elation b of Protein an eXponen hound, H01 Vere reduC‘ tein decree decreased. fluid be co 13 Application of Dye Binding Technique: Meat Products With evidence and success cited in other fields of study, the dye binding technique was applied to meat, meat products and meat meals. The work in this area is very limited. Most applications were attempted on meat and meat meals with little direct work on sausage emulsions or products. Bunyan (1959) attempted to analyze crude protein content in "meat" meals and fish meals using Orange G. However a great variation between samples was noted by this author. For example, standard deviations of $3.1 to 3.6% protein were found for meat meals, $4.2 to 4.4% whalemeal and i4.7 to 5.0% fish meal protein. And as has already been noted, that for the bound dye vs protein (%) none of the regression lines from this study passed through the zero point. Therefore, he felt that the speed and sim- plicity of the Orange G method did not outweigh the errors involved. Torten and Whitaker (1964) worked with a number of meat types ranging from beef and pork to cod and chicken. For Amido Black 103 (AB-l) the relation between the amount of dye bound (mg) per gram of sample and gram of protein was expressed better by means of a curved line. This indicated an exponential relationship between protein content and the amount of dye bound. However, a linear response might be elicited if the sample size were reduced to 0.3 gm. They also stated that as the concentration of pro- tein decreased the quantity of Orange G (AO-10) bound per unit of sample decreased. This is important for an empirical study, but they felt that it could be corrected in a routine analysis where the protein concentration 14 varied only 2"4%- As a possible explanation for this occurrence, they offer the physical blocking of the formation of the dye-protein complex by the lipid molecules present in the samples of lower protein content. The y-intercepts of the regression equations for breast of chicken and cod are close to zero but on the opposite side of zero, implying that more Orange G is bound per unit of protein in samples of lower protein content for these species. In explanation of this occurrence, they cite the presence of excess dye in solution and the possibility of physical adsorption of the dye molecules onto the protein. Or a problem of steric hindrance in the upper range of protein concentrations due to the bound molecules themselves may be involved. In spite of these difficulties, they found it reasonable to believe that the Orange G dye binding method could be satisfactorily adapted to routine estimation of the protein content of meats where the range of values does not exceed 2-4%. Bunyan and WOodham (1964) found that in the comparison of fresh meals, the sample with the higher lysine, arginine and histidine content did indeed bind more Orange G than other samples. Apparently they were seeking to estab- lish the same relation of protein "quality" to dye binding as previous work cited in the field of cereal chemistry. Moss and Kielsmeier (1967) worked with meat slurries and Amido Black 10B (AB-l). They found that accurate results could only be achieved with Orange G if scrupulous attention was paid to technique. Therefore they sought to use a dye with a higher molar optical density. However with AB-l, as the dye concentration increased, not only primary sites but 15 secondary sites on the protein were progressively occupied by the dye. Thus at the low protein levels, 370-430 mg of protein, more dye could be bound than would correspond to the stoichiometric ratio. The filtrate would contain less dye, appear lighter and be closer in optical density to the high.protein level filtrate. When the DBC was correlated with the protein content of the samples, a correlation coefficient (r) of 0.625 was obtained. Thus approximately one third of the variation in the dye binding capacity could be associated with the protein content of the samples. Cole (1969) reviewed work in which the ability of collagen to bind dyes was related to the extent of degradation that the collagen sustained. The collagen was highly purified hide collagen, thermally degraded and reacted with Orange C. It was found that as degradation increased the amount of dye bound increased. In the undegraded material less than half of the theoretically available anionic groups were reactive but more of them became available as the extent of degradation increased. Salthouse £5 31. (1971) reported on the binding of the vinyl sulfone dyes, remazol brilliant blue R, RH, RV and B to the lysine and hydroxy- lysine groups of collagen. However, their interests were directed towards in_gi££g staining and not quantitative analysis. Ashworth (1970) attempted to adapt dye binding procedures to a number of food types. His findings proved interesting since he tested a wide variety of meat and meat products. To discount the effect of lipid material and protein coagulation upon dye binding he heated ground beef patties to 160°C for 40 minutes with no effect on the DBC. Pork bound the ' o '7‘ N— *«5. 16 dye, Acid Orange 12, to the same extent as beef, thus species variation was not discernible through_DBC. However, chicken meat had a higher DBC than either pork or beef, but there was no significant difference between light and dark.meat. A further disclosure was presented in the demon- stration that cooked chicken luncheon meat and raw contaminated chicken meat from a mechanical deboner bound less dye than isolated muscle tissue. Nature of Sausage or Meat Emulsion Products The work of Torten and Whitaker (1964) and Ashworth (1970) appeared to indicate the futility of using dye binding technique to quantitate the protein content of fresh meat products. Yet the successful appli- cation of the technique to milk which is an oil-in-water emulsion system by Udy (1954, 1956, 1971), Ashworth £5 31. (1957, 1958, 1960, 1965, 1966 and 1970) and Sherbon (1970) seemed to indicate that a sausage emulsion might be the proper medium for analysis. Sausage emulsions, frankfurter, bologna, braunschweiger, etc., are considered oil-in-water emulsions. Saffle (1968), Swift §£_§l, (1961), Hansen (1960), Ivey gt_§l, (1970) and others have referred to the un- processed sausage product as an emulsion. Saffle (1968) and Kramlich (1971) consider the sausage emulsion to consist of a continuous phase of water with a discontinuous phase of fat or lipid material. The proteins are said to form an interface between the continuous and discontinuous phases. The application of the dye binding technique to an emulsion product infers a study of the parameters of that product. E 17 Particle Size Saffle (1968), Becher (1965) and Kramlich (1971) find it difficult to place a sausage emulsion in a category with "true" emulsions. Osipow (1962) and Becher (1965) place a size limit for the particles in an emulsion at 0.1 micrometers to 50 micrometers.r From the photographs of Hansen (1960) and Helmer and Saffle (1963) of commercial meat emulsions the fat particles are larger than 50 micrometers. However, Borchert £5 .31. (1967) have shown that some of the particles can be as small as 0.1 micrometers. “ Ackerman gt a1. (1971) demonstrated that fat parti- cles with a diameter greater than 200 micrometers tended towards fat separation. Therefore one must conclude that a sausage emulsion possesses particles which range from approximately 0.1 micrometer to less than 200 micrometers. Membrane Formation The particle size of a sausage emulsion is important since it deter- mines the amount of protein which will potentially be able to bind a dye added to the emulsion. Particle size is the determinant since a major fraction of the protein forms a membrane around the fat globule. Swift gt a1. (1961) found that 59.5 to 75% of the water soluble proteins and 79.8 to 84.1% of the salt soluble proteins were removed from the emulsion to form membranes. The term water soluble protein includes the sarcoplas- mic proteins which are soluble in water or dilute salt solutions. The term salt soluble proteins encompasses the myofibrillar proteins (Lawrie, 18 1966). Saffle (1968) has emphasized the division between water and salt soluble proteins is not very clearly delineated when dealing with pro— teins in an emulsion. Hegarty e£_a1, (1963) stated that actomyosin and myosin in their native states were completely utilized in the interface of oil and water emulsions. He found further that the presence of salt (0.3M KCL) greatly depressed the effectiveness of actin. It was found that the salt soluble proteins formed thicker membranes than the water soluble proteins (Swift, 1961). Saffle (1968) demonstrated the formation of this membrane in an aqueous solution of myofibrillar proteins into which was introduced a lipid material, in the form of a globule. In this system the thickness of the membrane was found to be a function of time and the membrane appeared to grow by accretion of protein layers around the fat globule. Grau (1971) stated that the stability of an oil-indwater emulsion of salt-free—sausage meat was improved by the formation of a protein film around the freely dispersed fine fat particles. The meat protein served as the emulsifier and its emulsifying effect depended upon the solubility of the protein in aqueous solutions. Membrane Proteins Much work has been done on the nature of the membrane that surrounds the fat globule. Prior to the use of the dye binding technique, membrane study was limited to microscopic techniques (both light and electron beam) and biochemical assays of membrane constituents. Via the dye binding technique further study of intact membrane proteins is possible. Hansen 19 (1960) stated that the membrane was formed entirely from salt soluble proteins and at least in part of myosin and actomyosin. Saffle (1968) reports that with one exception, the salt soluble proteins have been reported by all research workers to be superior to water soluble protein in the amount of fat emulsified. Swift and Sulzbacher (1963) emphasized that salt soluble proteins were 30-400% more effective in emulsification capacity than water soluble proteins. Or in another study that water soluble proteins were only 70% as efficient in emulsifying capacity as salt soluble proteins (Carpenter and Saffle, 1965). These authors titrated the salt soluble proteins and found sharp drOps in the pH curve at 3-5, 10-12 and 6-7 which would indicate the presence of carboxyl, amine (a and e) and imidazole groups, respectively. Since the nature of the dyes used in this study necessitate the presence of positively charged groups, the presence of the amine (e) and imidazole groups exposed to the environment present reactive sites for the dye. Concerning other proteins which might be present, Trautman (1964) found that water insoluble residue (connective tissue)8PPeared to possess very little emulsifying power. In fact, they were rated similar to the water soluble proteins. However, Van Eerd (1971) noted from his work with mutton in sausage emul- sions that the insoluble proteins contributed to the emulsifying proper- ties of meat, but to a lesser extent than the water and salt soluble proteins. Basic Amino Acid Residues Available in Protein Membrane Since it has been emphasized in the study of the dye binding techniques that azo dyes tend to bind specifically with the basic amino acid residues 20 of the protein component, it is worthwhile to note the quantity in which these groups are present in the system. Bodwell and McClain (1971) have illustrated that the basic amino acids represent between 12-24% of the total residues present in the constituent muscle protein. Acidic residues range from 14-28% and the amino acid residues with polar side chains represent the largest component in a range from 33—55% of the total resi- dues present. They also report that myosin from beef and pork contain 19.7 and 21.1% basic amino acid residues, respectively, in the total amount of residues present. Golovkin g£_§1, (1967) reported that the mg% of the basic amino acid residues increased with storage in a -2°C temperature for periods in excess of two weeks. They reported critical values of 9.19% for histidine, 6.78% for lysine and 4.26% for arginine, which increased to 12.31%, 7.15% and 6.15%, respectively, after 14 days. And after 20 days at the same temperature, the respective values increased to 14.6%, 9.15% and 9.49%. Effects of Non-Protein Components Upon Membrane Formation Swift 25 31. (1961) disclosed that the amount of fat emulsified in- creased as the rate of fat addition increased. He explained the occurrence as the result of near "instantaneous" protein membrane formation with the fat dispersed by mixing and stirring in the forming emulsion. Townsend §t_§l, (1968) determined that the low melting fat exhausted the emulsify- Tng capacity of the meats, consequently any increase in the temperature of comminution of the emulsion produced additional fat liquification which 21 led to instable protein membrane formation. Christian and Saffle (1967) found that the amount of emulsification was greater with shorter chain triglycerides than with longer chain triglycerides and that monoenoic fatty acids were preferred over dienoic fatty acids. However, if the carbon chain was the same length more mono- or dienoic fatty acids were emulsified than saturated fatty acids. In addition they found that more triglycerides can be emulsified than the corresponding free fatty acids. Swift 25 a1. (1968) reported that emulsion stability was higher for beef fat over pork fat at comparable temperatures during comminution. However, Townsend £5 31. (1971) demonstrated that the high melting point of beef fat required higher temperatures for thorough, uniform dispersal in the emulsion. Morrison gt_§l, (1971) found emulsion stability highly dependent upon the level of added water. They add that the fat and lean can be varied greatly but the level of water is narrow and critical. Source of Membrane Protein The type of protein available for membrane formation in a sausage emulsion depends upon the state and the condition of the meat as well as the type of meat. Borton gt_al, (1967) found that the emulsifying capa- city of the various meat trimmings decreased with increasing fat content. They also found that samples from various species that had a similar lean content ranked approximately the same in emulsifying capacity, i.e., the emulsifying efficiencies of proteins from the various red meat sources are similar if the fat contents are similar. Earlier, Saffle and Galbreath I ‘1‘ '1" b.l: 1:1 a.