COVALENT BINDING OF METHIONINE AND TRYPTOPHAN TO SOY PROTEIN fey LEANDROS PANAGIS VOUTSINAS B.Sc., University of Thessaloniki, 1969 A THESIS SUBMITTED IN PARTIAL FULFILLMENT THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n THE FACULTY OF GRADUATE STUDIES (Dept. of Food Science) " We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October, 1978 @ Leandros Panagis Voutsinas, 1978 In presenting th i s thes is in pa r t i a l fu l f i lment of the requirements for an advanced degree at the Univers i ty of B r i t i s h Columbia, I agree that the L ibrary sha l l make it f ree ly ava i l ab le for reference and study. I fur ther agree that permission for extensive copying of th i s thes is for scho lar ly purposes may be granted by the Head of my Department or by his representat ives. It is understood that copying or pub l i ca t ion of th is thes is for f inanc ia l gain sha l l not be allowed without my writ ten permission. Leandros P. Voutsinas Department of Food Science The Univers i ty of B r i t i s h Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date October 23, 1978 i i . ABSTRACT One common method of improving the nutrit ional quality of certain food proteins is through fort i f i cat ion with necessary amounts of l imit ing essential amino acids. This simple and convenient method, however, is not the best. Several disadvantages are asso-ciated with the addition of free amino acids to food proteins, such as changes in flavor and color, losses of added amino acids during food processing or cooking, differences in s tab i l i ty and metabolism between free amino acids and amino acids in proteins. Covalent attachment of the l imit ing amino acids, however, should eliminate these problems, and moreover, could improve the nutrit ional and functional properties of food proteins. In this study, therefore, an attempt to improve the nutrit ional value of the soy protein was made by using the carbodiimide condensation reaction to covalently bind methionine and tryptophan to soy protein. In order to confine as much as possible the binding, of amino acid to the protein a-carboxyl groups the soy protein isolate (SPI) was part ia l ly hydrolyzed with pepsin to increase the number of a-carboxyls in soy protein. Various conditions of the carbodiimide reaction were analysed by a fractional factor ia l design in an attempt to determine the factors affecting the amino acid binding to soy protein hydrolysate (SPH). Of the factors investigated, pH, SPH concentration, carbodiimide concentration, activation time and reaction time were found to s ignif icantly affect the methionine binding efficiency, whereas pH, SPH concentration, carbodiimide concentration, amino acid concentration and reaction temperature were found to s ignif icantly influence the tryptophan binding efficiency to SPH. To determine the best level for each of the selected factors an optimization of the carbodiimide reaction conditions was conducted by carrying out another factor ia l experiment. Thus, under the best condition found, the methionine and tryptophan contents of methionine - and tryptophan-bound SPH samples were increased 7.7-fold and 18.0-fold, respectively. An jin v i tro pepsin-pancreatin digestion test demonstrated that the bound amino acids were readily released. In order to improve the low yie ld of the f ina l product, another analysis of the carbodiimide reaction conditions was carried out. Since the yie ld could not be markedly improved by this factor ia l design in which peptic SPH was used, SPI without preliminary hydrolysis was used as the starting material. A product with 95-99% yield was obtained and its methionine or tryptophan content was increased 6.3-fold or 11.3-fold, respectively. High d iges t ib i l i ty was s t i l l maintained for these products. Gel f i l t r a t i o n chromatography demonstrated that the carbodiimide reaction caused an increase in the molecular weights of soy protein fractions. Furthermore, gel f i l t r a t i o n chromatography revealed that., there was no selective amino acid binding among the different soy protein fractions during the carbodiimide reaction. i v . TABLE OF CONTENTS PAGE INTRODUCTION 1 LITERATURE REVIEW 5 A. The chemistry of carbodiimides 5 B. Physical and chemical properties of 14 soybean proteins a. Solubil i ty 14 b. Molecular size distribution 16 c. Association-Dissociation reactions 17 d. Subunit structure 17 e. Amino acid composition 19 MATERIALS AND METHODS 20 A. Materials 20 B. Preparation of soy protein hydrolysate 21 C. S ta t i s t i ca l analysis 21 D. Covalent binding of methionine and tryptophan to soy protein hydrolysate 21 E. Methionine determination 22 F. Tryptophan determination 23 G. Nitrogen determination 23 H. Amino acid analysis 23 I. Enzymatic digestion test 24 J . Gel f i l t r a t i o n chromatography 24 1. Preparation and operation 24 2. Solubil ization of samples 25 RESULTS AND DISCUSSION 26 A. Preparation of soy protein hydrolysate 26 B. S ta t i s t i ca l analysis 27 C. Amino acid analysis 61 D. Enzymatic digestion test 64 E. Gel f i l t r a t i o n chromatography 67 GENERAL DISCUSSION 80 CONCLUSION 83 LITERATURE CITED 85 APPENDIX 95 V. LIST OF TABLES TABLE PAGE 1 Factors and the i r assigned l eve l s invest igated 28 for t h e i r poss ib le in f luence on amino ac id binding to SPH. 2 Ana lys i s of var iance of the 2 - l eve l f r a c t i o n a l 29 f a c t o r i a l experiment for methionine binding to SPH. 3 Ana lys i s of var iance of the 2 - l eve l f r a c t i o n a l 30 f a c t o r i a l experiment for tryptophan binding to SPH. 4 Assigned l eve l s for fac tors in f luenc ing the amino 40 ac id b inding to SPH. 5 Ana lys i s of var iance of the 3 - l eve l f a c t o r i a l 41 experiment fo r methionine binding to SPH. 6 Analys i s of var iance of the 3 - l eve l f a c t o r i a l 42 experiment for tryptophan binding to SPH. 7 Factors a f f e c t i n g amino ac id b inding and t h e i r 56 best l e v e l s . 8 Assigned l eve l s for fac tor s invest igated by a 57 3 - l eve l f a c t o r i a l experiment for t h e i r poss ib le in f luence on f i n a l product ' s y i e l d . 9 Ana lys i s of var iance of the 3 - l eve l f a c t o r i a l 58 experiment fo r s e l e c t i o n of the factors i n f luenc ing the methionine-bound SPH y i e l d s and opt imizat ion. 10 Ana lys i s of var iance of the 3 - l eve l f a c t o r i a l 59 experiment fo r s e l e c t i o n of the fac tor s i n f l uenc ing the tryptophan-bound SPH y i e l d s and opt imizat ion. 11 Factors a f f e c t i n g f i n a l product ' s y i e l d and the i r 60 best l e v e l s . 12 Amino ac id composition (g/16gN) of SPI and amino 62 acid-bound soy p ro te in samples with the highest coupl ing y i e l d s . Liberated amino acid (g/16gN) from SPI and amino acid-bound soy protein samples with the highest coupling yields by pepsin-pancreatin digestion iji v i t ro . Proteins used in cal ibration of Sephadex G-200 column eluted with 0.1M sodium phosphate buffer, pH 7.0, and containing 0.1% SDS. v i i . LIST OF FIGURES FIGURE PAGE 1 A mechanism proposed by Khorana (1955) fo r 6 the carbodiimide-mediated peptide synthes i s . 2 E f f e c t of pH on the rate of methionine and 32 tryptophan binding to SPH. 3 E f fec t of SPH concentrat ion on the ra te of 33 methionine and tryptophan binding to SPH. 4 E f f e c t of EDC concentrat ion on the ra te of 34 methionine and tryptophan b ind ing to SPH. 5 E f f e c t of a c t i v a t i o n time on the ra te of methionine 35 binding to SPH. 6 E f f e c t of reac t i on time on the ra te of methionine 36 binding to SPH. 7 E f f e c t of amino ac id (tryptophan) concentrat ion 37 on the rate of tryptophan binding to SPH. 8 E f f e c t of reac t i on temperature on the ra te of 38 tryptophan binding to SPH. 9 E f f e c t of pH on the ra te of methionine and tryptophan 43 b inding to SPH. 10 E f f e c t of SPH concentrat ion on the rate of methionine 46 and tryptophan binding to SPH. 11 E f f ec t of EDC concentrat ion on the ra te of methionine 47 and tryptophan binding to SPH. 12 E f f ec t of amino ac id (tryptophan) concentrat ion on 48 the rate of tryptophan b inding to SPH. 13 E f f e c t of a c t i v a t i o n time on the ra te of methionine 50 binding to SPH. 14 E f f ec t of reac t ion time on the rate of methionine 52 binding to SPH. 15 E f f e c t of reac t i on temperature on the rate of 54 tryptophan binding to SPH. Calibration curve of Sephadex G-200. Gel f i l t r a t i o n pattern of SPH solubil ized with SDS on Sephadex G-200. Gel f i l t r a t i o n pattern of tryptophan-bound SPH solubil ized with SDS on Sephadex G-200. Gel f i l t r a t i o n patterns of SPH and tryptophan-bound SPH on Sephadex G-200. Gel f i l t r a t i o n pattern of SPH solubil ized with NaOH (at pH 12.0) on Sephadex G-200. Gel f i l t r a t i o n pattern of methionine-bound SPH solubilized with NaOH (at pH 12.0) on Sephadex G-200. Gel f i l t r a t i o n patterns of SPH and methionine-bound SPH on Sephadex G-200. Gel f i l t r a t i o n pattern (A280nm) °^ tryptophan-bound SPH solubil ized with NaOH (at pH 12.0) on Sephadex G-200. Gel f i l t r a t i o n pattern (A2]_0nm) °^ tryptophan-bound SPH solubilized with NaOH (at pH 12.0) on Sephadex G-200. ACKNOWLEDGEMENTS The author wishes to express h i s sincere appreciation to his supervisor, Dr. S. Nakai, Professor, Department of Food Science, for h i s constant advise, help and encouragement throughout the course of t h i s study, and i n the preparation of the th e s i s . He i s also thankful to the members of his graduate committee: Dr. W. D. Powrie, Head, Department of Food Science Dr. J . F. Richards, Professor, Department of Food Science Dr. J . Vanderstoep, Assistant Professor, Department of Food Science for t h e i r i n t e r e s t i n t h i s research and for the review of t h i s t h e s i s . F i n a n c i a l a i d , g r a t e f u l l y acknowledged, was provided by Canada Department of Agri c u l t u r e through support of the project. 