. sowsnm' OF ISOtATED sov mom INRUENCEDBYALDBNDE TREATMENT; _ ' ' ’ ” REDUCTION.oxmmonmp.“ . ' z , :. NEUTRAL SALT. Thesis for me Degree (if-MS. i - ' - MICHiGA’N STATE UN’WERSW . RONALD T. KURPSUS. .' 1.971 ffifi'é's LIBRfiRY NI ifhig h S {a ”313 University MS 3‘43? ABSTRACT SOLUBILITY OF ISOLATED SOY PROTEIN INFLUENCED BY ALDEHYDE TREATMENT, REDUCTION-OXIDATION, AND NEUTRAL SALT By Ronald T. Kurpius The solubility of isolated soy protein in aqueous solutions was decreased utilizing three different methods. The methods were treatment with aldehydes, reduction- oxidation—heat treatment, and incorporation of a neutral divalent cationic salt. The aldehydes showed limited effects at a very acid pH while the reduction-reoxidation method decreased the solubility 80% at a pH near neutrality. The neutral salt method decreases the solubility instantan- eously about 50% at a neutral pH. Information from this study may be helpful in suggesting ways of imparting improved structure to soy proteins when used to produce fabricated food systems. SOLUBILITY OF ISOLATED SOY PROTEIN INFLUENCED BY ALDEHYDE TREATMENT, REDUCTION-OXIDATION, AND NEUTRAL SALT BY f- n “I ‘ I {30 Ronald waKurpius 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 1971 ACKNOWLEDGMENTS The author wishes to express appreciation to his wife for her encouragement, understanding, and self-sacrifice endured during the course of this graduate program. Appreciation.is extended to Dr. Walter M. Urbain for his guidance throughout the course of the program; appreciation is also extended to Dr. I» E. Dawson and Dr. D. E. Ullrey for their helpful suggestions and critical reading of the manuscript. The author is indebted to the National Institute of Health for providing the traineeship making it financially possible to complete this work. ' ii TABLE OF CONTENTS ACKNOWLEDGMENTS . . . . . . . . . LIST OF TABIES O O C O O O O O O INTRODUCTIOI‘I . o o o o o o o o 0 LITERATURE REVIEW . . . . . . . . Historical PerSpective of Soybean Protein Investigation . . . . . . . . Preparation of the Soybean for Protein Extraction . Characteristics of the Soybean Globulins Alteration of Proteins . . . . . . METHODS AND MATERIALS . . . . . . . Soybeans . . . . . . . . . . . Fat Extraction . . Preparation of the Crude Protein Extract Viscosity Measurements . . . Protein Solubility . . . Nitrogen Determinations . Heat Treatment . . . . RESULTS 0 g g 0 o o o O o O 0 Neutral Salts Effect on Soy Protein Solubility . Aldehydes and Heat: Their Effects on the Solubility and Relative Viscosity of Soy Protein Reduction, Oxidation, and Heat Treatment Using Cysteine and Potassium Bromate: Their Effect on Solubility and Relative Viscosity . Effect of Calcium Chloride on Soy Protein SOlubilitY o o o o o o o e 0 iii Page ii 11 14 17 l7 17 17 2O 22 22 23 24 26 3O 36 DISCUSSION Effect Effect Effect Effect SUMMARY AND CONCLUSIONS BIBLIOGRAPHY of Neutral Salts of Calcium Chloride of Aldehydes . of Reduction, Oxidation, iv and Heat Page 37 wwww \O CI)\'I\1 42 44 Table l. 2. 3. LIST OF TABLES Particle size distribution of the defatted soybean meal used for protein extraction . Sequence for protein extraction . . . . Solubility of soy protein at various ionic Strengths . g g o o o o o 0 o o Solubility of soy protein treated with aldehydes and heat . . . . . . . . Relative viscosity of soy protein after treatment with aldehydes and heat . . . Percentage soluble soy protein after re- duction, oxidation, and heat treatment . Change in pH of soy protein after treat- ment with cysteine, potassium bromate, and heat . . . . . . . . . . . Relative viscosity of soy protein after re- duction, oxidation, and heat treatment . Solubility of soy protein when treated with calcium chloride . . . . . . . . . Page 18 21 25 28 29 32 33 35 36 INTRODUCTION The direct use of plant protein for human food purposes is increasing. This trend will continue as the cost of conversion of plant protein into animal protein becomes increasingly expensive. The soybean has long been recOgnized as a source of high nutritional quality protein. The soybean contains approximately 35% protein, and it provides a relatively inexpensive source of all the essen- tial amino acids except methionine, which is the first limit- ing amino acid in the soy protein (Pomeranz, 1966). In addition, the production of soybeans as a cash cr0p in the United States of America has been phenomenal. It is now the No. 2 cash crop grown in this country (Rakosky, 1971). Traditionally in the United States, the major value of the soybean has been its oil content. The meal obtained after removal of the oil is utilized mostly as animal feed. During the past three decades, however, the interest in using the soy protein directly as food product has increased. At present there are three basic intermediate soy products derived from the defatted meal (Dimler, 1969). The U. S. Food and Drug Administration has assigned definitions which essentially classify the products by protein content 1 on a dry basis: 1. Soy flour or grits - 50% protein, 2. Soy concentrates - 70% protein, 3. Isolated soy protein - 90% protein. The main use of soy protein in foods for human consumption has been in existing food systems to incorporate certain functional pr0perties exhibited by these proteins. The soy protein is a very complex aggregate of many different fractions. These fractions singularly and collectively, exhibit diverse functional prOperties when incorporated into food systems. For example, soy proteins are used as aerating agents in whipped t0pping and frozen desserts. The cohesive and elastic properties of soy protein enhance the quality of baked goods. Soy protein concentrates and isolates are used in meat systems to bind water and fat where emulsion stability is required (Wolf, 1970). When soy protein concentrates are used in meat patties, the juices are retained resulting in less shrink and a more desirable product in terms of preparation and consumer satisfaction (Rakosky, 1971). As mentioned, most of the applications are of a functional nature; however, more recently, there have been developments that utilize the protein for nutritional as well as functional properties. These are the developments of the texturized soy protein product (a cooker-extrusion process) and the spinning process whereby the isolated soy 3 protein is prepared into a Spinning d0pe, extruded through spinnerettes, and the extruded protein coagulated in an acid-salt bath to form fibers. After conditioning, the fibers can be combined with carbohydrate, fat, color, flavor, and an edible binder such as egg albumin to form meat-like products. This study was undertaken to further the work in the area of the spun fiber technology. Product extruded through Spinnerettes and coagulated in the acid-salt bath, exhibit a persistent sourness that is hard to mask, and frequently it may be detected in finished products. Also, because of the nature of the soy protein, there is a tendency for the fabricated fiber food to melt or become soluble upon heating to cooking temperatures. One pro- cedure to correct both of these problems would be to discover a way to produce a coagulable protein at a pH value higher than 4.5--for example at a pH of 6.0. If this is possible and if