“ 22 (1964) found that the amount of fat had no effect on the percent protein that could be extracted. For meat with approximate lean percentages of 100, 80 and 60%, the percent salt soluble proteins extracted was 30.4, 30.0 and 30.4, respectively. Furthermore, an increase of pH from 5.5 to 6.5 in 0.5 pH increments yielded the following salt soluble proteins ex- pressed as percent of the total protein content: 35.9, 38.9 and 42.4, respectively. Trautman (1964), Saffle and Galbreath (1964) found a dif- ference in the amount of protein which could be extracted from pre-rigor and post—rigor meat (48 hours) with 50% more salt soluble protein extracted from the pre-rigor meat. The use of frozen meat was found to increase the minimum water requirement necessary for satisfactory physical pro- perties in an emulsion (Morrison 33 31., 1971). McCready and Cunningham (1971a) associated the greater emulsifying capacity of dark meat over light poultry meat to the higher pH of normal dark meat and the greater predominance of protein fractions associated with emulsifying characteris- tics that were released, even though the light meat had significantly higher amounts of total protein and salt soluble proteins (McCready and Cunningham, 1971b). In working with turkey meat, Neelakantan and Froning (1971) attributed the better emulsifying capacity of pre-rigor meat to higher pH and the presence of free actin and myosin. He also stated that the sarcoplasmic proteins contributed to emulsion stability in emulsions of post-rigor meat. Townsend £3.21: (1968) stated that emulsions prepared from frozen materials were as stable as those prepared from chilled mater— ials at corresponding temperatures. The work of Grauer g£_§1, (1969) 23 noted that at refrigeration temperatures the soluble protein values in- creased. They attributed the phenomenon to microbial contamination. Three gregared it and from th selected fc iomlatior rounds. Fa were in the it was gro' through th grinding y Send the; 10515) am evacuated blast fre days prio Pric each of t Substitm deSired 1 METHODS AND MATERIALS Meat Preparation for Sausage Emulsions Three different frankfurter emulsions (24%, 30% and 36% fat) were prepared from meat obtained from the central food stores at the university and from the abattoir located in the Meat Laboratory. The meat was selected for its particular lean to fat ratio to facilitate emulsion formulation. Lean beef (high protein, low fat) was represented by beef rounds. Fat beef (low protein, high fat) and pork (low protein, high fat) ’were in the form of plate beef and pork jowls. Once the meat was obtained it was ground by two passes through the 3/4" plate and a third pass through the 1/4" plate of a Toledo meat grinder, model No. 84181D. The grinding was done in a walk-in cooler maintained at 40°F (4°C). The ground meat was segregated according to type (lean, plate beef and pork jowls) and packed in Cryovac bags in 10 and 12 pound units. Air was evacuated from the bags, they were sealed and frozen in a -20°F (-28.8°C) blast freezer. The meat was allowed to thaw in the 4°C cooler for two days prior to use. Prior to sausage preparation, a proximate analysis was performed on each of the meat components. The results of the proximate analysis were substituted into formulation equations (Cunningham, 1971), to obtain the desired fat levels. Appendix II contains the formulations used. 24 25 Sausage Preparation All of the meat for a single sausage formulation (24 lbs), usually in a partially frozen state, was placed in the bowl of the Hobart Vertical Cutter Mixer, Model No. VCM4OE, approximately 60 lb capacity. Ice and seasonings were added directly to the batch; however, the sodium nitrate and sodium nitrite were dissolved in 15 m1 of water before addition to the ingredients. The cutter was operated at high speed for three minutes under a 22 inch vacuum with the direction of the agitator blade reversed at the end of two minutes. The temperature of the emulsion was measured prior to its transfer to an emulsion mill (Griffiths Mince Master with 1/16" diameter holes in plate and 3460 rpm). The emulsion passed once through the mill with the blade set at 150 inch lbs of torque. The temp- erature was again recorded prior to transfer of the emulsion to the E-Z Pak water stuffer. The No. 26 Nojax casing (Union Carbide Corporation) was stuffed to an approximate diameter of 21 mm. Samples were segregated at this point into processed and unprocessed emulsions. The unprocessed emulsion was stored in Whirl Pak bags in the freezer (-30°C) and cooler (4°C). The processed emulsion was stored in sausage link form in the freezer and cooler. Two or three strands of each different emulsion were prepared and weighed for smokehouse shrink data. The processing schedule was as follows: 1. Smokehouse pre-heated to 52-55°C (125—130°F). 2. 20 min at 52—55°C dry bulb setting and 32°C (90°F) wet bulb setting. 26 3. 30 min at 60°C (140°F) d.b. setting and 38°C (100°F) w.b. setting. 4, 39 min at 66°C (150°F) d.b. setting and 43°C (110°F) w.b. setting. 5. 30 min at 71°C (160°F) d.b. setting and 49°C (120°F) w.b. setting. 6. The time needed at 74°C (165°F) to raise the internal temperature of product to 70°C (158°F). Smoke was applied to the product during stages 2-4. A cold shower was applied to the product for 5-7 minutes which was then allowed to stand at room temperature until an internal temperature of 27-32°C (BO-90°F) was attained. The product was further chilled in the 4°C cooler for 24 hrs. The next day random samples from the three strands were selected for immediate testing and freezer storage. For analysis of protein content via the dye binding procedure and the Micro-Kjeldahl method, the randomly selected frankfurters were passed through a 1/8" plate on the small commercial meat grinder with 2 1b capacity(3 timesh The unprocessed emulsion was not ground. Preparation of a Model System Emulsion Crude protein extracts of the salt soluble proteins present in ground round and pork jowls were utilized in this study. Six different protein concentrations were prepared by adding 5, 10, 20, 30, 40 and 50 m1 of protein extract, respectively, to a 50 m1 volumetric flask and filling to mark with 0.03M phosphate buffer. The protein solutions were prepared in the 4°C cooler but the emulsification of the protein-oil mixture took place at ambient temperatures. This protein solution and 1.547 g of sodium chloride were added to a Virtis homogenizing cup (5—50 ml capacity). This {{aCi-prote seconds t:. milliliter room temp: roiled \‘i rheostat "high" an When appearanc: dye bindi; was teste The Of this F w“EYE! bler Waring b1 SEtting. 75 1:11 0.C 3aEnetic to a 250 in 3 30w 260° The 3'19- residr 27 NaCl-protein mixture was mixed at a rheostat setting of 20 for 15 seconds to insure the homogeneity of the mixture. Twenty-one and a half milliliters of Wesson Oil (a polyunsaturated soybean-cottonseed oil) at room temperature was added to the mixture and mixed on a rheostat con- trolled Virtis homogenizer. The complete mixture was homogenized at a rheostat setting of 60 for 2 minutes. (The Virtis rheostat was set on "high" and controlled completely through an auxiliary rheostat.) When emulsification was complete, as evidenced by a creamy white appearance, samples were taken for Kjeldahl protein analysis and for the dye binding analysis. For each form of analysis, a 5 m1 aliquot of sample was tested in duplicate. Protein Extraction Procedure The method of Helander (1957) was modified to meet the specific needs of this project. Thirty grams of ground round beef or ground pork jowls were blended in 210 m1 0.03M phosphate buffer, pH 7.4, for 1 minute in a Waring blender set at "low" speed and for 1 minute at the "high" speed setting. This slurry was transferred to a 500 m1 Erlenmeyer flask with 75 m1 0.03M phosphate buffer. The slurry was gently agitated with a magnetic stirrer for 4 hours at 4°C. The mixture was then transferred to a 250 ml centrifuge bottle and centrifuged at 1400 X G for 20 minutes in a Sorvall Superspeed RC2-B Automatic Refrigerated Centrifuge set at 2°C. The supernatant was filtered through cheese cloth and retained. The residue was re-suspended in the centrifuge bottle with 250 ml of 0.03! phOSP' was centrif chained vi soluble pm The r: M and ex pended as present th The r by vigorou the water- Micrc the amount Two 11 Ashworth . E 230 ml 0. Vere Weig! Weight of five mill: VhiCh Wag “‘38 centr 28 0.03M phosphate buffer and re-extracted for 4 hours. This re-suspension was centrifuged and filtered as previously described and the supernatant combined with the first supernatant. This fraction represents the water soluble proteins. The residue was resuspended in 1.1M KI, 0.1M phosphate buffer at pH 7.4 and extracted for 2 hours. It was centrifuged, filtered and resus- pended as described previously. The supernatants were combined and re- present the salt soluble protein fraction. The residue was resuspended in 0.03M phosphate buffer accompanied by vigorous shaking. The suspension was stored in this form and represents the water—salt insoluble protein. Micro-Kjeldahl determinations were run on each fraction to determine the amount of protein present in the fractions. Dye Binding Procedure Two methods were utilized in this study, the procedure proposed by Ashworth (1970) and a modification of this method. The Ashworth Method: Approximately 15 g of sample were blended in 250 m1 0.1M citric acid in a Waring blender for 2-3 minutes. Aliquots were weighed and placed in 50 ml polycarbonate centrifuge tubes. The weight of the sample was adjusted to 5.0 g with distilled water. Twenty- five milliliters of the dye reagent were added to the sample solution and which was shaken vigorously in a stoppered tube for 30 minutes. The mixture was centrifuged 5,860 X g for 5 min and filtered through medium porosity 29 paper. Absorption at 0.475 microns was measured and the amount of pro- tein present in the product read from a standard curve. The Modified Ashworth Method: To 51.0 ml 0.1M citric acid, 0.2M dibasic phosphate buffer, pH 5.5, was added 3.75 g of the emulsion sample in a Virtis homogenizing cup (5~50 m1 capacity). The mixture was homo- genized at a rheostat setting of 30 for 2 minutes. Two, 5.0 m1 aliquots of the mixture were transferred to Kjeldahl flasks via a wide mouth 10 m1 transfer pipette. The remaining mixture was transferred to a 250 m1 centrifuge bottle. The homogenizing cup was washed with 80.0 ml of dye reagent and collected in the centrifuge bottle. (The dye solution consisted of either A0-12, 1.3 mg/ml concentration or AO-lO, 0.04 mg/ml concentration dissolved in 0.1M citric acid, 0.2M dibasic phosphate buffer, pH 3.0). This mixture of dilute emulsion and dye reagent was gently agitated via magnetic stirrer for 30 minutes at ambient temperatures. The mixture was centri- fuged at 5,860 X g for 5 minutes in a Sorvall Superspeed RC2-B automatic refrigerated centrifuge set at 2°C. Following centrifugation, the super- natant was filtered through 11 cm S & S No. 597 filter paper and the filtrate transferred to a cuvette for an optical density measurement at 475 nm with a Beckman DU Spectrophotometer. Kjeldahl Protein Analysis The American Instrument Company (1961) Micro-Kjeldahl method was used with modifications. The sample was placed in a Kjeldahl flask with 30 approximately 0.5 g of sodium sulfate, 1 m1 of 10% copper sulfate and 7 m1 of concentrated sulfuric acid. Sample size varied with the type of system used. For the emulsion samples diluted in 0.1M citric acid, 0.2M dibasic phosphate buffer, pH.5.5, a 5 ml sample sufficed; however, for the model system 5 ml of sample and 5 ml of deionized water were used. Glass beads were added to the flask.and the mixture was digested over electric coils for 2-4 hours. Occasionally the flask was swirled to insure the complete digestion of the sample. The flask and its contents were cooled after digestion and approximately 15 ml of water (deionized) was added. Distillation was accomplished into ‘ 10 m1 of 2% boric acid and 2 drops of bromocresol—green indicator solution in a 125 ml Erlenmeyer collection flask. The digestion flask and the collection flask were pro- perly situated on the apparatus prior to the addition of 15 m1 of 40% cold sodium hydroxide to the digestion flask. Steam was directed from the boiling water flask through the closed system to the Kjeldahl flask. Distillation proceeded for 7 minutes. The collection flask was then lowered and the condenser was flushed for 4 minutes. The 2% boric acid solution was titrated to the green end point with 0.02N sulfuric acid. For protein calculations, a factor of 6.25 was used to determine the percent protein from the nitrogen analysis. Moisture Determination The A.0.A.C. (1970) method of drying 2.5-3.0 g of sample for 16-18 hours at loo-105°C was used. The I saple was The 1 afinal c. acrylanidt (Barton 5‘: urea puri Polmeriz 891- Wat the gels Three dro dye. To PIOtein e the Contr Which had Murrent EIECtl'Oph readded t' by Sliding 81385 tub 31 Fat Determination The A.O.A.C. (1970) ether extraction (Goldfisch method) of the dried sample was used. Disc Gel Electrophoresis The method of Davis (1965) was modified. The running gel contained a final concentration of 6.5% cyanogum which replaced the acrylamide-bis- acrylamide used by Davis (1965). The spacer gel contained 5.0% cyanogum. (Horton gt al., 1967) and both the running and spacer gels contained 7.0M urea purified over MB-3 resin. The gels were placed in glass tubes and polymerized by flourescent light for 20 minutes after the addition of each gel. water was used to level the minuscus of the gels and removed after the gels polymerized. The tank buffer was a Tris-glycine buffer, pH 8.5. Three drops of brom thymol blue were added to the buffer as the tracer dye. To the tap of each gel, 0.1 m1 of the salt soluble or water soluble protein extract with added sucrose was added via a syringe, these were the control samples. A 0.1 m1 sample of salt and water soluble protein which had been reacted with AO-12(0.13 mg/ml) represented the test sample. A current of 2.5 ms per