1. INTRODUCTION The e f f i c i e n t use of a p ro te in by man or animal requ i res that i t contains the e s sen t i a l amino ac ids as w e l l as n i t rogen i n amounts and proport ions needed by the organism to meet i t s needs for s p e c i f i c as we l l as f o r general phy s i o l og i ca l funct ions . It i s we l l known, however, that with the poss ib le exception of only a very few, most prote ins do not contain the e s sen t i a l amino acids i n the amounts and proport ions needed by man or animal. Therefore, the i r u t i l i z a t i o n tends to be i n e f f i c i e n t , p a r t i c u l a r l y fo r p ro te in foods der ived from the vegetable kingdom (Bressani, 1975). The amino ac id d e f i c i e n c i e s of a p ro te in source i s usua l ly overcome (balanced), as has been w e l l documented, by the add i t ion of the appropriate amounts of the d e f i c i e n t amino ac ids ; these can be i n the form of e i ther c r y s t a l l i n e (synthet ic) compounds (prote in f o r t i f i c a t i o n or complementation) or as const i tuents of prote ins which are r i c h sources of the d e f i c i e n t amino acids (prote in supplementation). Add i t ion of synthet ic amino acids r e su l t s i n increased pro te in qua l i t y due to the improvement of the e s sen t i a l amino ac id pattern and due only to that , s ince there are no other parameters involved (Bressani, 1974). On the other hand, add i t ion of the d e f i c i e n t amino acids as p ro te in can make changes i n a l l three major parameters a f f e c t i n g pro te in u t i l i z a t i o n , namely, p ro te in q u a l i t y , p ro te in quant i ty , and the t o t a l energy value ( P e l l e t , 1974). F o r t i f i c a t i o n with amino acids i s the normal p r ac t i ce i n formu-l a t i n g animal feeds (Waddell, 1958). Studies by Howe et a l . (1965A, 1965B, 1967) have demonstrated the improvement i n qua l i t y which can be obtained by f o r t i f y i n g foods with amino ac ids . Bread supplemented with 2. lysine is a far superior source of protein than ordinary bread (Altschul, 1974). It has been estimated, moreover, that 10,000-20,000 tons of 1-lysine w i l l by 1980, be added to cereals intended for human food (Chem. Engng News, 1973). The rapid prol i ferat ion of meat analogs from soy and other vegetable protein sources w i l l i n -crease the use of methionine in foods based on meat and.milk model (Horan, 1974B). Bressani (1975), reported experimental results from human subjects indicating that addition of the l imit ing amino acids represents not only a better protein nutrit ion for the individual , but a significant economic saving as well , due to the increased efficiency with which protein is u t i l i z e d . Although the interest in amino acid for t i f i ca t ion of human foods is certainly on the increase, the u t i l i za t ion of this option, at the present time, is not suff ic iently used to make a significant contribution to human protein supply. F o r t i -f ication with amino acids, however, is expected to become an essential and regular adjunct of the human food system (Altschul, 1974). In this way, the technology w i l l contribute more f l e x i b i l i t y to counteract the pressures of greater populations and higher expectations. The fort i f i cat ion of food proteins with essential amino acids is an important, simple and convenient method for improving their nutrit ional quality. However, i t is not necessarily the best method of improving their nutrit ional quality. This is due to several defects associated with the addition of free amino acids to food proteins. F i r s t l y , the added amino acids can easily be lost during the subsequent food processing, cooking or through washing or discarding of the juices (Bressani, 1975; Fujimaki _e_t a l . , 1977). Secondly, i t may easily occur that the free amino acids suffer degradation or reaction with food ingredients (e.g. Strecker degradation or amino-carbonyl reaction) to affect the food qualit ies such as flavor and color (Yamashita et a l . , 1970). Thirdly, considering the methionine for t i f i ca t ion of foods, i t s addition is severely restricted, because of the unattractive flavor which methionine, in general, imparts to foods (Klaui, 1974). Furthermore, Beigler (1969) reported that methionine gives off an unpleasant odor upon high heating, such as s t er i l i za t ion . This finding was recently confirmed by Hippe and Warthesen (1978) who reported that methionine for t i f i ca t ion in products that receive high heat treatment is not feasible from a flavor standpoint because of the formation of methional and dimethyl disulf ide. A generally accepted reaction scheme for the production of flavor compounds from methionine is Strecker de-gradation (Ballance, 1961). Methional, a product of this reaction, has been reported to have a flavor threshold in the ppb range (Patton and Josephson, 1957). Fourthly, a difference in s tab i l i ty between free methionine and methionine in peptides has been reported by Arai (1974) , suggesting the existence of different s tabi l i ty characteristics between free amino acids and amino acids covalently bound to proteins. F i f th ly , the metabolism of amino acids ingested as proteins differs somewhat from that of free amino acids. More rapid intest inal transport of amino acids from small peptides than from the equivalent amount of free amino acid or amino acid mixture has been shown to occur in man and several other mammalian species by many techniques including tolerance tests, the use of tied loops of intestine or intest inal perfusion (Craft et a l . , 1968; Matthews et a l . , 1968; Adibi , 1971), 4. and measurements of uptake or transmural transport by small intestine in v i tro (Cheng et a l . , 1971; Rubins and Auricchio, 1971). Thus, differences have been reported in intest inal transport between ol igo-peptide methionine and free methionine (Lis e_t a l . , 1972) as well as between oligopeptide lysine and free lysine (Burston et a l , , 1972). One possible method for removing or lessening the above defects is to supplement the desirable amino acids in a form covalently bound with proteins or peptides in foods. The high nutrit ive value of the soybean protein has been recognised for many years and has been responsible for the enormous growth of i t s use in the animal feed industry (Horan, 1974A). Insofar as the food industry is concerned, soy proteins unt i l recently were selected or used primarily for their functional effects in imparting desirable properties to fabricated foods. Now, hoever, they are being incorporated for their nutrit ional value (Wolf, 1972A). Soy protein offers the only expanding commercial source of protein able to meet the growing demand for a nutri t ional ly balanced, high protein food suitable for both human and animal feeding (Coppock, 1974). Soy protein is a good source of a l l the essential amino acids except methionine and tryptophan (Wolf, 1972A). The specific objectives, therefore, of this work were: (1) Optimization of the carbodiimide condensation reaction used as a means for covalently binding methionine and tryptophan to soy protein; (2) evaluation of the nutrit ional quality of the chemically modified soy protein obtained under the best condition; and (3) determination of the effect, i f any, of the carbodiimide condensation reaction on the molecular weight distribution pattern of the soy protein. 