the protein is also heat coagulable, it may be possible to overcome the sourness and melting problems. This study consists of four parts. First, the relationship between ionic strength and the solubility of the native protein was studied. Secondly, the effects of selected aldehydes on the physical properties of the native soy protein were monitored as an index of the chemical modification. Changes in the solubility and viscosity of the protein were monitored. Thirdly, the native protein was alternately reduced and then reoxidized followed by heat treatment in order to attempt to change the physical characteristics of the protein in the manner suggested previously. Again, changes in solubility and viscosity were monitored. The last part of the experiment consisted of observing the effects on solubility of the native protein when a calcium salt was incorporated into an aqueous soy protein system. LITERATURE REV LEW Historical Perspective of Sgybean Protein Investigation The domestically cultivated soybean originated from a naturally wild growing plant of eastern Asia. The cultivated Species is thought to have emerged through gene mutations without any change in chromosome number. Accord- ing to international botanical rules, the scientific name of the soybean is Glycine max (L.) Merrill (Morse, 1950). The soybean has long been a part of tradition and culture in China. As early as 2838 B.C., King Chan Nonog of China mentioned soybeans in a medical treatise (Pomeranz, 1966). It has only been the last fifty years that the soybean has become of significant commercial value in the United States. As mentioned earlier, soybean production in the United States now represents the No. 2 cash crop in this country (Rakosky, 1971). The Glycinig Fraction There has been considerable scientific work accom- plished in trying to elucidate the nature of soy protein. The first quantitative work of significance indicated that the soybean contains one major protein fraction called glycinin and two smaller fractions named phaseolin and 5 _legumelin (Osborne and Campbell, 1898). The protein extract was obtained by first coarsely grinding the bean, removing the outer seed coat, and finally grinding to a fine powder. The protein extract was obtained by an aqueous extraction of the meal. Upon dialysis, it was found that 16.6 g of protein precipitated from a 100 g of the aqueous extract. The precipitate could be largely solublized in brine; hence this fraction, glycinin, was considered to be a globulin protein. Solubility studies suggested that phaseolin was a globulin type protein, but it was more soluble than glycinin. The legumelin fraction was consid- ered a soluble albumin-like protein which would begin to coagulate when heated to 55°C. Osborne and Campbell suggested that the globulin portion remained soluble in the soybean seed by virtue of the potassium phosphates contained therein; however, Smith g§_gl., (1938) reported that this was questionable based on their work of dispersing the isolated soy protein in various concentrations of neutral salts. They found that salts, up to a limited concentration, inhibit the dispersion of the protein. Nagel et al., (1938a) related the insolubility of the extracted globulin fraction to the removal of a peptization agent when the meal or protein extract was treated with an ethanol-water mixture. These workers suggested that the removal of the lecithin fraction , with alcohol lowered the protein diapersibility. Wolf 7 g§_g;., (1963) added lecithin to protein previously extracted with ethanol, but they could not reestablish the level of solubility before alcohol treatment. Recent work describing the polymerization characteristics of the globulin protein probably indicates more correctly the reason for the change in solubility of the extracted globulin (Nash gt_gl., 1967; weir et al., 1964; Wolf g§_al., 1963). The globulin fraction, which is normally soluble in 0.5 ionic strength buffer, is approximately 20% less soluble when redispersed from a precipitated extract. It is believed the insolubility is due to formation of di- sulfide polymers when subunits of the globulin are in close molecular proximity. Characterization of the Songrotein Briggs and Mann (1950) have presented evidence indicating the complexity of the soy protein. Using electro- phoretic techniques, they were able to distinguish at least seven peaks. The glycinin component described by Osborne and Campbell was found to be a mixture of at least three individual proteins. Sixty % of the glycinin fraction was found to consist of a fairly homogeneous protein that would precipitate from solution if an extract was held at approximately 2°C for several hours. This cold insoluble fraction (CIF) showed a single peak on electrophoretic analysis. This was the first success this writer knows of in isolating a homogeneous protein from a mixture of soy protein components. The work of Briggs and Mann has been followed in rapid succession with refined studies using the 30phistication of ultracentrifugation, chromo- tography, and immunochemical techniques (Wolf, 1969). Ultracentrifuge data show a pattern of four major resolvable fractions having approximate 520,w values of 2, 7, 11, and 158 (Wolf, 1970). The term 520,w is a sedimentation coefficient corrected to the value it would have had in a solvent with the viscosity and density of water at 20°C, and S is the Svedberg unit of sedimentation rate equal to 10'13 seconds (Chervenka, 1969). These four fractions contain numerous protein species. The globulin components of the original glycinin are contained mainly in the 7 and 118 units (Wolf, 1970). The immunochemical and immunoelectrophoretic tech- niques have been used to completely characterize the separate soy proteins. Catsimpoolas g§_gl., (1968) have demonstrated this technique. Soybean proteins were separated into at least 12 antigenically distinct components by immunoelectrophoresis in agar gel. Pre ration of the So bea for Protein Extraction Preparation of the soybean for protein extraction requires some combination of grinding and oil extraction. The normal laboratory procedure is to crack, dehull, and flake the soybeans prior to solvent oil extraction, followed 9 by grinding to a desirable size. A wide variety of solvents has been used to extract the soybean oil prior to grinding. Smiley and Smith (1946) note that with the exception of Csonka and Jones who used ethyl ether, most researchers have used either petroleum ether or benzene. Smiley and Smith (1946) used ethanol which they claim gave a much lighter colored protein, suggesting that some additional impurity had been removed. The extraction was accomplished in a modified Soxhlet type extractor where the ethanol was cooled in a condenser prior to contact with the meal, thereby minimizing heat denatur- ation. Belter and Smith (1952) report that in commercial Operations most denaturation of soy protein arises from Operations after oil extraction such as during vapor desolventization. Other work has shown that hot or cold methanol or ethanol extraction reduces the subsequent extractability of soy protein with water and salt solutions with the globulin portion being the most sensitive (Mann and Briggs, 1950). In the extreme case, nitrogen