gel was maintained for protein electrophoresis. Electrophoresis was terminated when the leading brom thymol blue band reached the end of the gel. The gel was then removed from the glass tube by sliding a hypodermic needle along the interface between the gel and glass tube wall and forcing a photovolt solution into this space. The g Slack NBR Tse gel wa then held Data grass (Hit of descrir sented da Tet: progr 1.53:1 pro for basic squares, Zion accc to deter: error mea Coefficie also anal ETESSIOn Variables Vthh WOU data VOUL deified f1 An e} binding PT 19.,“ of : “guinea, ”3588 Of design We: 32 The gel was then placed into a test tube and immersed in 0.4% Buffalo Black NBR dye solution consisting of water, methanol and acetic acid (5:5:1). The gel was given twoSdestaining washes in 7% (v/v) glacial acetic acid and then held overnight in the same solution. Statistical Analysis Data were analyzed via standard computer center library statistical pro- grams (Michigan State University Agricultural Experiment Station STAT Series of descriptions). The programs applied a least square analyses to the pre- sented data and calculated linear models of regression equations for the data. Two programs were utilized on the CDC-3600 computer. They were the LSADD and LSDEL programs. The LSADD (least squares addition) program analyzed the data for basic statistical parameters (coefficients of correlation, means, sums of squares, etc.) and then added dependent variables to the least squares equa- tion according to a predetermined criterion (Fbi’ anéF-test for each parameter to determine if the addition of this variable to the equation would lower the error mean square of the AOV for the regression data while increasing the coefficient of determination(R2)). The LSDEL (least square deletion) program also analyzed the data for basic statistical parameters then calculated a re- gression equation utilizing all of the dependent variables. The dependent variables were then deleted according toapredetermined criterion, e.g. Fbi’ which would determine if the error mean square of the AOV for the regression data would increase and the R2 decreasesignificantly if a parameter were deleted from the equation. An experimental design was proposed only for the sausage emulsion-dye binding phase of this study. This design was a 3x2x2 factorial for the fat level of the emulsion, the dyes and two methods, respectively. In calculating significance for the results u-levels of 0.05 and 0.01 were used. The other phases of this study entailed empirically derived data and no experimental design was proposed for them. RESULTS AND DISCUSSION Dye Parameters Tsugo §£_§l, (1966) found Acid Orange II (AD-7) and Diacid Light Green GS (AG-25) suitable for mass protein determinations. Because they were available in this country, they were used in this study. Orange G (AD-10) and Brilliant Orange (AD-12) were shown by many investigators to be very effective for use in protein determinations. These four dyes I also provided an opportunity to ascertain the effect of molecular structure upon the dye binding mechanism. No.0,SQ-Mfl g “30,30 lflaflo Acid Orange-7 Acid Orange-10 30,». fl On» 9 i“ ' 5 m Jug“. Acid Orange-12 Acid Green-25 Figure 1. Molecular structure of azo and anthroquinone dyes selected for protein determination study. The difference in the molar optical density is also attributable to the molecular structure of the dye (Torten and Whitaker, 1964). 33 34 pH Since all of the dyes are "acid dyes", the reaction environment should be acidic. The acid medium permits the formation of the salt linkage between the acid group of the dye and the basic groups of the protein amino acid residues (Cockett and Hilton, 1961). It is the sub- stitution of polar groups such as the sulfonic acid groups present on the dyes used in this study that enable combination with the polar sites on the protein (Roberts and Caserio, 1965). Therefore, adherence to Beer's law in environments of pH 1.9 and 3.0 is essential since this medium is necessary for combination between the dye and protein. In figure 2 at pH 1.9, the slopes of the plots between the optical density of AO-12, AG-25 and AO-7 and the dye concentration (pg/ml) for these respective dyes diverge from a linear trace at 10 ug/ml. For the dye A0-12, the linear trace terminates at a dye concentration of 12 ug/ml. No effect due to molecular structure is apparent at pH 1.9. However, at pH 3.0 (figure 3) the departure from linearity is fore- stalled for AO-lO and AO-12 until a dye concentration of 80 ug/ml is attained. This represents an eight fold increase in the effective con- centration of the dye for an increase of 1.1 pH units in the environment. For AO-7 and AG-25, the same increase in the pH resulted in an extension of the linear portion of the plots to concentrations of 60 and 100 ug/ml. The dye AO-7 which exhibited the smallest increase of effective dye con- centration had its sulfonic acid substituent on the benzyl moiety of the dye molecule, the other dyes had substituents on the naphthyl or methylated benzyl moiety. However, increasing the pH to 3.0 served to increase the apparent effective concentration of all four dyes. Figure 2. Optical density 35 T ‘ A0-7, 480 um I AO-lO, 475 nm . AO-lZ, 475 nm 4 AG-25,.r 645 nm 6 7 8 9 10 11 12 Dye concentration (pg/ml) Adherence of dyes to Beer's Law at relevant wavelengths, pH 1.9, 0.1M citric acid, 0.2M dibasic phosphate buffer. :&VUCQT Faurvar—C Figure 3 36 3.0 _ ‘ A0-7, 480 nm II A0-10, 475 nm 2.8 - O A0-12, 475 nm 4: AG-25, 645 nm 2.6 ‘ 2I4 b 2.2 b 2.0 r 104 b - 1.2 P Optical density o a a .o a :n -n on Dye concentration (pg/ml) Figure 3. Adherence of dyes to Beer's Law at relevant wavelengths, pH 3.0, 0.1M citric acid, 0.2M dibasic phosphate buffer. 37 Absorbancy Index (Extinction Coefficient) A comparison of the absorbancy index values (extinction coefficients) for the dyes mentioned above demonstrated agreement between various studies and the relation of the dyes used in this study to previous work. There appeared to be agreement between our work and that of Tsugo 35 11° (1966)(Tab1e 1). Ashworth and Chaudry (1962) present a value for AO-10 that was in agreement with the other two studies yet for the absorbancy index of Amido Black 10B (AB-1) they strongly disagreed with Tsugo gt a1. (1966). This difference could indicate a difference in dye purity or that a different dye was used by each investigator. Table 1. A comparison of absorbancy index values (extinction coefficient) Tsugo g£_§l, Ashworth g£_§1, Dye ,pH 1.9 ,pH 1.9 1.9 3.0 Acid Orange 7 40.8 -— 38.5 51.7 Acid Orange 10 50.2 46.9 50.6 36.1 Acid Orange 12 -- -- 17.9 18.5 Acid Green 25 17.4 -- 15.1 9.6 Acid Black 1 49.6 81.5 -— -- In a comparison of the absorbancy index values at pH 1.9 and 3.0 differences were apparent. At pH 1.9 the absorbancy index values were close to the values reported in the literature; however, it became apparent in working with this ,system that there was some variation in the values. The standard deviation of the values for AO-7, AO-lO, AO-12 and AG-25 were $2.28, $1.32, i1.66 and $2.55, respectively. At pH 3.0 the standard deviation for AO-7 increased to +2.83 and it decreased slightly for AG-25 to +1.70. However for AO-lO and AO-12 these values 38 were reduced to +.58 and +0.10, respectively. Thus it appears from this finding that AO-10 and A0-12 would be the most useful for protein deter— minations since they would minimize the variation due to the dye itself. Retention by Filter Paper Since filtration of the free dye is an essential step in the dye binding procedure, it is imperative that the amount of dye retained by the filter be known. Table 2 illustrates the amount of dye retained by the filter paper in comparison with results reported in the literature. The large discrepancy between the percent retention reported by Tsugo gt El: (1966) for AG-25, 4.3% and the value obtained in this study 49% is irreconcilable, but could represent impurities in the dye, precipita- tion or some other instability. All of the values reported in this study Table 2. Percent retention by filter paper (Whatman No. 40) of azo and anthroquinone dyes at pH 1.9 and 3.0 Dye Tsugo §£_§l,, pH 1.9 pH 1.9 pH 3.0 Acid Orange 7 16.5 24.0 32.0 Acid Orange 10 2.9 3.0 3.9 Acid Orange 12 -- 29.0 15.5 Acid Green 25 4.3 49.0 45.0 represent an average of five values. The effect of increasing the pH to 3.0 is evident. The amount of dye retained by the filter for dyes A0-7 and AO-lO increased by 30%; whereas, the amount of dye retained by AO-12 and AG-25 decreased 45% and 8%, respectively. The amount of dye retained by AO-lO was acceptable and compares favorably with values 39 reported in the literature. The values for A0—7 and AG—25 were completely unacceptable since one-third to one-half of the dye was retained by the filter paper. A0-12 had a high degree of retention in the filter paper but still less than A0-7 and AG-25. It appeared that AO-10 and A0-12 would be the two best dyes to uti- lize in this study from the dye parameters examined. Thus the dyes AO-lO and A0-12 were a used in the protein determination; however, whenever possible the azo and anthroquinone dyes A0—7 and AG-25 would also be tested. Dye Binding and Protein Systems Hammerstein's Casein The ability of the dye to bind particular proteins was ascertained by reacting the dyes with Hammerstein's casein. Casein was selected because of the success with which the dye binding procedure had been applied to milk-protein systems(Ashworth, 1957, 1960, 1966; Udy, 1954, 1956; Dolby, 1961). Figure 4 illustrates the effect that the concen- tration of the dye has upon its ability to bind with proteins in solu- tion. At a concentration of 1 mg/ml of dye, only A0-12 and AG-25 bound to the protein in sufficient quantities to be detected by a spectrophoto- meter. The remaining two dyes may bind with proteins but not enough of the free dye was removed from the solution to be measureable. This may be attributed to the high molar optical density of the two dyes. Torten and Whitaker (1964) encountered the same difficulty with Amido Black 10B (AB-1). However with a ten fold more dilute concentration, it was 40 3.0 r49 ' iKiF;_____:i‘.| ‘_1.I 2.8 F 2 6 b I A0-7, 480 nm 1 I AO-lO, 475 nm 2-4 ‘ Q A0—12, 475 nm 4- AG-25, 645 nm 2.2 L 2.0 P h1.8 h . {J 'H 2 1.6 .- Q) 'U 7.; 1.4 - U H U D 8‘1-2 100 b :.~~~~ 008 b *~~ .‘ ‘ fit 0.6 L '~.*l‘..‘n~ ‘. “‘ ‘ 0.4 p» {I 002 p -- ------+--------; .5 6 7 8 9 10 11 12 13 14 15 16 Casein concentration (mg/m1) Dye concentration 1 mg/ml H 0.1 mg/ml .---.. Figure 4. The effect of two different dye concentrations upon the free dyes in a casein solution. 41 apparent that the protein concentration was far greater than that of the dye. In this situation most of the dye has been removed from the solu- tion. However, AO-lO appeared to be able to bind to the protein at this dye level but not at the higher level. Through the use of trial and error techniques the proper dye con- centration for protein binding was obtained for all four dyes. From figure 4 it is apparent that the optimum concentration for A0-12 to bind with casein was approximately 1 mg/ml. Similar methods disclose that the optimum dye concentration for A0—10 was below 0.1 mg/ml. For the remaining dyes AG-25 and A0-7, the optimum concentrations appeared to be 1 mg/ml and less than 0.5 mg/ml, respectively. Bovine Serum Albumin (BSA) Through the modification of a procedure developed by Daniel and Weber (1966), the average number of dye molecules bound per protein molecule could be empirically determined. Their method involved the use of BSA as the binding medium for fluorescent dyes and measurement of the fluorescent spectra. Our modification continues to use BSA (molecular weight - 66,000) as the binding medium and the binding of the azo and anthroquinone dyes was measured in the visible spectra. The average number of dye molecules bound per protein molecule, 5, was represented by the equation below. Where x0 = the initial concentration of the dye, x = the concentration of the dye remaining (free dye) and po = the protein ‘5 = [x0] - [ x ] (1) [Po] 42 concentration. If the molarity of the dye and the protein solution was known, then 3, the average number of dye molecules bound per protein molecule could be determined. The effect of saturation by the protein upon the number of dye molecules bound to the protein was evident for each of the four dyes (Figure 5). Initially, when there was an excess of dye in relation to protein, both binding and adsorption took place. The values for A0-7 and AO-lO, AO-12 and AG-25 in an environment where the dye concentration was 10 times greater than the protein concentration were 319, 120, 350 and 230 molecules of dye/protein molecule. The E values continued to decrease asymptotically until the protein concentration was 2.5 times greater than the dye concentration. Increasing the protein concentra- tion further only serves to dilute the amount of dye present in the system and would indicate that there was no longer any significant binding between the dye and protein. Each curve had a region where the 3 value appeared to be constant within a narrow range with respect to the increasing protein concentration. This region of the curve would allow us to estimate the number of dye molecules bound by the protein. In the BSA system the -* 1. values for AO-7, A0-12, A0-12 and AG-25 were 37.0, 13.0, 46.9 and 34.9 molecules of dye/protein mole- cule, respectively. Assuming the optimum binding values to be correct, some deductions could be made. The azo dyes A0-7 and A0-12 were positional isomers yet approximately 27% more A0-12 dye molecules were bound than A0-7 molecules. Furthermore, as figure 1 depicts, A0-10 was structurally similar to 43 .mohm oaoaaavonnuom was one onu moans A0-7=>A0-10. Thus it seems that an azo dye with a single sulfonic group had a greater affinity for the binding sites on the protein than a dye with two sulfonic groups. On the AO-lO molecule, the possibility of the sulfonic groups interfering with each other at the binding site offered an explanation for its poor affinity toward the protein. It also appeared reasonable that the location of the sul- fonic group on the dye molecule influenced its ability to bind to the protein. For A0-7 the sulfonic substituent was located on the benzyl moiety of the dye