5. LITERATURE REVIEW A. The chemistry of carbodiimides In 1955 Sheehan and Hess introduced a new method for the formation of peptide bonds using N,N'-dialkylated carbodiimides, preferably N ,N' -dicyclohexylcarbodiimide. This reagent became, in a short time, the most important coupling reagent. The carbodiimide method can be called the most popular method of peptide synthesis. The main reason for i t has probably been the high react ivity of carbodiimides, since this assures that the desired coupling reactions proceed smoothly, usually to completion of the acylation within a short time (Bodanszky and Ondetti, 1966). In addition to this , the simplicity and convenience of this technique as well as the commercial ava i lab i l i ty of the reagent are also an explanation for this unprecedented popularity. Racemization is slight (Anderson and Callahan, 1958) to negligible (Rebek and Fe i t l er , 1975) and this fact offers an additional advantage. The carbodiimide condensation reaction can be expresssed in the following equation: R-COOH+H2N-R' + R"-N=C=N-R" + R-CONH-R' + R"-NH-CO-NH-R"  The reaction mechanism was f i r s t discussed by Khorana (1953, 1955) and was more recently thoroughly reinvestigated by Detar and his coworkers (1966A,B,C). Khorana (1953) suggested that the reaction proceeds through the activation of the carboxyl group (I), ini t iated by the addition of i ts proton to the double bond of the carbodiimide(II), followed by the attack of the carboxylate anion to form the unstable reactive inter-mediate (III), the O-acylisourea ( F i g . l ) . This O-acylisourea(III), in turn, can undergo either an intramolecular rearrangement, v ia an oxygen to nitrogen (CKN) acyl transfer, to an N-acylurea(V) or conversion to two r o II R-C-OH + N-R' II [I] C II N-R' + R-C-0 N-R' I , H-N-R' o II II R-C-O-C I NH-R:' [II] [III] II 'I l-C-O-C N-R' I NH-R* [III] „ NH-R' R-C-O-C RCOO 0 II R-C-N-R' I 0=C I NH-R' [V] H N-R" NH-R' [IV] (a) H„N-R" X NH-R' NH-R' C=0 NH-R' [Va] 7 T ^ R-CO-NH-R" + C=0 * I [VII] NH-R' [Vb] + R-C=0 \ 0 / R-C=0 [VI] ] (b) H2N-R" R-CO-NH-R" [VII] + M R-C-OH — [I] Fig . 1. A mechanism proposed by Khorana (1955) for the carbodiimide-mediated peptide synthesis. 7. activated forms (IV and VI) of the original carboxyl component (e.g. carboxylic acids, peptides or proteins). In addition to the above mentioned poss ib i l i t i e s , Franzblau et_ a l . (1963) suggested that in certain cases, the 0-acylisourea could cycl ize. The f i r s t alternative, that is the formation of N-acylurea, has a chance to occur when there are no further protons available (Schroder and Lubke, 1965) or when no other nucleophile is present or the reaction of those present is slow or hindered (Kopple, 1966). If excess carboxyl component is the only ava i l -able reactant a second addition of proton takes place and the 0->-N acyl migration w i l l be prevented. Stabil ization occurs by way of formation of the symmetrical anhydride (VI) and disubstituted urea (Va). Although the intramolecular rearrangement of O-acylisourea is probably the most common route to the formation of N-acylureas, an alternative mechanism for their production, namely, by the subsequent acylation of the i n i t i a l l y formed disubstituted urea (Va) by the acid anhydride (VI), was discovered in experiments carried out by Smith et^ a l . (1958). It has a chance to occur when carbodiimide and carboxyl component are kept for a long time at room temperature and no other reactants are present. This poss ib i l i ty , however, has been subsequently excluded by De Tar and Si lverstein (1966A). The synthesis of peptide derivatives (VII), according to Khorana (1955), may proceed through one or more of the following three routes: (a) (IV)+NH2-R" -* (VII), since O-acylisourea has been found to be sensitive to bases and reacts readily with nucleophiles (Khorana, 1952); (b) acid anhydride (VI)+NH2-R" -> (VII); and (c) N-acylurea (V)+NH2-R" + (VII). Support for the above mechanism has been provided by Doleschall and Lempert (1963), who isolated a cycl ic O-acylisourea and examined i t s reactions with 8. different nucleophiles. The N-acylureas of the type (V) appear to be quite stable and show l i t t l e tendency to undergo condensation with an amino acid ester under mild conditions and thus conversion into desired peptide der i -vative (VII), (Khorana, 1953). It has been subsequently suggested, however, that N-acylureas are not activated compounds and therefore are unreactive toward amines and do not form peptide bonds. The N-acylurea (path c) has, thus, been excluded as an acylating agent (Diehl, 1957; De Tar and Si lverstein, 1966A, Klausner and Bodansky, 1972). Although the O-acylisourea is highly unstable and has not been isolated except in one very special case (Doleschall and Lempert, 1963), i t s occurrence is inferred from kinetic data and from the appearance in the reaction mixture of both the urea of the carbodiimide (as a by-product of peptide synthesis) and the N-acylurea resulting from the rearrangement of the proposed O-acylisourea (Steinman et^ a l . , 1966) . Recently a mechani-st ic study of the carbodiimide-mediated peptide synthesis has been reported by Rebek and Fei t ler (1973), who have determined the reaction intermediates. Their evidence indicated that the i n i t i a l acylating agent in the carbo-diimide reaction was distinguishable from the carboxylic acid anhydride under conditions normally encountered during peptide synthesis in solution. Their results ind icated a maximum of 55% (5°C) to 60% (30°C) as the extent of anhydride's participation during peptide synthesis in solution with N,N'-dicyclohexylcarbodiimide. Delayed addition of the amine component was found to decrease the importance of O-acylisourea (path a) and increase the extent of anhydride's participation (path b) in peptide synthesis. One advantage of N,N'-dicyclohexylcarbodiimide claimed by Sheehan 9. and Hess (1955) was the apparent lack of racemization in the formation of optically active peptide derivatives. However, the fact that coupling reactions promoted by carbodiimides may sometimes be attended by appre-ciable racemization was indicated by the finding of Hofmann e_t a l . (1958) and was promptly reaffirmed by Anderson and Callahan (1958), who have demonstrated an influence of solvent and temperature on the racemization. Low temperatures (-5°C) reduce signif icantly the amount of racemic mixture (DL-form). Recently Rebek and Fei t ler (1975), however, working on the peptide synthesis with carbodiimide, reported that racemization was in the range of 0.01 to 0.1%. From the racemization point of view, symmetrical anhydrides may be undesirable intermediates, since racemization might be expected from the intermediate anhydride on the basis of the ready azlactone formation (Bodanszky and Ondetti, 1966). Intermediary azlactone formation is the most l ike ly course that racemization takes when occurring during a condensation reaction (Schroder and Lubke, 1965). With the widely used dicyclohexylcarbodiimide reagent, racemization increases with temper-ature and is more pronounced in polar solvents such as dimethylformamide than in nonpolar, e .g . , C ^ C ^ (Bodanszky et_ a l . , 1976). The most efficient method to reduce racemization in coupling with carbodiimide is to use additives such as N-hydroxysuccinimide or 1-hydroxybenzotriazole.. Unfortunately the use of additives cannot be taken as a guarantee against racemization (Bodanszky e_t a l . , 1976) . The undesired formation of. N-acylurea is frequently observed (Khorana, 1955; Merrif ie ld and Woolley, 1956; Shankman and Schvo, 1958; and Sheehan e_t al^., 1956). It is dependent upon a number of factors. Low temperatures reduce i ts formation (Helferich and Boshagen, 1959; Schneider, 1960); therefore the reaction is normally carried out at a temperature of 0°C 10. or lower (du Vigneaud t^_ a l . , 1957; Katsoyannis and Suzuki, 1962; Sheehan and Yang, 1958). However,in a few cases the reaction is carried out at higher temperatures (Theodoropoulos and Folsch, 1958; Previero et_ al., 1973). The formation of N-acylurea is suppressed in methylene chloride (Merrif ield, 1958; Merrif ie ld and Woolley, 1958; Sheehan et a l . , 1956) or acetonitri le (Benoiton and Rydon, 1960; Sheehan et a l . , 1956). Khorana (1955) pointed out that the relative proportions of N-acylureas formed during the coupling reaction w i l l depend largely upon the particular peptide derivative used and the solvent employed and that their formation even in relat ively small amounts may complicate the isolation of the desired products in a satisfactory y ie ld . Note was made too, that, since the rearrangement of O-acylisourea(III) to the N-acylurea (V) is intramolecular whereas the reactions (III) -»- (IV) -> (VI) and (IV) •> (VII) are bimolecular (intermolecular), the formation of the desired products should be favored by an increase in the concentration of the reactants. Tometsko (1973) i l lustrated that the activated carboxyl groups during