extracted by water was reduced 70% with hot ethanol at 73°C. Cold methanol at 20°C reduced extractable nitrogen by 24%. Roberts and Briggs (1963) and Wolf et al., (1963) report that ethanol at various concentrations causes denaturation of the globulin fraction of the soy protein. Sixty % ethanol added to wet protein curd causes the most denaturation, 10 and the 7S ultracentrifugation component is the most sensitive to alcohol denaturation. The action of the ethanol is 80% complete in 15 minutes. The criterion for denaturation.was insolubility in a standard phOSphate buffer after ethanol treatment. After oil extraction, the defatted meal is finely ground in preparation for protein extraction. Smith gt_§1., (1938) found that the quantity of nitrogenous materials extracted from meal increases as the particle size of the meal decreases when various sized sieves are used in a Wiley mill. The smallest average particle size can be obtained from a pebble mill or ball mill. Nagel g§_gl., (1938b) confirmed these results, but suggested that pro- longed dry grinding in the ball mill to a 200 mesh size can result in lowered nitrogen extractability because of mechanical shock causing protein denaturation. The common method for grinding the protein is the use of a Wiley type mill with a 0.5 mm screen (Smith g§_g1., 1938; Mann and Briggs, 1950; Briggs and Mann, 1950). ProteiniExtractigg There are four methods used for soy protein ex- traction (Smiley and Smith, 1946). 14 Extraction with water and acid precipitation. 2. IExtraction with dilute alkali and acid pre- cipitation. 3. Extraction.with salt solution and precipitation by dialysis. 11 4, Extraction with water or salt solution, precipitation by saturating with ammonium sulfate, rediSpersing in water, and pre- cipitation by dialysis. Methods 1 or 2 are essentially those of the commercial industry while methods 3 and 4 closely represent the original methods of Osborne and Campbell. All the methods typically yield mainly the globulin fraction while the minor proteins remain in the whey. None of the above methods yields a homogeneous protein; the product contains all the fractions of the globulin as demonstrated by the electrophoretic study of Briggs and Mann (1950). Characteristics of tfie Soybean Globulins Briggs and Mann (1950) point out that 75% of the water extractable protein consists of globulins. These globulins have been the subject of much research because of their peculiar nature. Neutral Salt Effects Smith et al., (1938) discovered that the soybean meal yielded unusually high water peptization of nitrogen- ous constituents, but the rate of peptization decreased rapidly as increasingly higher ionic strength solvents were used in extraction. Water will extract about 90% of the nitrogenous material in soy meal while there is a steady decrease in extraction rate when salt concentration is increased from 0 to 0.1 N NaCl when.minimal extraction 12 of 46% occurs. The extraction rate then gradually in- creases to 86% as the salt concentration is increased from 0.1 to 1.0 N NaCl. At saturation with NaCl, 73% of the nitrogenous material is extracted from the meal. Similar results were obtained when the salt concentrations of water extracts were varied. These researchers also found that the minimum on the extraction curves varied over a wider range for univalent cations than for divalent cations. Wolf and Briggs (1956) using ultracentrifugal techniques determined that components or mixtures of components with 520,w values of 11 and 155 were reSponsible for solubility changes with varying ionic strengths. The 11 and 158 components represent most of the globulin fractions (Wolf and Briggs, 1959). Smith and his associates reported that Ca012 pro- duced a white flocculent precipitate immediately when an aqueous protein solution was made 0.0175 N with reSpect to CaCl2. A white precipitate also formed when a protein water extract was made 0.1 N with respect to NaCl, but the precipitate formed slowly. Wolf and Briggs (1958) suggest that the effect of CaC12 on precipitation of protein is not just an ionic effect since they could not obtain precipitation with NaCl in any range from 0.0 to 0.5 N at pH 7.6 This may indicate the ability of the calcium cation to complex with the protein. 13 Polymerization and Heating Effects Much research has shown the ability of the globulins to polymerize via disulfide links (Nash and Wolf, 1967; Kelley and Pressey, 1966; Wolf g§_gl., 1964; Wolf g§_gl., 1963). This usually can take place when the globulins are in close proximity; proximity can be caused by either isoelectric precipitation or dialysis. The result of polymerization is reduced solubility. It is mainly the 7 and 118 components that polymerize (Nash and Wolf, 1967). Reports differ as to the effect of heat on the globulins. wolf and Tamura (1969) heated a purified llS component from 25 to 100°C. After reducing the disulfide links, they found aggregation and precipitation occurred; if the sulfhydryl groups were blocked, precipitation was prevented. Hahn and Briggs (1950) found that heating the 118 component to 75°C for 5 hours or less, without modifi- cations of Wolf and Tamura, did not cause precipitation, but they were able to cause a fraction of the proteins to precipitate from an aqueous extract containing all the water extractable protein. Interestingly, they found that this extract would precipitate while standing at room temperature for 5 to 7 days while protected from bacterial contamination and growth. This may suggest formation of disulfide polymers via atmoSpheric oxidation causing aggregation leading to precipitation. 14 Briggs and Wolf (1957) report that the 118 component is capable of forming dimers, trimers, and tetramers linked by disulfide bonds through oxidation of sulfhydryl groups existing at the surface of the 118 component. Alteration of Proteins One of the several chemical modifications of proteins is accomplished with aldehydes. A familiar reaction of aldehyde with protein is the Formal titration where two moles of formaldehyde add to a free amino group (White g§_g;., 1968). Gustavson (1949) reported that the main reaction of aldehydes is their reaction with the e-amino group on lysine to form mostly -NHCH20H structures or less often, a methylene bridge. Using formaldehyde as an example of aldehydes, the frequently encountered reaction is the addition to a compound having an active hydrogen atom with the formation of a hydroxymethyl compound, equation (1). The hydroxy compound formed is usually reactive and may condense with another active hydrogen group forming the methylene bridge, equation (2) (Dexter and Edsall, 1945): (1) RH + CHZO—F—‘R-CHZWH), (2) R-CH2(0H) + HR'=-=!RCH2-R' + H20. Fraenkel-Conrat et al., (1950) discribed the main reactive groups in protein for binding formaldehyde to be the "primary amines". Also, they reported that carbonyl, peptide, or "secondary amines" groups do not react appreciably 15 with aldehydes. Because of their many reactive groups, proteins can combine with aldehydes and crosslinking can be pro— duced (Bjorksten, 1951). Crosslinkage via the methylene bridge is one example. Solubilization with Alkgli Limited amounts of alkali have the property of - dispersing soy protein without