molecule; whereas, A0-12 had the substituent located on the naphthalene moiety. The binding of the anthroquinone dye AG-25 appeared to lend support to the inference concerning the interference of two sulfonic groups in close proximity. AG-25 possessed two sulfonic substituents which were widely separated in comparison to the substituents on A0-10. The two sulfonic substituents could bind to sites on two different proteins or to two sites on a single protein molecule. The azo dye AO-lO had an ability to bind to the protein that was 37% less than the ability of the anthroquinone dye AG-25. Figure 6 depicts a plot of the log of the protein concentration as it varied with the log 5. Similar slopes of the linear plots would indicate similar affinities or degrees of binding between the dye molecule BSA concentration (M) 45 ‘ A0—7, 480 um I AO-lO, 475 nm 2 I A0-12, 475 nm 4' AG-25, 645 nm 10 1 2 3 4 5 6 7 3 9 1o 2 3 4 _ 101 102 n = average number dye molecules bound/protein molecule Figure 6. A log—log plot of BSA concentration vs 5, the average number dye molecules bound/protein molecule for all four dyes. 46 and the protein. The slopes of the data plotted for A0-12 and AG-25 were very similar, while the slopes of the data plotted for A0-12 and A0-7 differed slightly. An analysis of variance performed on E values of the four dyes with respect to their ability to bind proteins demonstrated what is depicted in figure 6, that only AO-lO is significantly different from the other dyes (P S 0.05) in its ability to bind proteinsr The remaining dyes demonstrated no significant difference among the dyes themselves to bind proteins. Meat-Lipid System A procedure was developed to ascertain the ability of the azo and anthroquinone dyes to react with a meat-lipid system. Torten and Whit- aker (1964) worked with fresh meat samples, Ashworth (1966, 1970) dealt with fresh and processed meats and Bunyan (1959) investigated meat meals. Bunyan and Woodham (1964) and Torten and Whitaker (1964) found that the dye binding procedure was feasible but highly variant in reactivity. Ashworth (1966, 1970) encountered little difficulty with the method. The procedure attempted in this study utilized mixtures of lean meat, beef round (lo-20% fat), intermediate fatdmeat (30-40%) plate beef and beef fat (93% fat). These components were ground and combined in various proportions which yielded a range of protein percentages from 4.0 to 22.4%. The Ashworth method was applied to this system using all four dyes and two concentrations, 200 pg/ml and 10 ug/ml. The 200 ug/ml dye concen- tration proved to be too high and registered little discernible difference; 47 therefore, only the 10 ug/ml dye concentration would yield reasonable results. The erratic nature of these results is illustrated in figure 7. A similar pattern of binding was observed for each dye in relation to the protein concentration. There also appeared to be two points of maximal dye binding by the protein for all four dyes at 10 and 20% protein. These points of maximum dye binding may represent transition points in the type of binding taking place between the dye and the protein, e.g. from adsorption of the dye molecules on the protein to the formation of a salt linkage or ionic bonding as the protein concentration increased. The erratic nature of the dye binding appeared to support the con- clusions of Torten and Whitaker (1964) and Bunyan and Woodham (1964) that the dye binding procedure exhibited a wide variation in binding as the protein concentration increased. That is, for a product with a broad protein range (> 10%) the dye binding technique would yield irregular results. The utilitarian aspect of this method may lie in the two linear portions of the graph but these trials demonstrated the variation that was present in an attempt to apply this method to a fresh meat product. Torten and Whitaker (1964) stated that the reaction between the dye and the protein involved multiple equilibria affected by the dye and pro- tein concentrations. The effect of the multiple equilibria upon dye binding might explain the results depicted in figure 7. At this point it was deemed unnecessary to proceed with all four dyes. Statistical tests demonstrated that no significant difference 48 0-9 ; AO-7, 480 um I AO-lO, 475 um I A0-12, 475 nm «b AG-25, 645 nm 0.8 0.7 0.6 0.5 Optical density 0I4 0.3 at , 4, 7 d + 0.2 , / OIl + A j I A l I l l n A i 2 4 6 8 10 12 14 16 18 20 22 24 Protein concentration (Kaeldahl, mg/ml) Figure 7. Meat-lipid system. Effect of protein concentration upon free dye concentration, (0.1 mg/ml dye concentration for all four dyes). 49 (P 5 0.05) existed between AO-12, AO-7, and AG-25 but that A0-10 was sig- nificantly different from the other dyes in protein binding ability. Coupled with the relative similarity that exists in the protein binding ability of these dyes was the fact that AG-ZS was inadequately soluble at pH 3.0 in the citrate-phosphate buffer. For these reasons and those cited earlier, further study was limited to the two azo dyes, A0-12 and A0-10. These two dyes offered “ essentially different affinities under similar conditions which would assist in determining the efficacy of the dye binding technique and the mechanism of dye binding. Meat and Protein Component Systems Protein Component Systems With the procedure developed by Helander (1957), various classes of proteins were extracted from ground meat. These fractions consisted of the sarcoplasmic proteins (water soluble), the myofibrillar proteins (salt soluble) and connective tissue proteins (stroma)(Lawrie, 1966). These three classes of protein, as extracted solutions (see Materials and Methods .were exposed to A0-12 (1.3 mg/ml) and AO—lO (0.04 mg/ml). These proteins were diluted with 0.03M phosphate buffer to provide different concentrations to expose to the dye system. The results of exposing A0-12 and AIFlO to the various protein classes is illustrated in figure 8. For A0-12, all three protein classes bound increasing amounts of dye as the protein concentration increased; however, for the stroma and sarcoplasmic proteins a point of maximum dye binding was attained and for concentrations above this point the amount 50 1.3 o o o I O 0 MYOfibrillar (KI) Sarcoplasmic 1.2 --..-O Stroma I AO-lO 1.1 . A0-12 1.0 nm ,0.9 O (I) O \1 Optical density at 475 c> c: u: as O b 0.3 OI2 0.1 ‘ ‘~. .000I0000...oooooooooout? 0.8 9.0 1.0 1.0 1.2 Protein concentration (Kjeldahl, mg/ml) Figure 8. The dye binding ability of three different protein fractions and the dyes, AO-12 and A0-10. 51 of dye bound decreased with increasing protein concentration. Because of the vigorous shaking that this method entailed during the exposure of the dye to the protein, the relative insolubility of the stroma proteins did not affect their ability to react to the dye. The relationship be— tween the Optical density and the protein concentration for myofibrillar proteins was curvilinear. The absorbance decreased with an increase in protein concentration and approached the protein concentration asymptoti- cally. This indicated that as the protein concentration increased the amount of dye bound increased and approached a point of maximum and con- stant dye bound. A0-10 exhibited very little affinity for any of the meat protein classes. ‘ . However, if greater concentrations of A0-10 were used the amount of dye bound by the proteins can not be measured. Table 3 presents the average DBC (U8 dye bound/mg protein) of each of the dyes for the three protein fractions and it illustrates the apparent saturation of A0-10 by the protein concentration. Table 3. The effect of muscle protein type upon the DBC (ug dye bound/mg protein i s.d., n = 30) of A0-10 and AO-12. Protein fraction Dye Sarcoplasmic Myofibrillar Stroma A0-10 3.6 i .91 3.2 i .90 1.8 i .6 Statistical treatment dischased that there was no significant differ- ence (PiS 0.05) in the DBC of A0—12 for any of the protein fractions. The lack of significance was found for the DBC of AO-lO among the three 52 protein fractions, however, there was a significant difference in reactivity between the dyes (P i 0.01) in each protein fraction. It was noted that A0-12 with only one sulfonic group showed a DBC approximately one order of magnitude greater than that of AO-lO with two sulfonic groups, but there was 30x more A0-12 than A0-10. Effect of Anions Upon Dye Binding The effect of both chloride and iodide salts upon the reactivity between the dyes and the salt extracted myofibrillar proteins was investi- gated. Figure 9 demonstrates the curvilinear relationship was retained in the A0-12-protein system regardless of the anion of potassium salt (u - 1.4) used to extract the protein. The DBC of A0—12 in KCl and K1 extracted myofibrillar proteins was 40.5 and 36, respectively. For A0-10 the DBC was 3.8 and 3.2 for the K01 and K1 extracted proteins, respect- ively. There was no significant difference in DBC attributable to the type of salt used to extract the proteins. Effect of Dye Binding on Disc Gel Electrophoresis Utilizing the sieving ability of the polyacrylamide gel (Davis, 1965) and the charge separation abilities of the electrophoretic method (Orn- stein, 1965) the gel electrophoretic method was used to evaluate the effect of dye binding upon the various protein classes. It was assumed that since the dye bound with the protein via ionic bonding the "charge" on the bound protein would be altered by the extent of the affinity of the dye for the protein binding sites. The direction of mobility in the 53 .. .. .. _ Myofibrillar (KI) 80 Myofibrillar (KCl) I A0-12 . AO-lO 70 P 0‘ O l U 0 I DBC (pg dye bound/mg protein) b c: b.) O 1 20 - 10 ' I1 .2 .3 I4 .5 I6 .7 I8 .9 1.0 1.1 1.2 Protein concentration (Kjeldahl, mg/ml) Figure 9. The effect of anion type used in myofibrillar protein extraction upon the DBC of the azo dyes, A0-10 and A0-12. 54 electrophoretic device is toward the anode; therefore, the addition of dye would alter the protein "charge", making it more negative thus in- creasing the rate of migration of the protein. Additionally, the binding of the dye molecules to the protein should increase the effective diameter of the protein and contribute to increasing the migration of the protein. Figure 10 illustrates the effect of AO-12 upon myofibrillar and sarcoplas- mic proteins. The most notable effect was the apparent disappearance of band 2 (Rm = 0.07 to 0.22) from the sarcoplasmic protein preparation. The effect of A0-12 upon the myofibrillar protein preparation involved the apparent disappearance of two bands (Rm 0.46 and 0.71) from the K1 extracted pro- teins and three bands (Rm 0.36, 0.46 and 0.71) from the K01 extracted proteins. Both types of salt soluble extracts exhibited an increase in optical density in the upper band (Rm 0.14). This band represents the slowly migrating myosin band and the band at am 0.46) is probably trapo- myosin (Rampton, 1971). From the criteria established initially, it appeared as if the dye possessed a greater affinity for myosin in the myo- fibrillar proteins. The components in the other bands underwent some diminuition, possibly some proteins were retained in the Rm 0.14 band. Apparently the increased diameter had a greater effect upon the proteins than the increased electrOphoretic mobility. The constituents bound in the sarc0plasmic proteins were unknown. Model Emulsion Systems A model emulsion system using sarcoplasmic and myofibrillar proteins in a dilute solution was studied. Aside from ascertaining the DBC, the ability of 55 .mGOHuomum afiououm saunas umHooHuumm mo moauuomoue ofiuouonmonuomao can con: malom out one can mo uoommo oeH .oa ouswfim ohm ohm ohm Houuaoo nuHB Houuaoo cuw3 Houuaoo Spas o.H m.o w.o 1 5.0 1 0.0 1 1| m.o owouuooflo o>fiufimoe + «.0 1 m.o 1 N.o 1 H.o 1 J L 0.0 I omouuooao o>Humwoa I HE J— 1 I II” AHuMV maflououm aHMv mafiououm msfimuoum mason em uwaaauafimomz anHHunamo%z oaammamoopmm 56 a model system to duplicate actual sausage emulsions was also studied. Previous workers (Swift gt_al., 1961; Hegarty g£_al,, 1963; Carpenter and Saffle, 1965; Trautman, 1964; Saffle, 1968; Ivey g£_al,, 1970) worked with model systems. Saffle in a review cautioned workers on the relevance of such model system data to the actual system or product. Previous studies have cited the myofibrillar proteins as the primary constituent of the protein membrane surrounding the lipid material in a sausage emulsion. The myofibrillar proteins were implicated by Hansen (1960), Swift gt_§l, (1961, 1968), Carpenter and Saffle (1964), Saffle (1968), Hegarty g£_§1, (1963), Borton 25 21° (1967) and Van Eerd (1971). Other investigators have demonstrated that the sarcOplasmic proteins do participate in the emulsification process but not to the extent of the myofibrillar proteins. These proteins did bind the dyes in a linear manner up to a protein concentration of 0.7 mg/ml. Hansen (1960), Swift st 21. (1961), Carpenter and Saffle (1964), Hegarty g£_§1, (1963) and Van Eerd (1971) all cite the relative inferiority of the sarcOplasmic proteins in emulsifying capacity in comparison with the myofibrillar proteins. The emulsifying capacity of the sarcoplasmic proteins could be enhanced by the addition of 3% NaCl to the mixture (Swift and Sulzbacher, 1963). The stroma proteins were found to contribute very little to the emulsification of the lipid material in an aqueous system; therefore, they were not used in this study. Van Eerd (1971) noted the insoluble proteins appear to contribute to a very limited extent when mutton was used for sausage production. 57 Effect of Lipids on Dye Binding The model emulsion system was used to determine if lipid material pre- sent in an emulsion would bind azo dyes. If the lipid material does bind the dyes, it will bias the protein measurement and would have to be accounted for in the protein determination. Table 4 illustrates that the presence of lipid material in the myofibrillar protein emulsion system had little apparent effect upon the DBC of the proteins. Table 4. The effect of varying the lipid concentration upon the DBC of myofibrillar beef proteins for two azo dyes at two concentra- tions at a pH 5.0 in a 9% protein solution. Concen. Dye binding capacity (g_dye bound/mg_protein) Dye ug/ml 20% fat 25% fat 30% fat 35% fat 40% fat A0-10 80 4.5 4.5 4.5 4.5 4.5 A0-10 200 — - - - - A0-12 80 13.6 13.4 13.2 13.2 13.2 A0-12 200 33.8 33.9 33.9 34.1 34.1 It was apparent from the constancy of the DBC values that varying the lipid concentration had little effect. This range of fat would ap— proximate the range anticipated in a sausage product with a median of 30% Which was the limit set by current Federal Regulation (MID Reg. 319.180). Since the myofibrillar proteins form the protein membrane around the lipid material, it was imperative that little interference from lipid material be present. The increase in DBC for A0-12 was a direct reflection of the increase in the concentration of the dye. In- creasing the dye concentration 2.5 fold caused a commensurate 2.5 fold increase in the amount of dye bound by the protein membrane system. 