peptide synthesis are not stable for long periods of time, since the formation of N-acylurea derivatives results in a loss of a c t i -vated carboxyl component and therefore decrease in peptide bond formation. The inactivation is a complex process that could also involve solvent impurities, temperature effects, etc. One problem associated with the use of the original carbodiimides, part icularly N,N'-dicyclohexylcarbodiimide, was the purif ication of the desired products. On one hand, the N-acylureas have structures and so lubi l i ty rather similar to those of the desired peptides and therefore i t is sometimes d i f f i c u l t to remove N-acylureas in order to obtain a pure 11. product. On the other hand, as pointed out by Khorana (1955), the insoluble co-product (dicyclohexylurea) which is formed during the implication of dicyclohexylcarbodiimide as the condensing agent in peptide syntheses could sometimes prove undesirable as, for example, in cases involving the synthesis of the less soluble higher poly-peptides, since the co-product and the peptide derivative may have similar so lubi l i ty properties, thus complicating the isolat ion procedure. In order to overcome the purif ication problems caused by the formation of by-product ureas and N-acylureas, other carbodiimides with more favorable so lubi l i ty properties have also been investigated (Sheehan and Hlavka, 1956; Sheehan et a l . , 1961). With these reagents, the by-product ureas and N-acylureas as well as any excess of reagent are easily removed by washing with dilute acid or water. Peptides have been synthesized in high yields and in very pure forms with these carbodiimides (Sheehan jit a l . , 1961). The most generally useful of them have been found to be 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (hereafter abbreviated to EDC) and the corresponding hydrochloride salt (Sheehan et a l . , 1961). These and the other water-soluble carbodiimides (WSC) are commercially available and the reaction can s t i l l be carried out under mild conditions (Greenstein and Winitz, 1961). The reaction mechanism of these water-soluble carbodiimides has been assumed to be similar to that proposed in 1953 by Khorana (Hoare and Koshland, 1966). Although a l l water-soluble carbodiimides react s imi larly , the smaller reagents might be expected to have greater access to part ia l ly buried carboxyl groups. This expectation appeared correct in a recent study with tobacco mosaic virus protein (Means and Feeney, 1971), in which upto three additional carboxyl groups could be modified with EDC which did not react with l-cyclohexyl-3-(2-morpholinyl-4-ethyl) carbodiimide metho-p-toluenesulfonate (the other most widely used water-soluble carbodiimide) . Sheehan ejt a l . , (1965) working with EDC demonstrated that the synthesis of pure oligopeptides was extremely rapid and fac i l e . Although coupling was usually about 80% after 1 hr at room temperature, the time required for complete coupling varied from 24 to 66 hr. The use of EDC has continued to expand because of i ts su i tabi l i ty to peptide synthesis and i t s high s tab i l i ty . The reaction between a water-soluble carbodiimide and the functional groups of a protein could be both numerous and complex. It is of obvious interest, therefore, to review the reactions of these unusual reagents with proteins. Sheehan and Hlavka (1957) published the f i r s t studies of the reaction between a protein and a water-soluble carbodiimide (in the absence of an added nucleophile), stating that the reaction produced cross-linking in gelatin. This cross-linking was presumably due to amide formation occuring between side-chain carboxyl and amino groups of the gelatin molecule. On the other hand, by carrying out the reaction between a protein and a water-soluble carbodiimide at pH values (pH 4.0), where coupling by nucleophilic groups (such as e-amino group of lysine) on the protein should be total ly inhibited, Franzblau e_t aJL. (1963) have noted that the reaction between gelatin and a water-soluble carbodiimide, in the presence of hydroxylamine as nucleophile, yielded hydroxamic acid derivatives of side-chain aspartic acid and glutamic acid residues. These workers also reported that a-carboxyl groups reacted more rapidly and quantitatively to produce hydroxamic acid derivatives than 6- or y-carboxyl groups which exhibited lower yields (about 50% and 40%, respec-t ive ly ) . Of great interest is the work of Previero e_t a l . (1973), who 13. reported that EDC could affect the carboxyl groups of an acidic peptide in different ways depending on their environment, namely, they demon-strated that the same reagent could be used for both selective activation of the C-terminal and blocking of the side-chain carboxyl groups in a single operation. These workers, thus, presented evidence which indicated that direct condensation without preincubation of an acidic peptide with EDC resulted in the coupling of two times as many amino acid ester moles as in the case of condensation after preincubation with EDC for 90 min. at 40°C. Their proposed explanation was that, when condensation took place after preincubation, the O-acylisourea i n i t i a l l y formed isomerized to the inert N-acylurea derivative at the side-chain carboxyl group, while the C-terminal remained activated as an Oxazolinone (Azlactone) which could subsequently react with a suitable nucleophile. Direct condensation without preincubation, on the other hand, resulted in the participation of both C-terminal and side-chain carboxyl groups in the formation of peptide bonds. Riehm and Scheraga (1966), by studying the reaction between ribonuclease and a water-soluble carbodiimide (in the absence of nucleophiles), have investigated a l l the possible reactions between a protein and a water-soluble carbodiimide. They demonstrated that the reaction between ribonuclease and carbodiimide, at low pH (4.5) did not involve direct reaction between the reagent and amino or imidazole groups (nucleophilic groups) in the protein. This finding might be expected since at pH 4.5 most ( i f not a l l ) of these groups would be protonated. On the other hand, the reaction between the protein and the water-soluble carbodiimide at high pH (9.5), followed by amino acid analyses, resulted in a product which contained fewer lysine residues 14. than did a ribonuclease hydrolysate. Therefore, i t was concluded that the reaction between a protein and a water-soluble carbodiimide, at high pH, results in the guanidination of e-amino groups, due to reaction between carbodiimide and e-amino groups of lysine residues. The same authors also excluded any reaction between carbodiimide and side-chain hydroxyl groups of threonine, serine, and tyrosine residues. This finding is in accordance with the results of Sheehan j2t al. (1955, 1956), who reported that the hydroxyl function of threonine, serine, and hydroxyproline does not require masking prior to condensation. Hoare and Koshland (1966) have used a water-soluble carbodiimide for the selective modification of carboxyl groups in proteins in the presence of high concentration of nucleophiles. They have reported a procedure leading to rapid and quantitative modification of carboxyl groups under mild conditions. Horinishi et a l . (1968) have also explored the reaction of water-soluble carbodiimides with proteins in the presence of nucleo-philes. They reported that with undenatured proteins, part ia l reaction is commonly observed and seemed to be an indication of the degree of exposure of the carboxyl groups, since only the more accessible or reactive carboxyl groups reacted. In the presence of denaturants (e.g. guanidine), however, quantitative modification of protein carboxyl groups could be obtained. B. Physical and chemical properties of soybean proteins (a) Solubil i ty The majority of soy-proteins are c lass i f ied as globulins. Such proteins are insoluble in water in the region of their isoelectric 15. points, but w i l l dissolve in the isoelectric state when salts such as sodium or calcium chloride are added. If the pH is above or below the isoelectric point, a globulin w i l l also dissolve in aqueous solutions in the absence of salts (Wolf and Cowan, 1975). When defatted meal is dispersed in d i s t i l l e d water at pH 6.5, nearly maximum protein so lubi l i ty occurs. Raising the pH with a l k a l i increases the so lubi l i ty s l ight ly , but adding acid reduces the so lubi l i ty abruptly to a minimum in the range of pH 4-5, which is the isoelectric region. At lower pH values, the proteins become positively charged and redissolve (Wolf, 1972A). Solubil i ty of the soybean globulins is also influenced by the co-existence of phytate in soybeans (Smith and Rackis, 1957). Another factor that affects so lubi l i ty of soybean proteins is disulfide polymerization (Wolf, 1972B). Commercial soyprotein isolates vary in so lubi l i ty because of processing variations (Nash and Wolf, 1967). Part of the insolubi l i ty of isolates arises during the isoelectric precipitation step (Nash et a l . , 1971). One reaction responsible for this insolubi l i ty appears to be the formation of protein polymers linked by disulfide bonds. When the isolates are treated with sulfhydryl compounds such as cysteine or mercaptoethanol, depolymerization occurs and part ia l so lubi l i ty is regained. A portion of the isolates, however, remains insoluble even with added sulfhydryl compounds (Nash e_t a l . , 1971). Solubi l i ty of the proteins is often required in order to obtain desired functional properties, and because soluble ingredients are always easier to formulate into foods. Consequently, almost a l l soy protein concentrates and isolates are neutra-l ized and sold as proteinates, with sodium proteinate the major form available. 