degradation. Kelley and Pressey (1966) were able to solubilize a globulin sample ‘ 2! after isoelectric precipitation. Due to disulfide poly- merization, normally the protein would be 20% less soluble upon redispersion. Lundgren (1949) also noted the ability of alkali to solublize the protein. Kelley and Pressey noted an increase in viscosity of a protein solution when alkali was added. They claim the NaOH caused unfolding of the globular proteins. They also suggested that dissociation took place because of a shift in the ultra- centrifugal pattern when the s values showed a shift 20,w from 2, 7, 11, and 158 to 3 an 58 material. S -d su fi e te ct o The cysteine and cystine residues form the basis for formation or unlinking of the disulfide polymers found in the soy protein. The conformation and structure of the globulins can be altered by reducing the disulfide bonds with an excess of sulfhydryl compounds (Wolf and l6 Tamura, 1969). Kelley and Pressey (1966) suggest that the sulfhydryl groups can interchange with the disulfides naturally present in the globulins thus breaking and form- ing new disulfide bonds. Jensen (1959) described a chain type reaction whereby one sulfhydryl initiator molecule could react with a disulfide bond forming a new disulfide bond and a new sulfhydryl initiator. This sequence can repeat, and, acting like a zipper, cause a whole series of sulfhydryl-disulfide interchanges. Henika and Rogers (1965) have described a series of f pr0posed reactions taking place in the disulfide-linked matrix of wheat flour protein when both an added reducing and oxidizing agent are present. The following reactions are considered: (3) RSH + PSSP—éPSH + RSSP, (4) RSH + asap—apes + RSSR, (5) KBro3 +- 3RSH + 3PSH——->3RSSP + KBr +3H20, (6) 1(er3 + 6RSH——>3RSSR + KBr + 31120, (7) KBrO3 + 6PSH—->3PSSP + KBr + 31120, where RSH is cysteine and P is a flour protein. Reactions 3 and 5 would allow unfolding of a protein, and reactions 6 and 7 would permit reformation of the disulfide matrix. METHODS AND MATERIALS Soybeans A sufficient amount of 1970 cr0p, Michigan grown, Harosoy (Var. 63) soybeans was frozen for three days at -2000. Freezing facilitated grinding and minimized protein denaturation due to heat generation. Fat.Extractio The frozen soybeans were cracked by subjecting them to a Waring blendor treatment for 10 sec. The cracked frozen beans were ground in a Thomas-Wiley Mill model ED-5. The 2 mm sieve was used. The ground full-fat meal was held at 2°C until extracted. Extraction was conducted for 48 hours using anhydrous ethyl ether in a Soxhlet extractor to remove essentially all the oil while minimizing the possibility of heat denaturation (Toriumi, 1970). The extraction thimbles used were Whatman 43 x 123 mm double thickness cellulose. ration of t e Cru e ote in Extrac The defatted meal was reground in the Wiley mill using the 0.5 mm sieve. The resulting finely ground meal represented the starting material for protein extraction. 17 18 Table 1 indicates that approximately 50% of the meal was 100 mesh or smaller. Its protein content, determined by the Kjeldahl method, was 49%. A nitrogen conversion factor of 6.25 was used. TABIE 1.--Particle Size Distribution of the Defatted Soybean.Mea1 Used for Protein Extraction Sieve Size Sieve Weight U. 8. Alternate Opening Fraction, % Mesh Designation Microns 100 149 48 50 297 28 30 595 22 >30 -—— 2 A 50 g sample of meal and 250 ml of deionized distilled water were combined in a Waring blendor. The slurry was blended until a temperature of 40°C was reached (approximately 10 minutes) while adding 1 N NaOH dropwise to yield a pH of 8.3. The slurry was then centrifuged for 30 minutes at 18,000 x g in a Sorvall RC2-B refrig— erated centrifuge using the GSA rotor with a minimum radius at tip of 5.75 inches. The temperature of the mixture during centrifugation was held in a range from 30-3500. After centrifuging, the supernatant was retained as the crude protein extract. The protein was obtained from the supernatant by either one of two methods. 19 Ammonium Sulfate Precipitation The "salting out" phenomenon forms the basis of this procedure (White g£_n;., 1968). When treated with reagent grade (NHZ4)2804 at a concentration of 75% of saturation, the maximum amount of protein contained in the extract comes out of solution. The protein-salt mixture was cen- trifuged for 30 minutes at 18,000 x g with the same centrifuge configuration as before to precipitate the protein. The supernatant was discarded, the protein pellet was washed with deionized distilled water, and then diapersed in deionized distilled water with the final protein concentration about 6%. The protein was dialized for one week against deionized distilled water at 2°C. After four days, dialysis against water gave negative results for tests for both the NHI ion and the 302 ion (Layde, 1961). The dialysis tubing was 1 7/8 inch cellulose casing obtained from the Food Products Division of the Union Carbide Corporation. After dialysis, the protein was considered to be at zero ionic strength. This material was the protein used to determine the solubility of the native protein at various ionic strengths. Isgelegtrig Ezegipitntion Except for the ionic strength-solubility experiment, all the protein used in this study was obtained by the isoelectric method, which is the method used in commercial preparation (Wolf, 1970). The extract is brought to its 20 region of minimum solubility by dr0pwise addition of l N HCl to adjust the pH from about 6.7 to 4.5. The acidified protein was spun for 3 minutes in an International Model CS centrifuge at 2,400 x g using the 277 rotor with maximum radius at the tip of 7.88 inches. The supernatant was discarded and the protein pellet resuSpended in deionized distilled water, mechanically stirred, and reprecipitated by centrifuging as before. The process of washing was repeated 5 times while keeping the pH at 4.5. After the final wash, the protein pellet was resuSpended in deionized water to provide a native protein material of various {concentrations depending on the experiment. Likewise, the pH was adjusted with acid or base depending on the desired pH for a particular test. Table 2 summarizes the sequences. Viscosity Measurements The viscosity measurements of protein solutions in this study were made using an Oswald type capillary bore viscometer (Triebold and Aurand, 1963). The u-tube viscometer was mounted in a water bath maintained at 25°i= 0.100. Five ml samples were employed, and the flow times through the viscometer were measured using a stopwatch. The specific gravities of the samples were obtained at the same temperature as the flow rates based on a ratio of weight of equal volumes using a pycnometer. The Specific gravity values were converted to relative density (wt./vol.) by multiplying the specific gravity values by the density of 21 TABLE 2.