58 No definitive statement was possible concerning the mechanism of binding between the dye and the protein from these data. Both the adsorp- tion mechanism and the selective binding hypothesis could explain the increase in DBC with increased dye concentration. This was possible since even the greater dye concentration (0.2 mg/ml) was still less than 10% of the concentration of the protein (5.7 mg/ml). Myofibrillar Proteins Figure 11 illustrates the effect of varying the protein concentra- tion had on the DBC of A0-12 and A0—10 in a model system with a constant lipid level of 30%. It is apparent from the graph that a curvilinear relationship existed between the DBC and the protein concentration. This relationship was not species specific since both beef and pork myofi- brillar proteins demonstrated the same relationship. However the DBC of beef myofibrillar proteins appeared to be less than that of the pork myofibrillar proteins. The discrepancy between beef and pork with respect to the DBC was from 6.3 to 1.2 units lower for beef than for pork with an average distance of 2.1 units. This difference was not significant and appeared to support the findings of Ashworth (1970) that pork bound the same amount of dye as beef. However he worked with meat in a patty form rather than an emulsion. The dye AO-lO did not exhibit the same relationship as A0-12 in the protein-lipid system. There was a reversal with beef myofibrillar proteins demonstrating a higher DBC than the pork myofibrillar proteins. This is Figure 11. 250 N O O I [—1 U1 0 I 100 f DBC (Hg dye bound/mg protein) 59 I fi‘ — Pork ti - —- - Beef 50 ' '- t \ \ ' A0-10 t -u‘-..‘ g.— ‘ J—- ‘ ‘ ‘ ‘ Li .5 1.0 1.5 2.0 2.5 3.0 3.5 Protein concentration (Kjeldahl, mg/ml) Comparison Of the dye binding capacity of beef and pork myofi- brillar proteins in a model emulsion system with A0-12 (1300 ug/ml) and A0—10 (40 ug/ml). 60 the reverse of the relationship demonstrated by A0-12. The discrepancy between the DBC values of the myofibrillar proteins for both species was higher than for A0-12, with a range and average of 8.7 to 1.2 units and 3.3 units (DBC), respectively. This reversal may have been the result of a concentration effect or configurational differences between the molecules. However earlier tests have demonstrated that these dye concentrations for A0-12 (1.3 mg/ml) and A0-10 (0.04 mg/ml) were the most effective for each dye. This does not exclude the influence of the dye concentration from consideration but the extenuating circumstances should be noted. The curvilinear relationship in figure 11 is interesting when com- pared to figure 9. Figure 9 represented the ability of the dyes to bind proteins with no lipid material present and in an excess amount of dye; whereas, figure 11 represented a protein-lipid emulsion in which the dyes were not present in excess. In figure 9 the slope decreased be— tween the protein concentrations of 0.5 to 1.2 mg/ml and appeared to be approaching a constant DBC value. In both figures the curvilinear relationship was apparent. However for A0-10 at the same protein concen- trations the relationship shifted from a near linear (constant) relation- ship in figure 9 to a curvilinear relationship in figure 11 with lipids present. Sarcoplasmic Proteins The sarcoplasmic proteins of beef and pork were emulsified in the model system and exposed to the dyes. These data are presented graphi- cally in figure 12. Difficulties in precipitating these proteins were 61 150 _ Pork 140 —-—- Beef 130- 120r 110' O O \O O O‘ \l O O t I DBC (pg dye bound/mg protein) u: a: c: c: 40’ 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Protein concentration (Kjeldahl, mg/ml) Figure 12. Comparison of the dye binding capacity of beef and pork sarco- plasmic proteins in a model emulsion system with A0-12, 1300 ug/ml) and A0-10 (40 ug/ml). 62 anticipated because of their solubility but the binding of the dye and the pH differential readily precipitated the sarcoplasmic proteins. The relationship between the DBC and the protein concentration was less consistent for this system than for the myofibrillar proteins; how- ever, a curvilinear relationship was apparent. This inconsistency was present for both dyes. The effect of the emulsification of these proteins was obtained by comparing the DBC values for the emulsified and unemulsified sarcoplasmic and myofibrillar proteins. The emulsified beef myofibrillar proteins had DBC values that were 13% greater than the unemulsified proteins for the dye A0-12 and 30% greater for A0-10. The emulsified beef sarco- plasmic proteins demonstrated an increase in the DBC values over the unemulsified proteins of 1% and 17%, respectively, for A0-12 and A0-10 dyes. The unemulsified pork proteins were not studied. The study with beef proteins did demonstrate that the emulsification process appeared to increase the DBC of the proteins, but the presence of lipids was found to have no effect upon the DBC. The increase in the DBC may have been due to the rearrangement of the protein structure to expose more polar groups to the dye molecules. The dye A0-12 was significantly different in its dye binding capacity (P S 0.01) than A0-10. However, if the correlations between the DBC and the protein concentration were compared there was not a significant differ— ence between the two dyes. The correlation coefficient (r) for AO-12 was 0.6766 and for A0-10, r - 0.6643. When the bound dye was correlated with the protein concentration, the results for A0-12 and A0—10 were 63 r = -0.0438 and 0.3814, respectively. If a similar comparison was made between the reciprocal of the DBC (DBC-1) and the protein concentration a significant difference was found. The correlations for A0-12 and A0-10 were r = 0.9987 and 0.7657, respectively. A regression equation utilizing the most significant parameters of the dyes with respect to the protein concentration was calculated. Y 1 -3.07 + 0.123 (DBC-1) + 0.002 (BD) (2) Where Y = protein concentration, and the coefficient of determination (R2) for the equation was 0.9993 for (P < 0.001), and the standard error of the estimate was 0.038. Sausage Emulsions and Dye Binding The use of sausage emulsions would provide a practical application of the dye binding technique and a means of assessing the validity of the model emulsion system. Torten and Whitaker (1964) and Bunyanamd Woodham (1964) reported the feasibility of this technique for evaluating the protein content of meat products under limited circumstances--narrow percentage range of protein and finely ground products. It was determined that the DBC values of both the processed (cooked, smoked and chilled) emulsion and the unprocessed (raw) emulsion be ascer- tained. Two methods of dye binding were used, the method suggested by Ashworth (1970) and our modification. Dilution of the Sample Both methods entailed the dilution of the sample in a buffer system; therefore, the correct product dilution had to be determined. Torten and Tnitaker and dye aulsior literati a value This 01 T1 buffer tratio The re 13 gm for ti Table and 1 Stand 1Vely SiSt 64 Whitaker (1964) and Sherbon (1970) found that the DBC varied with protein and dye concentrations. Only values on the dilution of the unprocessed emulsion will be reported here since no comparable values exist in the literature. Ashworth (1966, 1970) and Torten and Whitaker (1964) reported a value for frankfurters (processed emulsion) of 15 gm/250 ml buffer. This dilution was used in our study. The unprocessed emulsions were diluted in 100 ml citrate-phosphate buffer, pH 5.5, as follows: 3.75 gm, 7.5 gm and 15 gm. The dye concen- trations were 1.3 mg/ml and 0.04 mg/ml for A0-12 and A0—10, respectively. The results tabulated in Table 5 indicate that either the 7.5 gm or the 15 gm sample was sufficient to supply the proper amount of reactive sites for the dye molecules. The standard deviations for A0-12 with the 7.5 gm Table 5. The effect of sample dilution upon the DEC of protein in a sausage (unprocessed) emulsion. Gm of sample in Prot. con'n Bound dye (g/ml) DBC (mgfiprot.) 100 ml buffer mg/ml A0-10 AO-12 A0-10 A0-12 4.0 4.2 25.6 1267 6.0 308 7.5 9.2 24.7 1271 2.6 139 15.0 19.1 20.1 1296 1.1 68 and 15 gm samples were $6.8 and i8.0, respectively. AO-lO exhibited standard deviations for 7.5 and 15 gm samples of $0.41 and 10.89, respect- ively. The dye A0-12 had the lower pr0portionate standard deviation, since the lower DBC of A0-10 would yield a lower standard deviation. The 7.5 gm/100 ml buffer dilution was used since it appeared to be most con- sistently sensitive to protein variation and was nearly equal to the 65 dilution recommended by previous investigators for the processed emul- sion. Torten and Whitaker (1964) emphasized that the efficacy of the dye binding method was highly dependent upon both the protein and dye con- centrations. From Table 5 it is evident that an increase in protein concentration of 218% and 455% over the initial concentration resulted in corresponding decreases in the DBC of A0-12 of 55% and 78%, respect- ively. Similar decreases in the DBC of AO-lO occurred, they were 57% and 82%, respectively. DBC and Sausage Emulsions When the proper dilution was ascertained, the ability of the dye binding method to discriminate between different protein levels in the processed and unprocessed emulsion could be determined. Figures 13 and 14 compare the DBC and the protein concentration for both dyes and both states of the emulsion. For both dyes, it was apparent that for the un- processed product all three fat levels (24, 30 and 36% fat) possessed a higher DBC than the corresponding processed product. Yet at each fat level the processed product had a protein content that was 1-2% higher than} the protein in the corresponding unprocessed emulsion. For the A0-12 in the unprocessed emulsion with 24% fat and 11.5% protein, the DBC was 154% greater than in the processed emulsion with 14% protein. A similar trend was present in the other levels. The un- processed emulsions of the 30% and 36% fat levels with average protein contents of 10.4 and 9.3%, respectively, had DBC's which were -40 and 66 .AHE\w: oav OHIo< Sufi? mafiououm mo 0mm ofiu com: msoaufivcoo vommooouacs no vommoooum new mHo>oH vamfia unouomm«m manna mo poommo 0:9 Adiwa .Eouaomvo soHumuuaoosoo aflououm NH Ha OH m m n o m a m N H ‘ I i. - 1 I I E 1‘ I 1‘ I ‘ damaged wommoooumab II I I l I dogmasao pomwoooum dogmasao sow Mom A‘ I managed umm Nom I Ma dogmas—Bo own New I.. .mH enemas .H .N .m 11u0 an 3 1m) .H on Sn. 8 Am. n.. u N 1mm on d am. .4 E: H (ura r-! ,_q 67 .AHE\wa m.HV NHIo< pom aonHsam swamsmm m GH maHououm mo one one son: moumum :OHmHsam 3ou was mHo>oH vaHH ucoquMHw omega mo uoomwo ofiH .qH oustm HERE .HcmpHofiMV aoHumHusooaoo oHououm NH HH 2 m m H. o m 1V m m H 1 00H .. omH 0 H 0 1 com ml 3 p a .1 0mm 0. o n u p / 1 com .m d 1 m. 10mm m. aOHmHsao wommoooueab l.l..l .w W nOHmHoao mommoooum l / GOHmHsam now Nom 1.. (a ooq :OHmHoao mom mom I aonHoao umm New I it”. great 1151.017. silent vi mulsion The protein r one nemb' sore sir cessed e require that tr inate g of a 3 respec bEtVEE Torte: HOWevg Curvi: of the conceI was e; Ersat. 68 60% greater than the corresponding processed emulsion which averaged 12.6 and 10.8% protein, respectively. The dye AO-lO demonstrated the same effect with the DBC's that were 32%, 60% and 32% greater for the unprocessed emulsion at 24, 30 and 36% fat, respectively, than the processed emulsion. The results appeared to support a hypothesis which suggests that the protein membrane surrounding the lipid globule is layered, i.e., more than one membrane around each lipid globule (Saffle, 1969). It appeared that more sites for the dye to bind the protein were available in the unpro- cessed emulsion. In the unprocessed emulsion, the protein membrane layers may exist in a loose network allowing for the most convenient hydrophobic and hydrophilic interactions between components to satisfy thermodynamic requirements. This would be consistent with a hypothesis that suggested that the membrane would contract and denature on heating and thus elim- inate some binding sites. From figures 13 and 14, it appeared as if each graph was the result of a series of linear or near linear plots. For both dyes at their respective concentrations, there appeared to be a linear relationship between the DBC and the protein concentration in the 4-6% protein range. Torten and Whitaker (1964) published the same conclusion concerning AB-l. However the additive effect of the linear plots appeared to be the curvilinear relationship przsented in figures 13 and 14. From the aspect of the DBC, this curve possessed a negative slope, i.e. as the protein concentration increased the DBC decreased. The same basic relationship was exhibited by both dyes; however, the slope for the A0-12 system was greater than that for AO-lO. 69 The relevance of using model systems in this study was substantiated by comparing figures 11, 12, 13 and 14. It was reported in this study that both beef and pork myofibrillar and sarcoplasmic proteins demonstrated a negative curvilinear relationship between the DBC and the protein concentration. The curvilinear plots of the model and sausage emulsions did not exhibit slopes that were as great. 