16. (b) Molecular size distribution Undenatured soy proteins are a complex mixture of molecules that differ in size, charge, and structure (Wolf, 1975). The distribution in molecular weight ranges from 8,000 to about 600,000 as demonstrated by ultragentrifugation (Wolf, 1970) or by gel f i l t r a t i o n (Hasegawa et a l . (1963). An ultracentrifuge separates the water-extractable proteins of defatted meal into four major fractions. These fractions are designated 2,7,11, and 15S based on their sedimentation rates (coefficients). The 7S and 11S fractions comprise more than 60% of the total protein, and about 80% of the proteins have molecular weights of 100,000 and higher. The 2S fraction, representing about 20% of the total protein, contains the low molecular-weight proteins: several trypsin inhibitors , cytochrome C, allantoinase, and 2 globulins with unknown biological act iv i ty . The 7S fraction makes up more than one-third of the total protein and contains at least four different types of proteins: beta-amylase, four hemagglutinins, two or more lipoxygenases, and a protein called 7S globulin (Wolf and Cowan, 1975). The 7S globulin is a glycoprotein (Koshiyama, 1969) and makes up over one-half of the total 7S fraction (Wolf and Sly, 1967). About one-third of the total protein is found in the 11S fraction. So far only one protein called the 11S globulin or glycinin has been found in this sedimenting fraction. The 11S protein has a molecular weight of about 350,000, which is typical of major globulins of other seeds (Wolf, 1972B). The 15S fraction makes up the remainder of the protein. This fraction has not been isolated and characterized, but based on i t s sedimentation rate appears to have a molecular weight of over half a mi l l ion . Some workers have suggested 17. that this fraction is a polymer of the 11S protein (Wolf, 1975). The wide distribution of molecular sizes among the water-extractable soybean proteins is also observed by gel f i l t r a t i o n . Seven protein fractions have been fractionated (Hasegawa e_t a l . , 1963) as compared to only four in the ultragentrifuge. Obara and Kimura (1967), however, working on the gel f i l t r a t i o n fractionation of the water-extractable soybean proteins, separated only four protein fractions. Additional evidence for the complexity of the soybean proteins has been obtained by hydroxylapatite chromatography (Wolf and Sly, 1965), starch gel electrophoresis (Puski and Melnychyn, 1968; Shibasaki and Okubo, 1966), and Immunoelectrophoresis (Catsimpoolas et_ al,, 1968). (c) Association-Dissociation reactions A characteristic property of the 7S and 11S globulins is their ab i l i ty to undergo rapid and reversible association-dissociation re-actions under mild changes in ionic environment (Wolf, 1975). Thus, the 7S globulin at pH 7.6, 0.5 ionic strength exists as a monomer with a molecular weight of 180,000 to 200,000, but when the ionic strength is lowered to 0.1, the protein dimerizes to a 370,000 mol. wt. species (Koshiyama, 1968A). Likewise, the 11S globulin forms a faster-sedimenting form when ionic strength is changed from 0.5 to 0.1, but the extent of association is low (Naismith, 1955). When ionic strength is reduced furthermore, to 0.01, the 11S protein dissociates into smaller units that sediment as 7S and 2 to 3S entitles (Wolf and Briggs, 1958). (d) Subunit structure The 7S and 11S globulins possess an additional property that adds to their complexity. Both proteins are built-up of smaller units (sub-18. units) that interact very specif ical ly to form the parent globulins, and there appear to be no sub-units common to the two proteins (Wolf, 1972A). The 7S globulin contains 9 amino-terminal residues, a fact that indicates that the molecule contains at least 9 polypeptide chains (Koshiyama, 1968B). The 7S structure is disrupted by the usual reagents capable of dissociating proteins into subunits. When the protein is dissolved in 8M urea or 4M guanidine hydrochloride, the mol. wt. decreases from 180,000 for the native form to 22,500-24,000 (Koshiyama, 1971) . Dissociation into subunits also occurs on treatment with the anionic detergent sodium dodecyl sulfate (SDS). In SDS solution, the molecular weight of 7S globulin is 34,000 (Wolf, 1972A). Binding of about 40 molecules of detergent per subunit would account for the d is -crepancy in molecular weight under the different conditions of protein dissociation (detergent versus urea or guanidine hydrochloride). At high pH (0.01N NaOH) the 7S protein changes to a form with a sedimentation coefficient of only 0.4S (Koshiyama, 1968A). The 11S globulin has a quaternary structure consisting of 12 subunits. Isoelectric focusing of 11S protein in urea-mercaptoethanol solution, however, separates only 6 subunits. These results suggest that the 11S molecule (mol. wt. of 350,000) is actually a dimer of 2 identical monomers, each consisting of 6 subunits (Catsimpoolas, 1969). The quaternary structure of 11S mole-cule is disrupted under various conditions, including low ionic strength, high or low pH, high concentrations of urea, anionic detergents (e.g. SDS), and temperatures above 80°C (Wolf, 1972B). For reasons s t i l l unknown, the 11S.globulin is more susceptible to irreversible changes in structure than the 7S globulin (Wolf, 1972A). 19. (e) Amino acid composition The chemical composition of soy proteins is very important because of i t s relationship to nutrit ion and reactions of the proteins during processing. Soy proteins are a good source of a l l the essential amino acids except methionine and tryptophan (Wolf, 1972A). Methionine is the f i r s t l imit ing amino acid of soy proteins (Berry 'e t 'a l . , 1962) and this deficiency needs to be considered when the proteins are used for nutrit ional purposes (Coppock, 1974). Largely on the basis of animal experiments, i t has been generally assumed that the nutrit ive value of soybean protein products for human consumption would be enhanced by fort i f i cat ion with methionine (Liener, 1972). The available evidence indicates that meals of soybean products, f lour, or protein concentrates are improved in protein quality upon for t i f i ca t ion with methionine (Berry e_t a l . , 1962, 1966; Evans and McGinnis, 1946). Methionine has been added at 0.5% of soy protein in infant formulas and such formulas have been judged equivalent in protein value to the proteins of meat, egg, and milk (Fomon e_t a l . , 1973) . Some studies also suggest that tryptophan or threonine is the second l imiting amino acid of soy protein. The improve-ment in protein quality, however, brought about by adding methionine is so high that any other amino acid added causes only small increases (Bressani, 1975). Soybean proteins being high in lysine, the f i r s t l imit ing amino acid in cereals (Kato and Muramatsu, 1971), are useful as supplementing blends (Bookwalter et ' a l . , 1971; Horan, 1973). A mixture of soy protein concentrate and wheat flour is nutri t ional ly superior to either protein source alone; thus, a soy: wheat protein ratio of 3:1 gives an optimum protein efficiency ratio (Wilding et a l . , 20. 1968). The various soybean protein fractions differ in overall amino acid contents. Insofar as amino acid composition is concerned, the most significant differences between the 7S globulins and the 11S globulins are the 5- to 6-fold higher contents of methionine, tryptophan, and %-cystine in the latter (Wolf, 1972B). MATERIALS AND METHODS A. Materials Soy protein isolate used in this study was a product of General M i l l s , Inc. , Minneapolis, MN. Pepsin (Hog stomach mucosa 2X : crystal l ized and lyophilized powder), pancreatin (porcine pancreas grade VI) , L-trypto-phan, and 1-ethy1-3-(3-dimethylaminopropyl)carbodiimide hydrochloride, were a l l products of Sigma Chemical Company. L-methionine and p-dimethyl-aminobenzaldehyde (practical) were obtained from Eastman Organic Chemicals. Sodium nitroprusside was purchased from Fisher sc ient i f ic company. Pronase (B grade) was from Calbiochem, Los Angeles. 