--Sequence for Protein Extraction. W 1. Frozen Soybeans 1.1. Cracking 1.2. Grinding (2 mm screen) 1.3. Soxhlet oil extraction 2. Defatted meal 2.1. Grind (0.5 mm screen) 2.2. Aqueous alkali-meal slurry (5:1) 2.3. Clarification of protein extract 3. Protein removal from slurry 3.1. Salting out 3.11. (NH4) SO4 (75% of saturation) 3.12. Centr fugation for protein pre- cipitation 3.13. Surface wash of precipitate 3.14. DiSpersion of precipitate 3.15. Dialysis 3.2. Isoelectric precipitation 3.21. Addition of HCl to pH 4.5 3.22. Centrifugation 3.23. Repeated washing (5 x), dispersion, and centrifugation at pH 4.5 22 water at 25°C. The relative viscosity of a liquid, based on flow through a capillary, is a modification of Poiseuille's formula, and is represented by the formula given below, provided the same volume, temperature, and viscometer are used: ‘3: =:El£; =.relative viscosity (Triebold and Aurand, 1963). ’70 d0‘50 The <11 and t1 refer to the density and flow time through the capillary for the sample, and do and to are corrSSpond- ing values for water. The‘flb is equal to one. Protein Solubility The method used to determine the solubility of the soy protein in all experiments of this study is described below. Five to ten ml samples of the protein solution were placed in 12 ml centrifuge tubes and spun in a Lourdes Model AX centrifuge at 16,000 x g for 10 minutes using the 9RA-24 rotor. The protein remaining in the supernatant was considered soluble protein. The values were reported as follows: mg pzotgin in snpernntant x 100, Percentage Soluble Protein== mg protein in sample Nitrogen Determinations The micro-Kjeldahl method and the Biuret colorimetric method of Gornall et al., (1949) were used. The Kjeldahl procedure was used for dry samples and when reagents 23 caused interference in the normal color development of the Biuret reagent. The Kjeldahl nitrogen values were converted to percentage protein by using a factor of 6.25 times the nitrogen content. The formula for percentage protein is given below: Percentage protein_ me Of H01 titer me Wt N 6 2 O . '_ sample weig mg The colorimetric procedure was used in conjunction with the Bausch and Lomb Spectronic 20 SpectrOphotometer set at 550 nm to determine the percentage transmittance of the developed samples. The transmittance values were related to protein concentration (mg/ml) through a standard curve. The standard curve was constructed by using soy protein samples of known protein content determined by Kjeldahl analysis. Heat Treatment The treatment consisted of placing samples in test tubes and lowering them into a mechanically stirred oil bath held at 85-9500. The oil bath consisted of a 400 ml beaker fitted with a styrofoam t0p with holes to hold the test tubes. The beaker was placed on t0p a Corning hot plate. The length of time for the heat treatment varied with the nature of the test. RESULTS Neutral Salts Effect on Soy Protein Solubility This experiment was conducted using soy protein obtained by dialysis of an aqueous extract which was obtained by precipitation with (NH4)2804. By the end of the dialysis period (one week), the globulin fraction had precipitated out; the complete contents of the dialysis casing, however, were used for the solubility experiment. Sufficient reagent grade sodium chloride (Nallinckrodt) was added to 100 m1 volumetric flasks to obtain the ionic strengths required. The volumetric flasks were then filled to the mark with the dialized protein. Ten m1 aliquots were Spun in the Lourdes centrifuge according to procedure outlined earlier. For this test, as shown in Table 3, the soluble protein is recorded as mg/ml of protein remaining in the supernatant after Spinning. This experiment was conducted to assess the per- formance of the protein preparation as compared with that in the literature with reSpect to solubility in the native state. The results in Table 3 at pH 7.0, indicate solubil- ity is the greatest for zero ionic strength; it reached a low point with 0.01 N NaCl after which solubility increased 24 25 TABLE 3.--Solubility of Soy Protein at Various Ionic Strengthsa Solubility of protein.mg/ml ionic strength pH 0.00 0.01 0.05 0.10 0.50 3.5 36 24 29 29 9 4.0 31 19 27 23 8 4.5 4 6 8 5.0 3 2 4 5 8 6.0 8 5 8 7 9 7.0 28 11 18 22 10 aFigures are the average of two trials using dialized soy protein and NaCl to adjust ionic strength. until it was supressed by 0.5 N NaCl. It should be noted that while these results show the general trend reported by Smith et al., (1938), the region of minimum solubility occurred in a 0.01 ionic strength solution (for an univalent salt, normality is equal to ionic strength) which is a factor of 10 less than that reported by Smith and his coworkers. The use of a dialized protein extract in this present experiment instead of the crude salt or water extract used by Smith.g§_nl., (1938), may have accounted for the increased sensitivity of the protein at lower ionic strengths. 25 Aldehydes and Heat: Their Effects on the Solubility and Relative Ennngsity of Soy Protein Four aldehydes were selected for this experiment. Formaldehyde (Mallinckrodt) is the first of the series of aliphatic aldehydes and is very reactive. Glutaraldehyde (Sigma Chemical), a dialdehyde, is also very reactive by virtue of its ability to link with two free amino groups. Although these two aldehydes can not be used in a food system, a study of their performance might be valuable. Glucose (Mallinckrodt) and DL-Glyceraldehyde (Nutritional Biochemical) were studied to compare their results with the results of the more active aldehydes, while realizing the better suitability of these aldehydes in a food system. Effects on Solubility The aldehydes were used in a ratio of 7:1 protein to aldehyde. The initial protein concentration was 6%. The protein was obtained by the isoelectric procedure, and it consisted mainly of the globulin fraction. The effect of aldehydes on solubility was checked before and after heating. The aldehydes were added to dry 50 or 100 ml beakers, and the protein added to the beakers with stirring. The pH was adjusted to 7.0 with l N NaOH if necessary. After taking 10 ml aliquots for the heat treatment and non-heat treat- ment segments of the experiment, the pH was lowered to the next required value by dr0pwise addition of HCl and aliquots taken as before. This process was repeated throughout the 27 pH range of the SXperiment. After 10 minutes heating as described earlier, all samples were centrifuged according to the "Protein Solubility" procedure. Table 4 reports the solubility data as an average of two trials. The decreased solubility is shown at pH 3.0 for glutaraldehyde, formaldehyde, and glucose with glutaraldehyde affecting a marked decrease in solubility. All solubility values are minimal at pH 4.5 because soy protein exhibits minimum solubility at this point. Effects on Relative Vigcosiny The effect of the aldehyde treatment on the viscosity of the soy protein was monitored with the Ostwald type visco- meter. Soy protein, obtained by isoelectric separation, was combined with the aldehydes in a 7:1 protein to aldehyde ratio. The protein concentration was 1.1%. The pH was adjusted to 7.0 with l N acid or base, and a 10 minute heat treatment given to one half of the samples. Five ml aliquots were transferred to the viscometer and allowed to equilibrate to 2500. Triplicate samples were run for viscosity measurements. Table 5 indicates that generally, with the exception of glutaraldehyde, those samples given the heat treatment appear to have lower relative viscosity values. The dif- ference between individual aldehyde samples, however, with reSpect to heat and non-heat treatment is not different from the two control samples with the exception of glutaraldehyde. It is possible that glutaraldehyde caused some crosslinking 28 .Huu canon pdomeon 0p cfiouonm spas A.monm oaapooaoomav dfioponm Re no moHQSSm Ha 0H mo mamaap can no ommaopm one monsmfimw mm mm mm mm om om mm mm mm mm 0.5 we on mm as me em em mm ma em o.e m N m m m m m H o m m.e me an an me am me e a me am o.m seem - seen 02 seem . seem 02 seam . seam 02 seam . peas 02 seem . seem oz an Honpdoo mmoosao ocmnmoamnoomao oeaeoeaeneseao ochnocawsaom hpflaamsaom maoponm omwpcoohom g spasm eds neesnoeaa_spas cantata daoponm new no mundaneaomna.e mumaa 29 TABLE 5.