'Furthermore, at the levels of protein concentration in the model and sausage emulsion systems, the DBC of the aqueous systems was approaching a constant range in value. Torten and Whitaker (1964) alluded to a curvilinear relationship but other investigators (Ashworth, 1966, 1970; Udy, 1971) developed a linear relationship. A linear regression equation developed for a curvilinear relationship would have a large standard error. The existence of a curvilinear relationship would suggest that a limiting protein concen- tration exists beyond which there was no increase in bound dye though the protein concentration increased. This may explain the difficulty encountered by Torten and Whitaker (1964) in their work with fresh meat samples, since the protein levels might be beyond the limiting concentra- tion. Regression Equations Since.many* of the investigators assumed linearity in the relation- ship between a parameter of the dye, usually the DBC and the protein concentration, linear regression equations were calculated, however each investigator chose a different dye parameter. Udy (1971) chose an equa- tion of the form. 70 P = (C1 ' c) (3) k where c1 = stock dye solution, c = bound dye and k = a constant factor of adjustment for each product. For a fish meal patty and a"Gainesburger" the equations were P = (l.300—c)/0.0110 and P = (1.300-c)/0.0286, re- spectively. Torten and Whitaker (1964) calculated a regression equation for the DBC and bound dye (BD) against the protein concentration (Y). Using the DBC (mg dye bound/mg protein), they derived these equations for beef and pork with the AB-l dye. Beef Y 209.2 - 1.135 (DBC) (4) Pork Y 271.2 - 4.040 (DBC) (5) The large constant is a reflection of using the DBC in the regression equation. When the bound dye (BD) was used, the equations were as follows. Beef Y 8.18 + 0.301 (BD) r = 0.90 (6) Pork Y 5.43 4 0.367 (BD) r = 0.80 (7) Since each investigator chose a different dye parameter to relate to the protein concentration, our study tested a number of dye parameters. The dye parameters of A0-10 and A0712 tested were the DBC, DBC-1 (reci- procal of the DBC), bound dye (BD) and optical density of the free dye (0D). These parameters were correlated or regressed with the protein concentrations for each dye. These values were analyzed via basic sta- tistical tests for their relationship to the protein concentration and checked with an analysis of variance on the overall regression equation. The DBC-1 and DBC were correlated most highly with the protein con- centration. Data for the BD and OD did not yield consistently high correlations but some individual values were highly correlated. 71 Tables 6 and 7 represent a tabulation of the higher correlations between the dye parameters and protein concentration as measured by two different dye binding techniques for A0-12 and A0—10, respectively. Since we were dealing with the correlation coefficients (r) in tables 6 and 7, we were interested whether the parameter, protein concentration was interdependent with any dye parameter. This relationship is not to be confused with a regression of one variable on another. Our purpose was to estimate the degree to which these variables varied together (Sokal and Rohlf, 1969). No z-transformations were made for r. For the dye A0-12, there was a high correlation between the DBC—1 and protein concentration in the model system using the modified Ashworth method for analysis. This high correlation for the modified Ashworth method was present between the DBC.l and the protein concentrations of the processed and unprocessed emulsions, but the correlation was equally as high using the Ashworth method. The use of the DBC-1 yielded positive correlations as opposed to the negative correlations of the DBC, except for the data from the model system using the Ashworth method. A high correlation for the DBC or DBC.1 would be expected; since protein con- centration was used to calculate each. Therefore, only the bound dye (BD) or optical density (OD) parameters could be determined empirically. The correlation for the BD of A0-12 was highest in the unprocessed emul- sion using the modified Ashworth method. A similar trend was demonstrated in the correlations for the OD of A0-12 with the protein concentration. However the highly correlated parameter, the bound dye of A0-12, accounted only for 70% of the variation in the protein concentration for a 15 sample study with a standard error of the estimate of 0.566. moo~.r mme.o nmnn.l w¢m~.o memo.o mmwm.o NNHN.I mHmm.| aOHmHoao moxooocb oomo.l ommm.l mmmo.o HmHo.o mamm.o momo.o OHnm.I coho.| GOHmHoao poxoou Nmnm.l Heem.o «Hmm.o NHNN.0 mumm.o omHm.o mqoo.r mHHm.I Homoz venues wonuoe monuoa wonuoa wosuoa wozuoa monuoa monuoa amummm onuoum vonHmoz zuuo3£m< voHMHuoz nanosnm< voHMHvoz nuuossm¢ onMHvoz suuossmd huHmcow HmoHumo who mason Hlomn 0mm 72 muaoHOHmmooo aOHumHouuoo Auv .maoum%m cHououm ucouommHv oousu mo coHumuuaoosoo aHououm osu do OHIo¢ mo muouoamume ohm “sow aoum muaoHonwooo aOHumHmuuoo mo domHumnsoo .n mHan moqw.l mmwm.| Homm.o woqm.o owmm.o qum.o wHom.I o¢m¢.1 GOHmHoao moxoooas qoq~.1 omHo.o nHH~.o qomw.o Noam.o womo.o «mom.| Humm.l aonHssw moxooo meH.I ooaH.I mmqo.l oqu.I nmmm.0 anm.I mono.l mwmq.l Homo: monuma wosuoa cosmos wosuoa woeuoa wocuoe woeuoa morass amumhm aHmuoum monHvoz :uu03£m< voHMHwoz euuosnm< onMHmoz £uuo35m< voHMHvoz cuuosbm< hUHmoov HmoHumo who wasom Hlomn 0mm mucoHOHmmooo sOHumHouuoo Auv msoumkm cHououq uaouomev moans mo aoHumuuGooaoo :Hmuoua may no NHIo< mo mnouoawumn ohm usom aoum musoHOHmmooo GOHumHouuoo mo somHuoeaoo .o oHan 73 No apparent trends of interdependence were noted for A0-10 with the correlations for the BD and the protein concentration. However, for the parameter, 0D the modified Ashworth method was found to have two high correlations for the processed and unprocessed emulsions but the Ashworth method had the highest correlation in the processed emulsion. The cor- relation between the OD and the protein concentration using the Ashworth method was negative for the cooked emulsion and positive for raw emulsion. A similar trend of sign reversal was apparent using the modified Ashworth method and determing the BD. This discrepancy along with the considera- tions of the other correlations seemed to indicate that A0-10 could not be used to estimate protein content with any degree of confidence apply- ing bound dye values in either method of analysis nor applying the optical density with the Ashworth method. A consistent trend was apparent for the OD of A0—10 with the modified Ashworth method for all three systems, model emulsion, cooked emulsion and raw emulsion. The DBC and DBC.1 of A0-10 followed a similarrtrend of high correlations. When bound dye and Optical density values per se are to be used to estimate protein concentration in emulsion products, the correlations obtained here indicated that with A0-12 the modified method measuring the BD would be the method of choice for the unprocessed emulsion. The modified Ashworth method using A0-10 and measuring the OD might prefer- ably be chosen for the processed emulsion. The regression equations derived from all of the variables that accounted for the greatest amount of variation for a cooked emulsion were as follows! 74 1 A0-10 Y = -8.008 + 0.037 DBC- + 0.219 (BD) R2= 0.9997 (8) S.E.E. = 0.050 A0-12 Y = -8.648 + 1.282 DBC—1 + 0.007 (BD) R2= 0.9998 (9) S.E.E. = 0.046 The probabilities were less than 0.0005%. The uncooked emulsion presented less well defined situation. For both A0-10 and A0-12, a third variable was added to each equation. These equations were unwieldy since they contained 75% of the original number of dye parameters. 1 AO-lO Y = 1.125 + 0.038 DBC- — 0.019 (BD) — 0.504 (00) (10) S.E.E. = 0.053 A0—12 Y = -7.674 + 1.523 (DBC—1) + 0.004 (BD) + 0.005 (DBC) (ll) S.E.E. 8 0.056 The coefficients of determination for these equations were 0.9995 and 0.9988, respectively. Comparing equations 11 and 9, it was apparent that the equation was only slightly modified by the addition of the DBC variable. These equations underscore the point that while previous investiga- tors were correct in their assessment in using particular dependent var- iables to explain the protein concentration, the complexity of the relationship entails a more complete equation. SUMMARY AND CONCLUSION Tests on the various dyes, AO—lO, A0-12, A0-7 and AG-25 of certain dye parameters, absorbancy index (extinction coefficient), filter paper retention and adherence to Beer's Law demonstrated that the dyes used in this study were similar to the dyes reported by other investigators. These tests dischosed further that some of the dyes were more suitable, A0-12 and A0-10, for the dye binding technique than otherddyes, A0-7 and AG-25. It was found that the functionality of the dyes A0—12 and A0-10 could be improved by increasing the pH of the system from 1.9 to 3.0. In tests with various proteins, Hammerstein's casein and bovine serum albumin, the dyes demonstrated different affinities for the pro- teins. It was evident from observations on the dye binding abilities of the proteins that the structure of the dye molecule was an important consideration. Of the azo dyes, the dye AO-lO which had two sulfonic groups was bound less by the proteins than A0-7 and A0-12 which had one sulfonic group. The anthroquinone dye AG-25 with two sulfonic groups was bound in a greater quantity than the azo dye A0-10. The two sulfonic groups on the anthroquinone dye are widely separated but on A0-10 they are in close proximity, thus the difference in binding even though they both have two sulfonic groups. These results indicated that possible steric factors may enter into binding considerations between the dye and the protein. 1 When all four dyes were introduced into a meat-lipid system, utilizing lean meat and fat in various pr0portions, all of the dyes demonstrated a 75 76 similar pattern of binding to the proteins present, but a difference was noted in the amount of dye bound by the proteins. Apparently all of the dyes bound to the protein components but steric factors determined the amount of dye bound. Dye Binding and Model Systems Exposure of both AO-12 and AO-lO to various types of proteins, myo- fibrillar, sarc0plasmic and stroma, demonstrated that these dyes would bind to each type of protein. For a narrow protein concentration range the binding of AO-12 and A0-10 with thessarcOplasmic proteins appeared to be linear. A regular curvilinear relationship was found for the myofibrillar puoteins, AO-12 system. The myofibrillar proteins also bound larger amounts of A0-12 dye than the sarcoplasmic and stroma pro- teins. However, statistical tests disclosed that there was no signifi- cant difference between the quantity of dye bound by each protein compon- ent. For the myofibrillar proteins, the type of salt, KI or KCl used A to extract the proteins had no effect upon the DBC of the proteins. Nor did a species difference (pork vs beef) influence the DBC. Through the use of disc gel electrophoresis, it was evident that the dye (AD-12) bound with components of the sarcoplasmic and myofibrillar proteins. The constituent bound in the sarcoplasmic proteins is unknown, but for the myofibrillar proteins the myosin and tropomyosin fractions have been tentatively identified as binding with A0-12. The addition of oil to the myofibrillar and sarc0plasmic components to develop a model emulsion system was found to increase the DBC for A0-10 77 and AO-lZ. However, the oil itself had little or no effect upon the dye binding. This was demonstrated by maintaining the protein concentration constant but varying the lipid concentration over a wide range with no increase in the DBC over the range. AO-12 had theggreater DBC which increased proportionately with an increase in the dye concentration. A definite curvilinear relationship existed between the DBC and the protein concentration for both the myofibrillar and sarcoplasmic proteins. Dye Binding and Sausage Emulsions Tests performed in this study demonstrated that the dilution of the sample with a buffer could effect the DBC of the dye; therefore, this dilution must be carefully controlled. Generally, even though the unpro— cessed emulsion (raw) contained less protein than the processed emulsion (cooked), the unprocessed emulsion bound more dye than the processed emulsion. This observation applied equally to AO-lO and AO-12. The processed and unpvocessed emulsions were from the same batch; therefore the variation apparently resulted from the processing (smoking, cooking and chilling). Over narrow ranges there appears to be a linear relation- ship between the DBC and the protein concentration, in both the unpro- cessed and processed product. However, if the total range of the protein concentration was considered it became apparent that the relationship was more curvilinear. Regression equations derived from the above results demonstrated that the relationships were definitely curvilinear, whereas previous investigators were assuming linearity. This resulted in large standard EIIO in t6 78 errors of the mean. These equations further demonstrated that the varia- tion in the protein concentration for the unprocessed emulsion could be determined by two variables, the reciprocal of the DBC and the bound dye. The processed emulsion-dye binding relationship proved more complex and yielded an equation with three variables. However, the three variables differed for each dye. Statistical comparison of the equations and experimental results demonstrate that the model system approach for emulsions was applicable to the actual system. The dye binding technique has been offered by previous investigators (Udy, 1971; Ashworth, 1970) as a rapid means for protein determination in meat products. For this situation, the bound dye concentration of A0-12 in the modified Ashworth method in an unprocessed emulsion would be the technique to use for an on-line quality control system. However, only 70% of the variation in the protein content of a sausage emulsion could be accounted for by using the bound dye measurement in a limited sample study. A more extensive study into the efficacy of this method would have to in- volve an in-plant experiment consisting of a large number of samples. BIBLIOGRAPHY Ackerman, S. A., Swift, C. E., Carrol, R. J. and Townsend, W. E. 1971. Effects of types of fat and temperatures of comminution on dispersion of lipids in frankfurters. J. Food Sci. 36:266. (Acton, J. C. and Saffle, R. L. 1971. Stability of oil-indwater emulsions. 