5-Sulfosal icyl ic acid (98%) was obtained from Aldrich Chemical Company, Inc.. Sodium dodecyl sulfate (SDS), specially pure 99%, was purchased from BDH Chemicals, Canada. The standard proteins, used in this study for molecular weight cal ibrat ion, and their sources were: ribonuclease A (bovine pancreas, type I-A, 5X crystall ized) , a- chymotrypsin (bovine pancreas, 3X crysta l l ized) , serum albumin (Bovine), and catalase (bovine l iver) were a l l products of Sigma Chemical Company; myoglobin (sperm whale) was from Schwartz/Mann. The Sephadex G-200 gel (particle diameter 40-120y), and the Blue Dextran 2000, were products of Pharmacia Fine Chemicals, Upsala, Sweden. 21. B. Preparation of soy protein hydrolysate 12 g of soy protein isolate was dispersed in 500 ml of d i s t i l l e d water. The pH of the mixture was adjusted with HCI to 1.6. To this dispersion was added 120 mg of pepsin dissolved in 100 ml of d i s t i l l e d water, and the mixture was incubated at 35°C for a specified time (5 or 16 hr) . The hydrolysis was stopped by neutralizing (pH 7.0) the incubation mixture with NaOH. The resulting product was, then, centrifuged at 9,150 x g for 15 min., and the supernatant was freeze-dried to obtain a powdered soybean protein hydrolysate. C. S ta t i s t i ca l analysis (Selection of inf luential factors and deter- mination of their best levels) . Various conditions of the carbodiimide condensation reaction were investigated by a 2-level fractional factor ia l experiment (Taguchi, 1957; Appendix), in an attempt to determine the factors affecting the amino acid binding to soy protein. To determine the best level for each of the selected inf luent ia l factors a 3-level factor ia l experiment was used (Taguchi, 1957; Appendix). In order to improve the low yie ld of the f inal product obtained ( i . e . amino acid-coupled soy protein hydrolysate), another 3-level factor ia l analysis of various conditions of the carbodiimide reaction was carried out, thus determining the factors influencing the f ina l product's yield as well as their best levels. A l l s ta t i s t i ca l data obtained in the above mentioned factor ia l experiments were analysed by a Monroe Model 1880 programmable calculator. D. Covalent binding of methionine and tryptophan to soy protein hydrolysate A specified amount of soy protein hydrolysate was dissolved in 15 ml of d i s t i l l e d water at room temperature. The pH of the solution was 22. adjusted to the appropriate level (set by the experiment). A specified amount of 1-ethy1-3-(3-dimethylaminopropyl)carbodiimide hydrochloride dissolved in 2 ml of d i s t i l l e d water was, then, added to the soy protein hydrolysate solution. Absolute ethanol (1.1 ml) was also added, i f re-quired by the experiment. After the specified time for the act ivat ion 3 of the protein carboxyl groups, a specified amount of L-methionine or L -tryptophan in solution^ was added to the reaction mixture. The pH of the mixture was checked and adjusted to the level set by the experiment. The reaction was allowed to proceed at a specified temperature for a specified time. Then, 1.8 ml of g lac ia l acetic acid was added to the mixture to stop the reaction (by destroying the excess EDC). Subsequently, the sample was dialysed in the cold (5°C) against 2 changes of 1M acetic acid for 1% day, against running tap water for % day, and f ina l ly against one change of d i s t i l l e d deionized water for % day. Then, the sample was neutralized (pH 7.5), st irred for 10 min (on a magnetic s t i r r e r ) , and freeze-dried. E. Methionine determination The methionine content of various samples was determined by the sodium nitroprusside colorimetric method of Lunder (1974/1975), with the following modifications: (a) 0.5 g sample was mixed with 2.5 ml HC1 20% into a hydro-lys is tube. The tube was sealed with vacuum grease, carefully closed with a metallic clamp, and then placed in an a ir over at 120°C for 10 hours; (b) 450 mg of activated charcoal (12 x 30 mesh, Witco Chemical, N.Y.) were used to discolourize the protein hydrolysate; and (c) 10 ml of 25% ethyl alcohol was used to wash the activated charcoal and elute the methionine adsorbed on i t . The remainder of the procedure If no activation was required by the experiment ( i . e . , activation time 0 min.) , then, the addition of amino acid solution to the soyprotein hydrolysate solution preceded the carbodiimide addition, k The amino acid solution was prepared by adding drops of IN HC1 to a st irred dispersion of a specified amount of amino acid in 4 ml of d i s t i l l e d water and bringing to the volume (5 ml) by adding dis t . water. 23. was carried out exactly as described by Lunder (1974/1975). F. Tryptophan determination The tryptophan content of various samples was determined by the spectrophotometric method of Spies (1967) after hydrolysis of samples with pronase (procedure W). The p-dimethylaminobenzaldehyde (practical) was recrystal l ized twice from ethanol by adding d i s t i l l e d water to a warm (about 50°C) ethanolic solution of p-dimethylaminobenzaldehyde unt i l crysta l l izat ion was just in i t iated and then cooling the solution. A suspension of s l ight ly greenish-white crystals was obtained. These crystals were harvested by suction f i l t r a t i o n , a ir -dr ied , and kept in a desiccator away from light unt i l used. G. Nitrogen determination The nitrogen content of soy protein isolate, soy protein hydrolysate, as well as methionine- or tryptophan-bound soy protein isolate and soy protein hydrolysate samples obtained under best reaction conditions was determined by the rapid micro-Kjeldahl procedure of Concon and Soltess (1973). The protein content of each sample was then calculated by multi-plying the amount of i ts nitrogen content by a factor of 6.25. H. Amino acid analysis The samples were hydrolysed with p-toluenesulfonic acid according to the method of Liu and Chang (1971), at 110°C for 24 hr. Amino acids were analysed on a single-column system (Durrum Chem. Corp., Palo Alto , CA.) attached to a Phoenix Model M6800 Amino Acid Analyzer (Phoenix Precision Instrument Co.) . 24. I. Enzymatic digestion test The d iges t ib i l i ty of various samples was determined by the original pepsin-pancreatin digest method of Akeson and Stahmann (1964) as modified by Stahmann and Woldegiorgis (1975), with the following change: to 5.0 ml of digestion mixture was added 5.0 ml of 14% sul fosa l icy l ic acid. The flask was then shaken for 15-30 minutes (in a shaking water bath at 37°C) and removed from the shaker. The pH was adjusted to 2.2 by adding drops of 6N NaOH. The digest was then transferred to a 15 ml graduated tube and brought to volume by adding pH 2.2 c i trate sample di luter . The sample was then centrifuged at 1,000 x g for 30 minutes, f i l tered through a dry Whatman No. 1 f i l t e r paper, then through a sintered glass UF f i l t e r direct ly into a sample tube, and stored in a freezer at -20°C or in a cold room (5°C) for amino acid analysis. J . Gel f i l t r a t i o n chromatography (1) Preparation and operation Gel f i l t r a t i o n was carried out on Sephadex G-200 in a glass chromato-graphic column (Fisher and Porter) with an internal diameter of 2.0 cm and a length of 100 cm. The packed column was stabil ized and equilibrated with 2 bed volumes of elution buffer (0.1M sodium phosphate buffer, pH 7.0, containing 0.02% sodium azide, to prevent microbial growth, and 0.1% SDS) in an upward flow direction. The operating pressure was adjusted to give a flow rate of about 10.4 ml per hour during equilibration and subsequent runs. The column was operated in an upward flow direction. The inlet tubing was joined to a 3-way valve which was connected to a funnel ( f i l l ed with elution buffer) and a 5 ml syringe. Sample containing 150 mg protein in 5.0 ml solution was applied to the column via the syringe. The valve 25. was then turned to connect the eluant to the column. The effluent was collected by a fraction collector (Gilson Medical Electronics, Wisconsin) set to take 5.2 ml fractions. The elution pattern of each protein sample was obtained by measuring the absorbance of each fraction in a 1-cm c e l l at 280 nm by a Beckman DB spectrophotometer, and plotting the value against the fraction's number direct ly on a recording chart by means of a Varicord recorder Model 43 (Photovolt Corporation, N.Y.) attached to the spectrophotometer. The void volume (V ) and the total bed volume (V^ _) of the column o t were determined from the elution volumes of Blue Dextran 2000 and tryptophan, respectively, using