--Relative Viscosity of Soy Protein after Treatment with.A1dehydeS and Heata Relative Viscosity Sample No Heat Heat Water 1.00 1.00 Control 1.42 1.40 Formaldehyde 1.51 1.49 Glutaraldehyde 1.41 1.52 Glyceraldehyde 1.43 1.41 Glucose 1.43 1.40 8Figures are the average of three trials of 1.1% solutions of 5 ml soy protein (isoelectric prep.) with protein to reagent ratio 7:1 and pH:= 7.0. 30 that changed the flow characteristics of the protein system. Reduction, Oxidation, and Heat Treatment Using Cysteine and Potassium Bromate: Their Effect on Solubility and Relative Viscosity Free-base L-cysteine (Sigma Chemical) was used as a reducing agent to cleave the disulfide bonds of the globulin fractions of the soy protein. Potassium bromate (Baker Chemical) was used as an oxidizing agent to reform disulfide bonds after reduction. The protein for this experiment was the globulin fractions obtained by isoelectric separation. Effects on Solubility This experiment involved adding cysteine and potassium bromate to a 3% protein solution (pH 7.5) in a 6:1 protein to reagent ratio. Preliminary experiments indicated a maximum decrease in solubility when this protein to reagent ratio was used. Generally the sequence of events was to add the protein to the cysteine contained in the bottom of a large test tube followed by a period of heat treatment. The protein-cysteine was then added to the potassium bromate contained in the bottom of another large test tube. The system was then given a period of additional heat treatment. Five m1 aliquots of the samples were then subjected to the protein solubility procedure as previously described. The steps in the procedure are listed below: (1) Protein added to cysteine and mixed, 31 (2) Heat treatment, (3) Protein-cysteine added to potassium bromate, (4) Heat treatment, (5) Percentage solubility determined. The results of this experiment are shown in table 6. Trial 1, the control, indicates the solubility of the protein with only a 5 minute heat treatment. Trials 2 and 3 show that either cysteine or potassium bromate singularly do not appreciably effect the solubility. Trial 4, however, shows an 80% decrease in solubility when both cysteine and potassium bromate are contained in the sequence. Trial 6 indicates that the heat treatment is not necessary after cysteine incorporation, but observation has shown the decrease in solubility is more rapid with both heat treat- ments after cysteine and potassium bromate addition. Trials 9 through 13 show the effect of varying the potassium bromate concentration. After the addition of potassium.bromate, visual observance showed that the protein system first became Opaque followed by precipitation. The pH change was measured when the various reagents were incorporated in the protein system. This indicated that the decrease in solubility was due to some factor other than a simple decrease in pH to the isoelectric region where soy protein has very limited solubility. Table 7 shows that a pH drOp of 0.7 of a unit occurs when both cysteine and bromate were added. 32 TABLE 6.--Percentage Soluble Soy Protein after Reduction, Oxidation, and Heat Treatment8 Sample Cysteine K—Bromate Heat reat. Soluble No. mg mg min Protein(%) 1 -- -- 5 87 2 25 -- 5 77 3 -- 25 5-5 87 4 25 25 5-5 20 5 25 25 2-5 29 6 25 25 0-5 13 7 25 25 1—5 12 8 10 25 0-5 56 9 25 5 1-5 73 10 25 10 1-5 58 11 25 15 1-5 28 12 25 20 1-5 21 13 25 25 1-5 12 aA11 samples are 5 ml of 3% solution of soy protein (isoelectric prep.). Each figure shown is an average of three trials. bHeat treat. numbers refer to treatment given after the addition of each reagent. Trial 3 indicates 5 min. heat given protein before and after addition of bromate. 33 TABLE 7.--Change in pH of Soy Protein after treatment with Cysteine, Potassium Bromate and Heat Heat Samplea Treat. pH P 0 6.7 P, Cys 0 6.8 P, Cys 1 6.7 P, Cys, Bromate O 6.0 P, Cys, Bromate 5 6.0 P, Bromate O 6.5 P, Bromate 5 6.5 aP = Protein; Cys:: Cysteine; Bromate = Potassium Bromate. Effect on Relative Viscosity The relative viscosity was monitored after treating 1% soy protein solution with cysteine, potassium bromate, and varying lengths of heat treatment. The protein was prepared by the isoelectric separation method to obtain the globulin fraction. The protein to reagent ratio for the tests was 6:1. I As in the previous experiment using cysteine and potassium bromate, the protein was added to the reagents and mixed. Three 35 ml protein samples were used. Five ml aliquots were used for viscosity measurements while approximately 25 ml aliquots were used for density deter- minations. The first sample was given a 5 minute heat 34 treatment followed by cooling in an ice bath. The viscosity and density measurements were then taken at 25°C. The second 35 m1 sample was given the initial 5 minute heat treatment, cooled, and then added to 60 mg of cysteine. The sample was mixed and given 1 additional minute of heat after which it was cooled before taking viscosity and density measurements. The third sample was given the same treatment as sample 2 through the second cooling period after which it was added to 60 mg of potassium bromate. Viscosity and density measurements were obtained without additional heating, and subsequent viscosity measurements were obtained after additional heat of 2, 6, and K>minute intervals. Table 8 shows the results obtained from typical samples subjected to the treatments described above. The results show that cysteine caused a decrease in relative viscosity from 1.32 to 1.25. Likewise, after treating with potassium bromate, the viscosity decreased initially to 1.19 before the heat treatment after which the viscosity eventually increased to 1.26. There was no evidence of the protein system starting to come out of solution during the test. Preliminary tests have shown that the cysteine- bromate-heat treatment will coagulate 1% soy protein, but uninterrupted heating of 20 to 30 minutes is required. 35 TABLE 8.--Relative Viscosity of Soy Protein after Reduction, Oxidation, and Heat Treatment8 W Treatmentb Sample Cumulative Heat Heat Relative Sample min. min. Viscosity P 5 5 1.32 P, Cys 1 6 1.25 P, Cys, Bromate 0 6 1.19 P, Cys, Bromate 2 8 1.24 P, Cys, Bromate 4 12 1.25 P, Cys, Bromate 4 16 1.26 aFigures shown are typical for viscosity changes occurring when 57ml samples of 1% solution of soy protein (isoelectric prep.) are treated in succession with cysteine, gotassium bromate, and heat. Protein to reagent ratio is :1. bP: Protein; Cys : Cysteine; Bromate 2 Potassium Bromate. 