2. Effects of oil phase volume, stability test, viscosity, type of oil and protein additive. J. Food Sci. 36:1118. American Instrument Co. 1961. The determination of nitrogen by the Kjeldahl procedure including digestion, distillation and titration. Reprint No. 104. Aizawa, H. 1969. Interaction of food colours with protein. VII. Binding capacity and free energy of binding on azo colour by protein. J. Jap. Soc. Food and Nutr. 22:231. Armstrong, J. A. 1971. Semi~micro, dye-binding assay for rabbit inter- feron. Applied Microbiology 71:723. Ashworth, U. S. and Seals, R. 1957. Factors affecting the dye binding of milk proteins. Proc. West. Div. Amer. Dairy Sci. Assn. Ashworth, U. S. 1958. A rapid method for the estimation of serum pro- teins in milk. Proc. West. Div. Amer. Dairy Sci. Assn. Ashworth, U. S., Seals, R. and Erb, R. E. 1960. An improved procedure for the determination of milk proteins by dye binding. J. Dairy Sci. 43:614. Ashworth, U. S. and Chaudry, M.A. 1962. Dye binding capacity of milk proteins for amido black 10B and orange G. J. Dairy Sci. 45:452. Ashworth, U. S. 1965. A rapid method for the determination of casein in milk by the dye binding method and for the detection of Mastitis. J. Dairy Sci. 48:537. Ashworth, U. S. 1966. Determination of proteins in Dairy Products by dye binding. J. Dairy Sci. 49:133. Ashworth, U. S. 1970. Proteins in meat and egg products determined by dye binding. J. Food Sci. 36:509. Association of Official Agricultural Chemists. 1965. Official Methods 2£_Analysis g£_the A.O.A.C. Tenth Ed. Assoc. Offic. Agr. Chemists, Washington, D.C., 20004. 79 80 Becher, P. 1965. Emulsions: Theory and Practice. Rheinhold Publ. New York, New York. Bodwell, C. E. and McClain, P. E. 1971. "Chemistry of Animal Tissues: Proteins" in The Science of Meat and Meat Products, ed. Price, J. F. and Schweigert, B. S. W. H. Freeman, San Francisco. Borchert, R. M., Greaser, M. L., Bard, J. C., Cassens, R. G. and Briskey, E. J. 1967. Electron microsc0py of a meat emulsion. J. Food Sci. 32:419. Borton, R. J., Webb, N. B. and Bratzler, L. J. 1967. Emulsifying capaci- ties and emulsion stability of dilute meat slurries from various meat trimmings. Food Technol. 22:162. Borton, R. J. 1969. The effects of four species of bacteria on some properties of porcine muscle proteins. Ph.D. Thesis, Michigan State University. Brand, L., Heitz, J. R. 1971. Relation of the auramine 0 binding site to the active site of horse liver alcohol dehydrogenase. Biochemistry 10:2695. Bunyan, J. 1959. Orange G binding as a measure of protein content. J. Sci. Food Agric. 10:425. Bunyan, J. and WOodham, A. A. 1964. Protein quality of feeding stuffs. Brit. J. Nutr. 18:537. Carpenter, J. A. and Saffle, R. L. 1965. Some physical and chemical factors affecting the emulsifying capacity of meat protein extracts. Food Technol. 19:1567. Chapman, L. M., Greenberg, D. M. and Schmidt, C. L. A. 1927. Studies on the nature of the combination between certain acid dyes and proteins. J. Biol. Chem. 72:707. Christian, J. A. and Saffle, R. L. 1967. Thefrelative amounts of plant and animal fats and oils emulsified in a model system with muscle salt-soluble protein and an industrial emulsifier. Food Technol. 21:1042. Cockett, S. R. and Hilton, K. A. 1961. DyeinggCellulosic Fibers and Other Related Processes. Academic Press, New York. Cole, E. P. 1969. Alternative methods to the Kjeldahl estimation of protein nitrogen. Rev. Pure and Appl. Chem. 19:109. Colenbrander, V. F. and Martin, T. G. 1971. Orange G dye binding for determination of protein of sow's milk. J. Dairy Sci. 54:531. 81 Cunningham, A. E. 1971. Controlling the fat and moisture content of sausage. Food Technol. 25:689. Daniel, E. and Weber, G. 1966. Cooperative effects in binding by bovine serum albumin. 1. the binding of 1-Anilino-8-naphthalene sulfate. Fluorimetric titrations. Biochemistry 5:1893. Davis, B. J. 1965. Disc electrophoresis II-method and application to Human Serum Proteins. Annals N.Y. Acad. Sci. 121:404. Dolby, R. M. 1961. Dye binding methods of estimation of protein in milk. J. Dairy Res. 28:43. Fraenkel-Conrat, H. and Cooper, M. 1944. The use of dyes for the deter- mination of acid and basic groups in proteins. J. Biol. Chem. 154: 239. Golovkin, N. A., Shagan, O. S., Vasllev, A. A., and Mil'tsina, E. M. 1967. The change in the content of free amino acids during storage of chilled meat. Prikladnya Biokhymiya I Mikrobiologiya. 3:184. Grau, R. 1971. Binding and emulsification problems in the preparation of salt-free dietetic meat products. Fleischwirtschaft 51:786. Grauer, M., Cahill, V. R. and Ockerman, N. W. 1969. The emulsifying capacity bovine muscle tissue as related to time postdmortem and microbial contamination. Food Technol. 23:1591. Hammond, E. G., Seals, R. G. and Rheinbold, G. W. 1966. Determination of proteolysis by dye binding. J. Dairy Sci. 49:504. Hansen, L. J. 1960. Emulsion formation in finely comminuted sausage. Food Technol. 14:565. Hart, H. V., Hutchinson, J. B. and Groves, J. 1961. Study of the Udy dye binding method and its ability to estimate wheat proteins. Res. Assoc. Brit. Flour Millers' Bulletin. 12:41. Hegarty, G. R., Bratzler, L. J. and Pearson, A. M. 1963. Studies on the emulsifying properties of some intracellular beef muscle proteins. J. Food Sci. 28:663. Helander, E. 1957. On quantitative muscle protein determination: sar- coplasmic and myofibril protein content of normal and atrophic skeletal muscle. Acta Physiologica Scandinavia, 41 Supplemtum 141. Helmer, R. L. and Saffle, R. L. 1963. Effect of chopping temperature on the stability of sausage emulsions. Food Technol. 18:1195. 82 Ivey, F. J., Webb, N. B. and Jones, V. A. 1970. The effect of disperse phase droplet size and interfacial thickness on the emulsifying capa- city and stability of meat emulsions. Food Technol. 24:1279. Jonas, A. and Weber, G. 1971. Presence of arginine residues at the strong, hydrOphobic anion binding sites of bovine serum albumin. Biochemistry 10:1335. Kastner, J. 1965. A rapid determination of nitrogenous substances in plant material X. Two simple colorimetric methods for estimation of proteins. Rostlimna Vyroba 11:889. Kaul, A. R., Dhar, R. D. and Swaminathan, M. S. 1970a. Microscopic and other dye-binding techniques of screening for proteins in cereals. Imprl. Plant Protein Nucl. Tech. Proc. Symp. (IAEA: Vienna, Austria). Kaul, A. K., Dhar, R. D. and Raghaviah. 1970b. Macro and micro dye binding techniques for estimating protein quality in food. J. Food Sci. Technol. 7:11. Klotz, I. M., Walker, F. M. and Pwan, R. B. 1946. The binding of organic ions by proteins. J. Am. Chem. Soc. 68:1486. Klotz, I. M. 1957. "Interaction of Organic Molecules with Proteins." in Molecular Structure and Biological Specificity, ed. Pauling, L. and Tauo, H. I. Amer. Ins. Bio. Sci.,No. 2, Washington, D.C. Kramlich, W. E. 1971. "Sausage Products" in The Science of Meat and Meat~ Products. ed. Price, J. F. and Schweigert, B. S., W. H. Freeman Co., San Francisco. Laiken, N. and Nemethy, G. 1971. A new model for the binding of flexible ligands to proteins. Biochemistry 10:2101. Lawrie, R. A. 1966. Meet Science. Pergmon Press, Oxford, Engl. Lawrence, J. M., Herrick, H. E. and Grant, D. R. 1970. Analysis of wheat flour proteins by polyacrylamide gel electrophoresis. Anal. Chem. 47:98. McCready, S. T., Cunningham, F. E. 1971a. Properties of salt—extractable proteins of broiler meat. J. Sci. Food Agr. 22:317. McCready, S. T., Cunningham, F. E. 1971b. Electrophoretic separation of salt-soluble proteins of meat. Poultry Sci. 50:64. McIlvaine, T. C. 1921. A buffer solution for colorimetric comparison. J. Biol. Chem. 49:183. 83 Michigan State University Agricultural Experiment Station. 1966. STAT Series Description No. 9: Stepwise Addition of Variables to Form a least Squares Equation. Michigan State University Agricultural Experiment Station. 1969. STAT Series Description No. 8: Stepwise Deletion of Variables from a Least Squares Equation. Moran, E. T. jr., Jensen, L. S. and McGinnis, J. 1963. Dye binding by soybean and fish meal as an index of quality. J. Nutr. 79:239. Moriguchi, I., Fushimi, S. and Kaneniwa, N. 1971. Spectroscopic studies on molecular interactions. VII. Spectral changes of various dyes by bovine serum albumin and by organic solvents. Chem. Pharm. Bull. 19:1272. Morrison, G. S., Webb, N. B., Blumer, T. N., Ivey, F. J. and Haq, A. 1971 Relationship between composition and stability of sausage type emul- sions. J. Food Sci. 36:426. Moss, V. G. and Kielsmeier, E. W. 1967. A method for rapid determination of protein in meat by dye binding. Food Technol. 21:351. Mossberg, R. 1965. Some problems concerning the determination of crude protein content in cereals by dye binding. Agri. Hort. Genetics 23:206. Mossberg, R. 1966. Some analytical criteria of quality in barley. Agri. Hort. Genetica 24:193. Mossberg, R. 1968. "Estimation of protein content and quality by dye binding." in Evaluation of Novel Protein Products, Proc. of the Internet. Bio. Programme and Wenner-Gren Cntr. Symposium, Stockhblm 1968. ed. Bender, A. E., Ldfquist, B., Kehlberg, R. and Menck, L. Pergamon Press, New York. Neelakantan, S. and Froning, G. W. 1971. Emulsifying characteristics of some intracellular turkey muscle proteins. J. Food Sci. 36:613. Ney, K. H. and Wirutoma, J. D. G. 1970. Determination of the available lysine in milk and processed cheese by the reactive dye Remazol Brilliant Blue R. Z. Lebensm.-Unters—Forsch. 144:92. Osipow. L. J. 1962. Surface Chemistry: Theory and Application, Rhein- hold pub., New York. Ornstein, L. 1965. Disc electrophoresis I-background and theory. Annals N.Y. Acad. Sci. 121:404. 84 Outen, G. E., Tilley, J. M. A. and Wilson, R. F. 1966. Estimation of protein in dried herbages using the dyestuff Orange G. J. Food Sci. Agric. 17:285. Pomeranz, Y. 1965. Evaluation of factors affecting the determination of nitrogen in soya products by the biuret and Orange G dye binding methods. J. FoOd Sci. 30:307. Povlovskaya, T. E., Pasynski, A. G. 1963. The effect of 02 on binding dyes by proteins under irradiation. Dok. Akad. SSSR 149:976. Putnam, F. W. and Neurath, H. J. 1943. Complex formation between syn- thetic detergents and proteins. J. Biol. Chem. 150:263. Rampton, J. H., Pearson, A. M., Walker, J. E. and Kapsalis, J. G. 1969. Disc electrophoresis of Weber-Edsall extract and actomyosin from skeletal muscle. Agric. Food Chem. 19:238. Roberts, J. D. and Caserio, M. C. 1965. Basic Principles of Organic Chemistry, W. A. Benjamin, Inc., New York. Rosen, S. W. and Klotz, I. M. 1957. Phosphorus analogs as inhibitors of the beef heart succinic dehydrogenase system. Arcu. Biochem. Biophys. 67:161. Saffle, R. L. and Galbraith, J. W. 1964. Quantitative determination of salt soluble protein in various types of meat. Food Technol. 18:119. Saffle, R. L. 1968. "Meat Emulsions" in Advances in Food Research, v. 16, ed. Chichester, C. 0., Mrak, E. M. and Stewart, G. F., Academic Press, New York. Saffle, R. L. 1969. Stability of Meat Emulsions. Proc. Meat Ind. Res. Conf. Amer. Meat Inst. Salthouse, T. N., Macdonald, B. G. and Aguello, D. 1971. Vinyl sulfate dyes. possible applications in histology. Stain Tech. 46:245. Schroder, H. E. and Boyd, S. N. 1957. "The effect of dye structure on affinity for fibers." in Molecular Structure and Biological Speci- ficity. ed. Pauling, L. and Tauo, H. I., Amer. Inst. Bio. Sci. Publ. No. 2, Washington, D. C. Sherbon, J. W. 1970. Dye binding method for protein content of dairy products. J. Off. Anal. Chem. 53:862. Sokal, R. R. and Rohlf, F. J. 1969. Biometry: The principles and practice of statistics in biological research. W. H. Freeman & Co., San Francisco. 85 Swift, C. E., Lockett, C. and Fryar, A. J. 1961. Comminuted meat emulsions --the capacity of meats for emulsifying fat. Food Technol. 15:468. Swift, C. E. and Sulzbacher, W. 1963. Comminuted meat emulsions: factors affecting meat proteins as emulsion stabilizers. Food Technol. 17:224. Swift, C. E., Townsend, W. E. and Witnauer, L. P. 1968. Comminuted meat emulsions: relation of the melting characteristics of fat to emul- sion stability. Food Technol. 22:117. '+~ Szeverengi, E. and Hazkoto, E. 1966. Protein determination in the beer ‘ f industry. Soripar 13:70. ' Tereda, N., Maeda, K., Funakoshi, K. and Kametani, F. 1971. Binding of Ponceau 3R to bovine serum albumin. Chem. Pharm. Bull. 19:355. Thomasow, J., Mrowetz, G., Delfs, E. 1971. Determination of the protein content of milk according to the amido black method using the Pro- Milk automatic instrument. Milchwissenschaft 26:474. Torten, J. and Whitaker, J. R. 1964. Evaluation of the biuret and dye binding methods for protein determination in meats. J. Food Sci. 29:168. Townsend, W. E., Witnauer, L. P., Riboff, J. A. and Swift, C. E. 1968. Comminuted meat emulsions: differential thermal analysis of fat transitions. Food Technol. 22:319. Townsend, W. E., Ackerman, S. A., Witnauer, L. P., Palmer, W. E. and Swift, C. E. 1971. Effects of types and levels of fat and rates of temperature of commution on the processing and characteristics of frankfurters. J. Food Sci. 36:267. Trautman, J. C. 1964. Fat emulsifying properties of pre-rigor and post- rigor pork proteins. Food Technol. 18:1065. Tsugo, T., Iwaida, M. and Kawaguci, Y. 1966. Application of various acid dyes for estimation of protein in milk by dye binding. J. Dairy Sci. 49:455. Udy, D. C. 1954. Dye binding capacities of wheat flour protein fractions. Cereal Chem. 31:389. Udy, D. C. 1956. Estimation of protein in wheat and flour by ion binding. Cereal Chem. 33:190. Udy, D. C. 1971. Improved dye binding method for estimating protein. J. Amer. Oil Chem. Soc. 48:29A. ‘Jan I 86 Van Eerd, J. P. 1971. Meat emulsion stability influence of hydrophilic- 1ipophilic balance, salt concentration and blending with surfactants. J. Food Sci. 36:1121. Volodin, V. I., Troshiva, K. A. 1968. Micromethod for determining the content of nitrogen in protein fractions with the use of Orange G. Nauch. Tr. Ves., Nauch-Issled. Inst. Zernabob Kult, 2:168. APPENDIX 87 Appendix 1. Composition of solutions used in this study. I. Dye Binding Reagents A. Citrate-phosphate buffer (McIlvaine, 1921) 1) 0.1M Citric acid 19.21 gm citric acid in 1000 m1 deionized distilled water 2) 0.2M Sodium phosphate (dibasic) 71.7 gm NaZHP04°7H20 in 1000 m1 deionized distilled water Mixed in the following proportions: xml A + yml B + deionized, distilled water = 100 m1 x y pH x y pH 39.8 10.2 3.0 26.7 23.3 4.5 32.9 15.1 3.5 24.3 25.7 5.0 30 7 l9 3 4.0 21.0 29.0 5.5 B. Orange G (AD-10) .04 gm Orange G dye in 1000 m1 citrate-phosphate buffer, at pH 3.0 C. Brilliant Orange (AG—12) 1.3 gm Brilliant Orange in 1000 m1 citrate-phosphate buffer at pH 3.0 D. Acid Orange II (A0-7) 1.0 gm Acid Orange 11 in 1000 m1 citrate-phosphate buffer at a pH 3.0 E. Alizarin Cyanine Green (AG-25) 1.0 gm of Alizarin Cyanine Green in 1000 ml citrate- phosphate buffer at pH 3.0 II. Protein solubilities A. 0.03 M phosphate buffer, pH 7.4 0.602 gm of KH2P04 and 4.391bms of KZHP04 were dissolved in 1000 m1 of deionized and distilled water. 