absorbances at 280 nm. The column was calibrated with standards proteins. (2) Solubil ization of samples The methionine- or tryptophan-bound soy protein hydrolysate or soy protein isolate samples with the highest coupling yields were insoluble in the phosphate buffer (eluant). Thus, different dissociation procedures were used to solubi l ize the amino acid-bound soy protein hydrolysate samples (which exhibited much greater so lubi l i ty than the corresponding amino acid-bound soy protein isolate samples). The tryptophan-bound soy protein hydrolysate sample was solubil ized as follows: in 5.0 ml of elution buffer were added 20 mg SDS and 1 drop of 2-mercaptoethanol. To this solution was dispersed 150 mg of tryptophan-bound soy protein hydrolysate and the mixture was st irred on a magnetic s t i r rer . Moderate heat was applied to fac i l i ta te the dissolution. The sample after being completely solubilized was f i l tered through a dry Whatman No. 1 f i l t e r paper (to remove any invis ible micro-part ic le ) , and then applied to the column. Although, the soy protein 26. hydrolysate (control) was soluble in the elution buffer, the same solubi l izat ion procedure was applied to i t in order to have a better control. Since the methionine-coupled soy protein hydrolysate sample exhibited slight so lubi l i ty under the above described conditions, the following procedure was used for i t s solubi l izat ion: 150 mg of methione-bound soy protein hydrolysate were added to 5.0 ml of d i s t i l l e d water. The pH of the dispersion was adjusted to 12.0 by NaOH and the mixture was st irred at room temperature for 30 min, then centrifuged at 3,090 x g for 15 min and the supernatant was applied to the column. The same method was used to solubi l ize 150 mg of soy protein hydrolysate (control) and 150 mg of tryptophan-bound soy protein hydrolysate sample. However, by this solubi l izat ion procedure the tryptophan-bound SPH sample was not completely solubil ized, and thus, only the supernatant was applied to the column. RESULTS AND DISCUSSION A. Preparation of soy protein hydrolysate Soy protein isolate was used as the starting material in this study, in order to eliminate any interference during the carbodiimide-mediated coupling reaction due to the presence of non-protein components. Various l inking modes are possible during the carbodiimide condensation reaction, i . e . , linkings between the amino group of the amino acid being bound and C-terminal a-carboxyl or side chain aspartic 6- or glutamic y - carboxyl groups of the soy protein molecule or between the N-terminal a-amino or e-amino groups of soy protein and the carboxyl group of the amino acid being bound. However, the nutrit ional ava i lab i l i ty of 3-a and y-a linkages 27. is uncertain and not well documented in the l i terature . Since i t has been reported (Franzblau e_t_ al. , 1963) that protein a-carboxyl groups react more readily and quantitatively with carbodiimides in the presence of nucleophiles to produce the corresponding protein derivatives than protein 3r or ycarboxyl groups, the SPI was part ia l ly hydrolysed with pepsin to increase the number of a-carboxyl and a-amino groups in soy protein so as to confine the amino acid binding mainly to the a-a peptide linkages. B. S tat i s t i ca l analysis Data concerning the reaction characteristics of each amino acid are of fundamental importance for providing optimum reaction conditions and thus obtaining high yields during peptide synthesis. In order to determine, therefore, the factors influencing the methionine and tryptophan binding to soy protein hydrolysate (SPH), various conditions of the carbodiimide condensation reaction were analysed by a 2-level fractional factor ia l design (Taguchi, 1957; Appendix). Table 1 shows the factors and their assigned levels investigated for their possible influence on amino acid binding to soy protein. Tables 2 and 3 show the analysis of variance of the 2-level fractional factor ia l design for methionine and tryptophan binding, respectively, to SPH. As i t can be seen in Table 2, pH, SPH concentration, reaction time, carbodiimide concentration, activation time and pH x SPH concentration interaction were found to s ignif icantly affect the methionine binding efficiency. On the other hand, Table 3 shows that pH, reaction temperature, amino acid (tryptophan) concentration, SPH concentration, carbodiimide con-centration, and pH x SPH concentration interaction signif icantly i n -fluenced the tryptophan binding to SPH. 28. Table 1 - Factors and their assigned levels investigated for their possible influence on amino acid binding to SPH. Factor Level 1. 5 8 2. Reaction temperature (°C) 22 40 3. Pepsin hydrolysis time (hr) 5 16 4. Amino acid concentration (mg) 80 140 5. SPH concentration (mg) 220 660 6. Carbodiimide concetration (mg) 80 140 7. Activation time (min) 0 60 8. Reaction time (hr) 1 24 9. Polarity (% EtOH) 0 5 Table 2 - Analysis of variance of the 2-level fractional factor ia l experiment for methionine binding to SPH. Source of Variation S.S. D.F. M.S. F-value pH (A) 1.93 1 1. 93 64. 33** SPH concentration (E) 4.48 1 4. 48 149. 33*?v Reaction time (F) 0.87 1 0. 87 29. 00** Carbodiimine concentration (G) 1.00 1 1. 00 33. 33** Activation time (H) 0.20 1 0. 20 6. 67* A x E 0.35 1 0. 35 11. 67** E r r o r 3 0.27 9 0. 03 Total 9.10 15 The sum of square values for reaction temperature (B), pepsin hydrolysis time (C), amino acid concentration (D) and polarity (I) were very low and, therefore, incorporated into the error S.S. * F 0 > 0 5 ( 1 , 9 ) = 5.12. * * F 0 . 0 1 ( 1 ' 9 ) = 1 0 - 6 0 -Table 3 - Analysis of variance of the 2-level fractional factor ia l experiment for tryptophan binding to SPH. Source of Variation S.S. D.F. M.S. F-value pH (A) 9.30 1 9.30 35.77** Reaction temperature (B) 2.10 1 2.10 8.08* Amino acid concentration (D) 1.62 1 1.62 6.23* SPH concentration (E) 7.02 1 7.02 27.00** Reaction time (F) 1.00 1 1.00 3.85 Carbodiimide concentration (G) 3.24 1 3.24 12.46* Activation time (H) 1.00 1 1.00 3.85 Polarity (I) 0.64 1 0.64 2.46 A x E 4.20 1 4.20 16.15** E r r o r 3 1.54 6 0.26 Total 31.66 15 The sum of square value for pepsin hydrolysis time (C) was low and, therefore, incorporated into the error S.S. * F 0 . 0 5 ( 1 , 6 ) = 5.99. * * F 0 . 0 1 ( 1 , 6 ) = 1 3 - 7 0 « 31. In order to find the level (of the two assigned) for each of the determined inf luential factors, which more favourably affected the binding of each amino acid to SPH, the response curve for each factor should be constructed. Figs. 2,3,4,5 and 6 show the effect of pH, SPH concentration, carbodiimide concentration, activation time, and reaction time, respectively, on the rate of methionine binding to SPH, whereas Figs. 2,3,4,7 and 8 show the effect of pH, SPH concentration, carbodiimide concentration, tryptophan concentration, and reaction temperature, respec-t ive ly , on the rate of tryptophan binding to SPH. As i t can be seen in these figures, a pH 5 was found to be more favourable than pH 8 for the binding of both amino acids (Fig. 2). Moreover, a low SPH concentration (220 mg) and high carbodiimide concentration (140 mg) were found to favor the rates of binding of both amino acids (Figs. 3 and 4, respectively). Figs. 5 and 6 show respectively that, activation time of 0 min., namely, no activation (direct condensation), and long reaction time (24 hr) were more favourable for the binding of methionine to SPH than activation time of 60 min., and a short reaction time (1 hr) . On the other hand, a high amino acid (tryptophan) concentration (140 mg) and a high reaction temperature (40°C) were found to favor the tryptophan binding to SPH (Figs. 7 and 8, respectively). In order to determine the best level for each of the selected i n -f luential factors an optimization of the carbodiimide reaction was conducted by carrying out a 3-level factor ia l analysis of variance (Taguchi, 1957; Appendix). Table 4 shows the three (3) levels assigned for each factor affecting the binding of methionine and tryptophan to SPH. Footnotes a and b indicate the level for each of the non-influential factors 4 5 6 7 8 pH Effect of pH on the rate of methionine (solid line) and tryptophan (broken line) binding to SPH. The vert ica l lines represent the 95% confidence l imits . The confidence l imit was calculated as follows: c L ± t ( 0 . 0 5 ) a t D.F. error / S . S . error < no. of data averaged D.F. error 33. 1°0 300 500 700 S P H c o n c e n t r a t i o n ( m g ) Fig. 3. Effect of SPH concentration on the rate of methionine (solid line) and tryptophan (broken line) binding to SPH. E D C c o n c e n t r a t i o n ( m g ) F i g . 4 . E f f e c t of EDC concentration on the rate of methionine ( s o l i d l i n e ) and tryptophan (broken l i n e ) binding to SPH. 35. £ 3 L A c t i v a t i o n t ime ( m i n ) F i g . 