36 Effect of Calcium Chloride on Soy Protein Solubility This experiment was conducted to assess the per- formance of CaC12 in decreasing the solubility of the globulin fraction. One molar CaCl2 was added to 5 ml samples of 6.6% protein solution at pH 7.0. The salt concentrations of the protein samples and the decrease in solubility are shown in Table 9. The maximum decrease in solubility occurs at the lowest salt concentration used. Subsequently, the solubility increases and then starts to decrease as the molar concentration of CaCl2 increases. A check of the pH after CaC12 treatment showed that it decreased to 6.9 from the initial value of 7.0. TABLE 9.--Solubility of Soy Protein when Treated with Calcium Chloride8 Percentage CaCl Protein Trial No. Molarigy Solubility 1 0.02 38 2 0.09 42 3 0.26 61 4 0.36 61 5 0.44 60 6 0.52 55 aFigures show the average of two trials of 5 m1 soy protein samples (isoelectric prep.). 'il‘l-‘Jliilll DISCUSSION The solubility of soy protein has been decreased by three methods including treatment with aldehydes, reduction and oxidation, and incorporation of a neutral salt. Effect of Neutral Salts The use of neutral salts in conjunction with proteins is not new, and salts are probably used in the coagulating bath to produce the soy protein fibers in the Spun fiber process. Reports in the literature mention the use of an "acid-salt" bath. Lundgren (1949), reported that the protein is "forced through spinnerettes into a precipitating bath consisting usually of aqueous solutions of acids, inorganic salts, and, frequently, a heavy metal ion." Effect of Calcium Chloride The observation by Smith n§_nl., (1938) that CaClg produced a white flocculent precipitate has been confirmed in this work. The instantaneous effect of the Ca012 leads this writer to agree with earlier work that indicated Ca012 may form complexes with the protein (Briggs and Mann, 1958; Zittle and DellaMonica, 1957). 37 38 Effect of Aldehydes As noted in the results, the only appreciable change occurred with glutaraldehyde at a pH of 3.0 where the soy protein would normally increase in solubility if the pH is adjusted downward from the isoelectric region. Visual inSpection of the protein pellet after centrifugation showed it to be a rubbery mass that exhibited limited elastic prOperties. This would indicate that crosslinking, probably between the e-amino groups of the lysine residues, had taken place. Reports in the literature indicate this possibility (Gustavson, 1949; Fraenkel-Conrat §§_nl., 1950; Bjorksten, 1951). Formaldehyde also decreased the solubility at pH 3.0 to a lesser degree which may indicate that some crosslinking via the methylene bridge occurred. Again the protein pellet after centrifugation showed in- dications of crosslinkage. The lower reactivity of glycer- aldehyde and glucose prevented crosslinking from occurring. It is interesting that the crosslinking was observed only at the very acid pH. The crosslinking may not occur until the protein precipitated out at the isoelectric region creating high protein concentrations in localized areas and enhancing crosslinking conditions. The protein to aldehyde ratio would be increased, but Gustavson (1949) pointed out that a vast structural intergrity can be imparted to a molecular system with very few covalent bridges being formed. When the pH was made more acid, the cross- linked protein resisted solubilization. 39 Effect of Reduction, Oxidation, and Heat The decreased solubility by the addition of cysteine, potassium bromate, and heat is most likely the result of disulfide bond cleavage of intramolecular (cystine residues) and intermolecular bonds followed by reformation of disulfide bonds which may not involve the identical locations of the original disulfides. If cystine residues were cleaved on adjacent peptide chains, it would be possible upon reoxidation for new intermolecular bonds to be formed. This also suggests the possibility of more than the original number of disulfide bonds. The linking of soy protein globulins by disulfide bonds, accompanied by heat, causes aggregation of the globulin subunits which leads to precipitation. The effect of reduction, oxidation, and heat on the viscosity is interesting; a possible explanation follows. One might eXpect that since cysteine decreased the viscosity, potassium bromate should increase it because of the disulfide bond formation creating a matrix that would impede inter- molecular flow. Kelley and Pressey (1966) explained that when soy protein is treated with alkali, chain unfolding results. When, as in this experiment, cysteine or another reducing agent is added in excess, there should be maximum cleaving of disulfides taking place. This should have the effect of lowering the viscosity because movement of peptide chains past each other would be easier. It may be possible that when 40 excess bromate was added, as described before, more than the original complement of disulfides could be formed at new locations. The disulfide bonds could cause a tightening of the peptide chains leading to configurational changes that would decrease intermolecular flow interference as indicated by lower viscosity measurements. The eventual increase in viscosity was probably the result of the heat causing aggregation which would be an intermediate step before precipitation occurs. Preliminary experiments showed, as mentioned earlier, that precipitation of 1% protein will occur when the heat duration is extended to 20 or 30 minutes. The combined action of the reduction, oxidation, and heat phenomenon along with the use of neutral salts might be adaptable to the soy protein spun-fiber process. As an example, the spinning dOpe would be made up in the con- ventional manner (Lundgren, 1949), except that a predetermined amount of cysteine would be added to the spinning dOpe before final mixing in a colloidal mill. The incorporation of cysteine would cause disulfide bond cleavage. If a more desirable reorientation or longitudinal alignment of the peptide chains resulted, then the incorporation of cysteine may be advantageous. The Spinning dope would be extruded as normal through the Spinnerettes into a coagulating bath which would not consist of acid, but it would be a combination of heated potassium bromate and a divalent cationic salt to reform the disulfide bonds and 41 cause insolubility. The coagulation via the cysteine- bromate-heat effect would probably be aided by incorporating the salt because the protein, not being at the isoelectric region, would exhibit a net charge which causes repulsion of the peptide chains. Zittle and DellaMonica (1957) indicate that, by incorporating a calcium salt into fi-lactoglobulin, the positive calcium ion reduces the net charge on the protein molecule and precipitation occurs at pH 7.4 with very small concentrations of the calcium salt. The investigation of these ideas could lead to improvements in the technology of the Spun fiber foods or they could lead to other departures not requiring the expensive spinning process. SUMMARY AND CONCLUSIONS Three methods were studied to decrease the solubility of soy protein. The protein used for the solubility studies was extracted and isolated from Harosoy soybeans using both the "salting out" and the isoelectric precipitation methods. The three methods used to decrease the solubility were: (1) Incorporation of aldehydes, (2) Reduction, oxidation, and heat treatment, (3) Treatment with a neutral salt. The effects of the experiments were monitored by measuring the change in percentage soluble protein and change in viscosity. All the methods were successful to varying degrees. Results of the aldehyde study demonstrated the ability of glutaraldehyde to form crosslinks and effectively reduce the protein solubility at a low pH. Glutaraldehyde also decreased the viscosity of soy protein from that of the control when both were given heat treatment. Form- aldehyde also showed some evidence of crosslinking at a low pH. Glyceraldehyde and glucose did not appreciably effect the solubility of the soy protein. 