88 B. 1.1M KI, 0.1M phosphate buffer, pH 7.4 182.6 gm of K1, 2.178 gm of KH2P04, and 14.631 gms of K2HPO4 were dissolved in 1000 ml deionized distilled water 111. Electrophoretic solutions - Disc gel solutions A. Running gel - made by mixing 6.4 m1 of solution 1, 1.6 ml of solution 2, and 2.67 ml of solution 3 for 8 tubes. 1. Solution 1 5 ml of 2N HCl, 7.62 gm of Tris, 0.10 ml TMED. 81.25 m1 of 10M urea were mixed and then diluted to 100 ml with deionized distilled water. 2. Solution 2 43.3 gm of cyanogum were dissolved in 25 m1 of 10M urea and then diluted to 100 ml with deionized distilled water 0 3. Solution 3 1 mg of riboflavin was dissolved in 35 m1 10M urea and then diluted to 50 ml with deionized distilled water. B. Spacer gel - made by mixing 1.6 m1 of solution 1, 0.4 ml solution 2, and 0.67 ml of solution 3 for 8 tubes 1. Solution 1 5 m1 of 2N HCl, 1.25 gm of Tris, 0.075 ml TMED and 81.25 ml of 10M urea were mixed and then diluted to 100 ml with deionized distilled water. 2, Solution 2 33.3 gm cyanogum were dissolved in 25 ml of 10M urea and then diluted to 100 ml with deionized distilled water. 3. Solution 3 Same as solution 3 used in the Running gel. C. Tank buffer 6.0 gm Tris and 28.8 gm of glycine were dissolved in 1000 m1 of distilled deionized water. 100 ml of this buffer was diluted to 1 liter with distilled deionized water to provide the buffer used for each electrOphoretic run. 89 Appendix II. Composition of Three Example Sausage Formulations The formulations used in this study varied with the content of the meat used for the sausage; therefore, these formulations only serve as examples of the type product made. Fat Component 24% 30% 36% Lean beef 9.6 lb 7.1 lb 4.8 lb Beef trimmings 3.8 lb 5.0 lb 6.2 lb Pork jowls 3.7 lb 4.8 lb 6.0 lb Water (ice) 6.2 lb 6.3 lb 6.2 lb Salt 173.0 gm 173.0 gm 173.0 km Sugar 33.0 gm 33.0 gm 33.0 gm Seasoning 44.0 gm 44.0 gm 44.0 gm Sodium nitrate 1.4 gm 1.4 gm 1.4 gm Sodium nitrite 1.4 gm 1.4 gm 1.4 gm Sodium ascorbate 4.0 gm 4.0 gm 4.0 gm Proximate Analysis of Sausage Formulations Type of sample 2 Fat % Protein Z H20 % Total-Ash Unprocessed-24Z 17.4 11.5 65.1 93.9 Unprocessdd-3OZ 23.3 10.4 61.4 95.2 Unprocessed-36Z 29.1 9.3 57.3 95.7 Processed—24% 2403 1400 5708 9601 Processed-30% 30.3 12.6 53.2 96.1 Processed-36% 33.3 10.8 48.9 95.9 Unprocessed-24Z 20.6 11.6 63.9 96.1 Unprocessed-BOZ 27.9 12.1 57.0 96.9 Unprocessed-36Z 30.7 10.1 56.3 97.0 Processed-24% 26.1 14.7 55.5 96.3 Processed-30% 34.3 10.8 50.2 95.3 Processed-36% 35.9 14.7 45.1 95.6 Unprocessed-24Z 29.5 9 1 57.2 95.8 Unprocessed-3OZ -- -- -- -- Unprocessed-36Z 30.9 10.9 51.3 93.1 Processed-24% 30.3 12.1 51.8 94.2 Processed-30% -- -- -- -- Processed-36% 22.2 17.2 54.1 93.5 90 Appendix III A. Model Emulsion System Ashworth modified method Identification of sample: PKMYO = Pork myofibrillar proteins BFMYO - Beef myofibrillar proteins PKSAR = Pork sarcoplasmic proteins BFSAR“II Beef sarc0plasmic proteins Dye binding capacity (pg dye/mg protein) A010. One decimal place (34.9) Reciprocal of the dye binding capacity A010. One decimal place (28.2) Bound dye A010. One decimal place (14.3). Opticalddensity A010. Three decimal places (1.130). Dye binding capacity (ug dye/mg protein) A012. One decimal place (300.0). Reciprocal of the dye binding capacity A012. One decimal place (3.3). Bound dye A012. One decimal place (1230.0). Optical density A012. Three decimal places (1.055). Protein concentration (Kjeldahl, mg/ml). Two decimal places (0.41). ~c-..—..~_..‘ . "U ‘ PKMyoa ..-.-~_—- m...— 1 'RFMVO? “pKMYlO wQFMyofi 'RFMyoq "PKSAQZ RFSAD? 1 PKMvOI PKMY02 PKMY/3 PKMyOfi PKMYOfi pKMYO PKMYOP PKMYOQ PKMY11 BFMyOI BFMyO3 RFMv04 RFMYofi PFMV07 RFMvOQ PKSAQI PKSARE PKSAQ4 PKSADH BFSAQI BFSAQR RFSAQJ RFSAQ4 91 MODIFIED ASHWOVTH LSDEL AND LSADD' 2 349 235 166 146 127 399 094 377 075 072 369 095 153 122 127 061 “69 036 032 030 142 135 116 116 082 313 123 140 116 3 0287 0426 0602 0685 0787 1010 1064 1290 1333 1389 1449 0110 3667 0820 0788 1640 1450 2780 3130 333o 0707 0741 0861 0860 1227 0319 0813 0716 0862 060 4 143 134 133 137 145 140 145 142 140 136 144 090 IFO 180 190 110 130 190 170 180 160 170 190 190 100 150 170 200 100 16557- 180 5 1134 117d 1173 1157 1123 1149 1129 113? 1151 1163 1121 111d 0903 0796 C736 1041 F956 075? P915 768 1023 097? 0926 0880 0893 107? 0966 384‘) home 6 oood 2247 1527 0836 0800 p 8? Oééq 0653 0599 0840 0847 0670 0640 02?? 0211 108? 0981 0802 0763 0543 0800 0864 h7ee 130?. 10851 1240. 1250. 0234- 2504. APPENDIX p935 p427’ 812700 319200 8 12300 12808 12216 12248 12320 12490 112390 12566 12510 .12340 12519 11800 .12500 12500 ,12200 12300 11700 '12700 12260 12360 12410 12430 '12630 12190 -12200 12370 12560 9 1234 0932 3760 0025 0662 0749 1003 0731 1101 0426 0910 3609 0678 0776 127? 1276 0550 1414 1204 1120 1074 0894 1538 1340 1200 0940 III 12630 0696 HFTHOD 10 105fi641 057 1179080 0995094 - 114 141 154- 184 187 189-: 209 no: VJ—J 100 148 150 182 189 9326 529-” 603 113 126 155 163 233 048 138 143 164 gee A 92 Appendix III B. Model Emulsion System 10. Ashworth method Identification of sample: PKMYO = Pork myofibrillar proteins BFMYO = Beef myofibrillar proteins PKSAR = Pork sarcoplasmic proteins BFSAR = Beef sarcoplasmic proteins Dye binding capacity (pg dye/mg protein) A010. One decimal place (35.4) Reciprocal of the dye binding capacity A010. One decimal place (28.2) Bound dye A010. One decimal place (13.1) Optical density A010. Three decimal places (1.145) Dye binding capacity (ug dye/mg protein) A012. One decimal place (330.8) Reciprocal of the dye binding capacity A012. One decimal place (2.9) Bound dye A012. One decimal place (1224.0) Optical density A012. Three decimal places (1.030) Protein concentration (Kjeldahl, mg/ml). Two decimal places (0.37) .5!!me ...§... 3.3.. $.14 .n 4 .5 n ,1 ¢ t U LSDEL AND LSADD PKMyO 3 28? 44? :70 725 4 131 I43 1?? 130 5 llaar 115A 113” 114” 93 ASHWOQTH METHOD APPENDIX 7 8 9 .901?2a1 =.511?30,191.063 .5?.1101 771220, 111 pKMyo,194 961140110’001910919308 PKMYO 391109C143112¢De111231290 PKNYO 089112314a1120p754139122oe 581 PKMYO1387 14fi1431132079212712901 PKMVO 373 3631351114069 "PKMyl U7912661501116067.1471273o' PKMyl1079 265153120106a11561246' PKMYI ,68 471151123005021oc1110o -8FMvo1133 42fi147115~1841 54 BFMyO-186 a3914n1161 _ RFMYo11403771h8114BAQZBI RFMY04101 99015311200929121 266544115 316143116506 4931301201 61314414260- 250“131 , -222133161flb 14 591180 141- 8191601023094910512411113C131 1~~ — ~vv wu~~ ”v ao71a3~93709001251 826194°96?077F12¢1240-1061160 935193388“066- th 1 0 11119059310 16912603. 101 099183P981oa59_191206; 150 400150073dp101.491219-151- 819171784n0896113124011341140 752904093n08221221259¢1201153 389P15094qpé20_391260 1-1 .— ..v “~_7.~.7 - -_- . _-. . ..._ .-—-_.. 11,, O91v193 ‘ BFSAQI 94 Appendix IV A. Ashworth Modified Method Processed and Unprocessed Sausage Emulsion Sample identification: a) 24CWSl - 24% fat emulsion b) 30CWSl — 30% fat emulsion c) 36CWSl - 36% fat emulsion C = cooked emulsion U = uncooked emulsion Dye binding capacity (pg dye/mg protein) A010. Two decimal place1(4.68) Reciprocal of the dye binding capacity, A010. No decimal (214) Bound dye A010. One decimal place (37.6) Optical density A010. Three decimal places (.086) Protein concentration (Kjeldahl, mg/ml). Two decimal places (08.02) Same as No. 1 above. Dye binding capacity (ug dye/mg protein) A010. Two decimal places(5.10) Reciprocal of the dye binding capacity A010. No decimal (196) Bound dye AOlO. One decimal place (36.2) Optical density A010. Three decimal places (1.406) Protein concentration (Kjeldahl, mg/ml). Two decimal places (7.09) 1 V 24Ch!51 24C1I'SE "24Cws3 24Cw54 BOCwsl '"30CH52 30CN53 30Cw§4 ”36CW81 36Cw5? 36CwsE 36CWS£¥ . ._95 . MODIFIED ASHWOQTH METHOD APPENDIX IV A . 'PPOCESSED AND UNPPOCESSFD SAUSAGE LSDEL AND LSADD 2 3 fléfiéla H61178 flBSRCS 613163 509155 76W130 637157 01024? $29004, 5431840 6231610 577258. .117 _12{'. 5 096 095 091 086 103 17a 100 111 109 6 bsoz 11670 0767 1140 0682 0591 0606 poao n51o oa7e 0580 0000 PQE ‘ZQUWSE “‘BOUWSI ‘30UW84 '7 24uw51 eauwsfl 24UWS4 24UWS5 BOUWSZ 3OUWS? BOUWSE 36UWSZ 36UWS3 36UWSI 36UW54 36UWSS 8 . 0510 0623 0621 0784 0730 0713 0895 0993 0991 0004 1110 1010 0812 0912 EMULSIONS 196 159 161”‘ 127‘ 137 140 11?? 112. 122 111 090 0011 123 110. 092» 109q 96 Appendix IV B. Ashworth Method 10. ll. 12. Processed and Unprocessed Sausage Emulsion Sample identification: a) 24CWSl - 24% fat emulsion b) 3OCWSl - 30% fat emulsion c) 36CWSl - 36% fat emulsion C = cooked emulsion U - uncooked emulsion Dye binding capacity (pg dye bound/mg protein) A010. One decimal place (89.3) Reciprocal of the dye binding capacity A010. Two decimal places (11.20) Bound dye A010. One decimal place (076.7) Optical density A010. Three decimal places (0.164) Protein concentration (Kjeldahl mg/ml). Three decimal places (8.586) Same as No. 1 above. Dye binding capacity 61g dye bound/mg protein) A010. One deciaml place (104.4) Reciprocal of the dye binding capacity A010. Two decimal places (9.58) Bound dye A010. One decimal place (071.1) Optical density A010. Three decinnl places (0.361) Protein concentration (Kjeldahl mg/ml). Three decimal places (6.809) 97 ASHWOQTH METHOD .. 11 APPENDIX IV R PROCESSED AND UNPROCPSSFD SAUSAGE FMULSIONS u LSDEL AND LSAUD l S 6 7 8 24nglr- 767016Ah696 aauw511044_ 24am92- 01690414 24Uw521064 316 .1 24Cw§° 01960414 Bauw531044 9560711 24cmq 01960600 aauwsqlnefi . 24Cmsli- 01660672 P4uw561054 a4qw71n 240wp6 9701140 10006720 ?auw561043 31cws11032 96 502457154 BOUWS11141 30ng 101: 023q730a souw521147 673060 “~ 1“. ancmg .02007630 aouwsa1172.ea snows 101 ”07647229 '300w541273 30Cws-, ~0223374553 3ouwsd1114 3970681 44 "“ ** 30Cw§ fiP307154-~- BOUW561208 028 36Cw91111a _.?37564l 36UW511282 36Cm9211°3 .‘02156641 360w521525. ' 1-1.1.1... ____ -' 36(31115: 1711966710”- 36U'.'.‘531332 36Cw9 -. :02185990 36uw541299 36Cw§5 9631036 02437610 36ow531a73. ‘~““*m~_~~36CmQ0110C ._q 0201671o—~*~_——36uw561275_ 98 Appendix V A. Ashworth Modified Method 10. 11. 12. Processed and Unprocessed Sausage Emulsion Sample identification: a) 24CWSl - 24% fat emulsion b) BOCWSl - 30% fat emulsion c) 36CWSl - 36% fat emulsion C = cooked emulsion U - uncooked emulsion Dye binding capacity (yg dye/mg protein) A012. No decimal (164) Reciprocal of the dye binding capacity, A012. Two decimal places (6.10) Bound dye A012. No decimal (1269) Optical density A012. Three decimal places (0.560) Protein concentration (Kjeldahl mg/ml). Two decimal places (7.73) Same as No. 1 above Dye binding capacity (fig dye/mg protein) A012. No decimal (182) Reciprocal of the dye binding capacity A012. Two decimal places (5.50) Bound dye A012. No decimal (1251) Optical density A012. Three decimal places (0.087) Protein concentration (Kjeldahl mg/ml). Two decimal places (6.92) 99 VODIFIED ASHWODTH MPTHOD APPENDIX v A PROCESSED AND UNPDOCPSSFD SAUSAC? EMULSIONS U LSDEL AND LSADD ’ 1 2 3 4 5 6 7 8 24Cw$116 1012690560 773 240w5118 24Cms?1615211?fi507efl 783 24uwsaes‘ 118196?9329 665 ~- ~ aauwsaes- 93 127634341190 aauwsagss 30CW9120s+ss1290174¢ 641 aauwsszac ‘M1 ““‘30CMSRF141671701157 603 » '“‘“-30UW51198.05 30Cwsql61117129a134_000 souwsazs 30CM94130169199919391000 souwsaas '~' 36Cw§1?6. 496 300w5429 36Cm5223¢ 2012901730 542 30095027 36Cm53231 F47 ' 360w51266 970 - w~~vw636005331 - 360w5327 36UWSAB€-. ‘v—~wnu~~s~+~~M~MW—nu-+ 6~~iw«—s~»fl-~w v«~36uws§31 q16 240-19410 .. ; .— —-—— n v— _ HLL—g >——..—— a k a“ 4‘.— WW--.“ +4 7 u... _ — - -_...»oA—-4_-- ‘.. _,~. .—. a a. __ ”___. 4 A 1.. _ ..___ A. w —— -w - -— —4 .- ”a -— *V—v—‘O‘—4 .. ~—.— ~ _ -w ——-. ~— _‘.. H? .. ~ H k- 4- --.¥—g—_.~ —- a if .. --- m—-——-- _.-— M m._._..~_ _ __. .4 i.— .uufi .. w _-.-.r ___. .. -_~- _...___.__-—-.- .9— ...— ._ - ”4 _____ .— F... -m u... .— ._ ____. - h - ".. —— ---___~.. . » x»—--- _ - ..... ._. ..-.. fl___.__~._._.—fl.. .,... -__.__ , .ai _11.-.._.._...._.....t...u _. _ .1 u..-»»....._ ...._..-__._...._ . .fi... . . ~...... _ . ._., -____$ ' —-‘ ._ — — .. M _. _. .. - k ._- A-.. * £11me9. .334, m. fiuafilh 1”, - 100 Appendix V B. Ashworth Method 10. 11. 12. Processed and Unprocessed Sausage Emulsion Sample identification: a) 24CWSl - 24% fat emulsion b) 30CWSl - 30% fat emulsion c) 36CWSl - 36% fat emulsion C = cooked U = uncooked Dye binding capacity (pg dye/mg protein) A012. One decimal place (214.5) Reciprocal of the dye binding capacity A012. Two decimal places (4.66) Bound dye A012. One decimal place (182.3) Optical density A012. Three decimal places (0.650) Protein concentration (Kjeldahl mg/ml). Three decimal places (8.50) Same as No. 1 above. Dye binding capacity (ug dye/mg protein) A012. One decimal place (256.4) Reciprocal of the dye binding capacity A012. Two decimal places (03.90) Bound dye A012. One decimal place (169.2) Optical density A012. Three decimal places (1.082) Protein concentration (Kjeldahl mg/ml). Three decimal places (6.599) 101 AquopTH METHOD APPENDIX v 0 poocrsssn AND Umoaoccssso SAUSAGE FHULSIONS u LSDEL AND LSADD ‘ l 2 3 7 10 11 12 24Cw8121450466 240ws1 .169210826599 24Cmsaalsro46r 24uwsl 167911367020 *~* ~ 24CM8320810470 daauwsr..r - 169210026459 24Cws4211 0474 24006116 50380169910606599 24cN56907 0402 74Uw8926213382167411356388 “”““'“””"24Cw5&209, 476 340w5;959203s6167411376458 30CM91245 ”40¢ Souwsl967b0373167412006250 30c052243 041? 3ouws -0190342164a129F5645 '~““w~m 3000532311046? 300ws 0860324165912275376 30CW9426030384 :00w54 7840350168411246048 300w;62463.406 3ouw9 0390352169311735961 7““"’*""“‘ 30(2111561249 401 7“ 3OU‘.’.‘ 973p3361718 108135779 ' ._.__.-_1 36cms1-saa 359 36uws1 1210320161611466164 36Cw82268 37? 360ws 0710326169610743530 ““** °“36Cws??63( :37m 36005-.0040333169611036646 y 36cws4967/139? .. 36UWS" 6920337170610046760 36Cwssa636,37q177c w . 3601.45w 0110332166511705530 ‘“”*“7‘“"36Cm5q2692 3731769 so —~~~w-~3suwsc 1010322167911095414—«~—um--- I EACSEWKI $185143. F”.- , r . . . 1 6 I. r MnCH 6AM EWATE UNIVEIRISITY 'IBRAFHES 23191 11 69 0314