42 43 The soy protein solubility was also effectively decreased with the cysteine-bromate—heat treatment. The cysteine acted as a reducing agent to cleave disulfide bonds while the bromate reoxidized the sulfhydryls to form new disulfide bonds. The treatment was sequential with cysteine being added before bromate, and heat treatments given after addition of each reagent. The treatment caused precipitation of the protein and decreased solubility about 80%. The treatment also caused changes in the viscosity of the protein. Cysteine reduced the viscosity of the protein. When bromate was added, it reduced the viscosity further followed by gradual increases after intervals of heat treatment. The soy protein solubility was also decreased by treatment with a divalent cationic salt. CaCl2 (0.02.M) decreased the percentage solubility about 50%. Information from this study may be helpful in suggesting alternate ways of imparting improved structure to soy proteins when used to produce fabricated food.systems. BIBLIOGRAPHY Belter, P. A. and Smith A. K. 1952. Protein Denaturation in Soybean Meal During Processing. J. Am. Oil Chem. SOC. 2931700 Bjorksten, J. 1951. Cross Linkages in Protein Chemistry. In.Advances in Protein Chemistry, Vol. 6 ed. Anson, 11'. L., Edsall, J. T., and Baily, in, p. 343. Acedemic Press, New York. Briggs, D. R. and Mann, R. L. 1950. .An Electrophoretic Analysis of Soybean Protein. Cereal Chem. 27:243. Briggs, D. R. and Wolf, W. J. 1957. Studies on the Cold- Insoluble Fraction of the WaterAExtractable Soybean Protein. 1. Polymerization of the 118 Component through Reaction of Sulfhydryl Groups to form Disulfide Bonds. Arch. Biochem. Biophys. 72:127. Catsimpoolas, N., Leuthner, E., and Neyer, E. W. 1968. Studies on the Characterization of Soybean Protein by ImmgnoelectrOphoresis. Arch. Biochem. Biophys. 127:33 . Chervenka, C. H. 1969. A Manual of Methods for the Analytical Ultracentrifuge p. 87. Published by Spinco Div. of Beckman Instruments, Inc., Palo Alto, California. Dexter, F. and Edsall J. T. 1945. The Reactions of Formaldehyde with Amino Acids and Proteins. In Advances in Protein Chemistry, Vol. 2 ed. Anson, M. L., Edsall, J. T., p. 281. Acedemic Press, New York. Dimler, R. J. 1969. Oilseed Proteins. Chem. Eng. Prog. 65(9):20. Fraenkel-Conrat H., COOper, N., and Olcott, H. S. 1945. The Reaction of Formaldehyde with Proteins. J. Am. Chem. Soc. 67:950. Gornall. A. G., Bardawill, J. C., and David, M. M. 1949. Determination of Serum Proteins by means of the Biuret Reaction. J. Biol. Chem. 177:751. 44 45 Gustavson, K. H. 1949. Some Protein-Chemical ASpects of Tanning Processes. In Advances in Protein Chemistry, Vol. 5, ed. Anson, M. L., Edsall, J. T., and Baily, K., p. 406. Acedemic Press, New York. Henika, R. G. and Rogers, N. E. 1965. Reactions of Cysteine, Bromate, and Whey in a Rapid Breadmaking Process. Cereal Chem. 42:397. Jensen, V. E. 1959. Sulfhydryl-Disulfide Interchange. Science 13081319. Kelley, J. J. and Pressey, R. 1966. Studies with Soybean Protein and Fiber Formation. Cereal Chem. 43:195. Layde, D. C. 1961. Introduction to Qualitative Analysis, pp. 290, 299. Allyn and Bacon, Inc., Boston. Lundgren, H. P. 1949. Synthetic Fibers Made From Proteins. In.Advances in Protein Chemistry, Vol. 5, ed. Anson, M. L., Edsall, J. T., and Baily, K., p. 305. Acedemic Press, New York. Mann, R. L. and Briggs, D. R. 1950. Effects of Solvent and Heat Treatments on Soybean Proteins as Evidenced by Electrophoretic Analysis. Cereal Chem. 27:358. {Morse, W. J. 1950. History of Soybean Production. In Soybeans and Soybean Production, Vol. I, ed. Markley, K. S., p. 3. Interscience Pub1., New York. Nagel, R. H., Becker, H. 0., and Milner, R. T. 1938a. The Solubility of Some Constituents of Soybean Meal in Alcohol-Water Solutions. Cereal Chem. 15:766. Nagel, R. H., Becker, H. 0., and Milner, R. T. 1938b. Some Physical Factors Affecting the DiSpersion of Soybean Proteins in water. Cereal Chem. 15:463. Nash, A. M. and Wolf, W. J. 1967. Solubility and Ultra- centrafugal Studies on Soybean Globulins. Cereal Chem. 44:183. Osborne, T. B. and Campbell, G. F. 1898. Proteids of the Soybean. J. Am. Chem. Soc. 20:419. Pomeranz, Y. 1966. Soy Flour in Breadmaking: A Review of its Chemical Composition, Nutritional Value and Functional Properties. Bakers Digest 40:45. Rakosky, J. 1971. Vegetable ProteinFMeat Mixtures. The National Provisioner 164:13. 46 Roberts, R. C. and Briggs, D. R. 1963. Characteristics of the Various Soybean Globulin Components with Respect to Denaturization by Ethanol. Cereal Chem. 40:450. Smiley, W. G. and Smith, A. K. 1946. Preparation and Nitrggen Content of Soybean Protein. Cereal Chem. 2382 0 Smith, A. K., Circle S. J., and Brother, G. A. 1938. Peptization of Soybean Proteins. The Effect of Neutral Salts on the Quality of Nitrogenous Constituents Extracted from Oil-Free Meal. J. Am. Chem. Soc. 60:1316. Toriumi, H. 1970. Effects of Gamma Radiation on water Holding Capacity and Nitrogen Solubility of Isolated Soybean Protein. M.S. thesis, Michigan State University, East Lansing, Michigan. Triebold, H. A. and.Aurand, L. W. 1963. Food Composition anthnalysis, p. 91. VanNostrand Reinhold C0., New York 0 White, A., Handler, P., and Smith E. L. 1968. Principals of Biochemistry, pp. 87-137. McGraw-Hill, New York. Wolf, W. J. 1969. Soybean Protein Nomenclature: A Progress Report. Cereal Sci. Today l4(3):76. Wolf, W. J. 1970. Soybean Proteins: Their Functional cgemécal, and Physical Properties. Agr. Food Chem. 1 :9 9. Wolf, W. J. and Briggs, D. R. 1956. Ultracentrifugal Investigation of the Effect of Neutral Salts on the Extraction of Soybean Proteins. Arch. Biochem. Biophys. 63:40. Wolf, W. J. and Briggs, D. R. 1958. Studies on the Cold Insoluble Fraction of Water Extractable Soybean Proteins. II. Factors Influencing Conformation Changes in the 118 Component. Arch. Biochem. Biophys e 76 3377. Wolf, W. J. and Briggs, D. R. 1959. Purification and Characterization of the 118 Component of Soybean Proteins. Arch. Biochem. Biophys. 85:186. ‘Wolf, W. J., Eldridge, A. C., and Babcock, G. E. 1963. Physical Properties of AlcoholAExtracted Soybean Proteins. Cereal Chem. 40:504 47 Wolf, W. J., Sly D. A. and Babcock, G. E. 1964. Denaturation of Soybean Globulins by Aqueous ISOpropanol. Cereal Chem. 41:328. Zittle, C. A. and DellaMonica E. S. 1957. The Viscosity and Opacity of Heated d-Lactoglobulin Solutions: The Effect of Salts, and Oxidizing and Reducing Reagents. J. Am. Chem. SOC. 7931260 HNI 366