5. E f f e c t of a c t i v a t i o n time on the rate of methionine binding to SPH. 36. R e a c t i o n t i m e ( hr) Fig. 6. Effect of reaction time on the rate of methionine binding to SPH. 37. « 4 0 8 0 12 0 16 0 A m i n o a c i d c o n e e n t r a t i o n ( mg ) Fig. 7. E f f e c t of amino acid (tryptophan) concentration on the rate of tryptophan binding to SPH. 38. X 3 Q L (f) o a a o o 1 r-a H 0 20 40 R e a c t i o n t e m p e r a t u r e ( °C ) Fig. 8. Effect of reaction temperature on the rate of tryptophan binding to SPH. which was kept constant in a l l experiments. The 3 levels assigned for each influencing factor were not chosen a r b i t r a r i l y , but selected so as to approach the optimal level of the factor found by the 2-level fractional factor ia l design. Tables 5 and 6 show the analysis of variance of the 3-level factor ia l experiment for methionine and tryptophan binding, respectively. As i t can be seen in Table 5, highly significant (P<0.01) differences in methionine binding were brought about by the three assigned levels of pH, SPH concentration, carbodiimide concentration, and activation time, whereas no s ta t i s t i ca l ly significant (P>0.05) differences in methionine binding were caused by the three levels of reaction time tested or by the interaction of pH x SPH con-centration. Table 6 indicates that highly significant (P<0.01) d i f f er -ences in tryptophan binding were produced by the tested three levels of carbodiimide concentration and reaction temperature, whereas no s ta t i s t i ca l ly significant (P>0.05) differences in tryptophan binding were caused by the three levels of the other inf luent ia l factors or by the pH x SPH concentration interaction. In order to find the best level of each influencing factor among the assigned three levels the response curves of these factors should be constructed. F ig . 9 shows the effect of pH on the rate of methionine and tryptophan binding to SPH. It can be seen that, pH 6.5 was found to be the best among the three levels for the efficient binding of both amino acids. This factor (pH) has been mentioned in several papers using carbodiimides (Hoare and Koshland, 1966; Means and Feeney, 1971; Riehm and Scheraga, 1966), as having an optimum at 4.75, namely, a pH which should normally l imit the reaction to the protein carboxyl groups, 40. Table 4 - Assigned levels ' for factors influencing the amino acid binding to SPH. Amino Acid Factor Methionine Tryptophan Level Level 1 2 3 1 2 3 1. pH 3.5 5.0 6.5 3.5 5.0 6.5 2. SPH concentration(mg) 160 220 400 160 220 400 3. Carbodiimide concentration(mg)100 140 200 100 140 200 4. Amino acid concentration(mg) 100 140 200 5. Activation time(min) 0 20 40 6. Reaction time (hr) 12 24 30 7. Reaction temperature (°C) 30 40 50 Level for each of the non-influencing factors which was kept constant in a l l experiments of methionine-binding reaction to SPH: pepsin hydrolysis time (16 hr) , amino acid concentration (120 mg), polarity (0% EtOH), and reaction temperature (22°C) . Level for each of the non-influencing factors which was kept constant in a l l experiments of tryptophan-binding to SPH: pepsin hydrolysis time (16 hr) , polarity (0% EtOH), activation time (0 min), and reaction time (16 hr) . Table 5 - Analysis of variance of the 3-level factor ia l experiment for methionine binding to SPH. Source of Variation SS. D.F. MS. F-value pH(A) 19.29 2 9.65 16.16** SPH concentration(B) 21.31 2 10.66 17.86** Carbodiimide concentration(C) 9.37 2 4.69 7.86** Activation time (D) 9.38 2 4.69 7.86** Reaction activation time(E) 3.66 2 1.83 3.07 A x B 5.09 4 1.27 2.13 Error 7.16 12 0.60 Total 75.26 26 * F 0 . 0 5 ( 2 ' 1 2 ) " 3 ' 8 9 ' F 0 . 0 5 ( 4 ' 1 2 ) -3.26. ** F o . o i ( 2 ' 1 2 ) - 6 - 9 3 > Fo > 0 1<4,12) , 5.41. V Table 6 - Analysis of variance of the 3-level factor ia l experiment for tryptophan binding to SPH. Source of variation S.S. D.F. MS. F-value pH(A) 1. 97 2 0, ,99 1, ,30 SPH concentration(B) 4. 06 2 2. ,03 2. ,67 Carbodiimide concentration(C) 71. 04 2 35. ,52 46. ,74** Amino acid concentration(D) 2. 08 2 1. ,04 1. .37 Reaction temperature(E) 44. 21 2 22. ,11 29, .09** A x B 4. 25 4 1. ,06 1. ,40 Error 9. 12 12 0. ,76 Total 136. 73 26 F 0 > 0 5 ( 2 , 1 2 ) = 3.89, F 0 ( ) 5 ( 4 , 1 2 ) = 3.26. F 0 > 0 1 ( 2 , 1 2 ) = 6.93, F 0 0 1 ( 4 , 1 2 ) = 5.41. 43. 44. since at this pH most ( i f not a l l ) of the protein nucleophilic groups (such as amino or imidazole groups) would be protonated (Riehm and Scheraga, 1966). On the other hand, Horinishi e_t a l . (1968) reported that the WSC-promoted reaction between bacitracin A (a cycl ic dodeca-peptide having two carboxyl groups of aspartic and glutamic acids in the double ring structure) and glycine ethylester (nucleophile), as well as the reaction between lysozyme (egg white) and glycine ethylester, both had a broad f lat optimum between pH 5.2 and 6.9. These workers also ascribed the drop of the degree of reaction at the acidic side to de-composition of WSC, and on the alkaline side to hydrolysis of glycine ethylester with a l k a l i and to deprotonation from the tert iary amino group of WSC ( i . e . , to the drop of the reactivity of the reagents). Holbek (1976), furthermore, working on the EDC-mediated binding of lysine to wheat gluten, has noted that pHs 7.5 and 10.0 resulted in considerably greater coupling yields than pH 5.0, pH 7.5 found to be the best among these three levels tested. The finding that at pH 6.5 the amino acid coupling was higher than at pH 5.0 can be explained by taking into account the fact that the isoelectric point of soy protein fa l l s in the range of pH 4.0-5.0. Therefore, with the precipitation of soy protein at pH about 5.0, there w i l l be less access ibi l i ty for both the WSC to be bound to soy protein carboxyl groups to form the reactive intermediates (0-acylisoureas) and the amino acid to attack these and other reaction intermediates to form the desired protein derivatives (products). Moreover, i t was recently suggested by Addy et a l . (1973A) that the pH used in the carbodiimide-mediated synthesis of peptides in aqueous medium should leave the amino group of the amino acid (being bound) 45. unprotonated, although the carbodiimide and postulated O-acylisourea should be protonated. Therefore, the yie ld of peptide bond formation should rise to a maximum when this condition exists and f a l l quickly when the pH is low enough to protonate the amino group of the amino acid, thus preventing i t from acting as nucleophile (since the un-protonated form of the a-amino group of the amino acid being bound is the real nucleophile). The pK values for the amino groups of methionine and tryptophan are 9.21 and 9.39, respectively, (West and Todd, 1961), that i s , much higher than the recommended pH of 4.75. At pH 6.5, however, both conflicting requirements, i . e . low pH to l imit the reaction to protein carboxyl groups and pH which should leave the amino group of the amino acid being bound unprotonated, are partly met. Since this pH (6.5) l ies between the isoelectric point and the pK values of the amino groups of both amino acids, therefore, there w i l l be a certain amount of unprotonated amino groups, although low, available for nucleo-phi l i c attack. On the other hand, at this pH the reaction between soy protein and EDC did not involve direct reaction between EDC and amino or imidazole groups in the protein, as w i l l be demonstrated from the amino acid analysis below. Figures 10, 11 and 12 show, respectively, that low SPH concentration, high carbodiimide concentration, and high amino acid concentration (although not s ta t i s t i ca l ly significant) resulted in higher amino acid coupling yields than the other two assigned levels of these influencing factors. These findings are in accordance with the original note made by Khorana in 1955 which stated that an increase in the concentration of reactants should favor the formation of desired protein products over the formation of undesirable side-products (N-46. CO e o I a o i CD O a> o o v . o •*-> CD 100 200 300 400 S P H c o n c e n t r a t i o n ( m g ) F i g . 10. E f f e c t of SPH concentration on the rate of methionine ( s o l i d l i n e ) and tryptophan (broken l i n e ) binding to SPH. 47 80 120 160 20 E D C c o n c e n t r a t i o n ( m g ) F i g . 11. E f f e c t of EDC concentration on the rate of methionine ( s o l i d l i n e ) and tryptophan (broken l i n e ) binding to SPH. 4 8 . A m i n o a c i d c o n c e n t r a t i o n ( m g ) F i g . 12. E f f e c t of amino acid (tryptophan) concentration on the rate of tryptophan binding to SPH. acylurea derivatives). Horinishi et a l . (1968) have demonstrated that the WSC-mediated coupling between carboxyl groups of acetylglycine or proteins and aliphatic amines at room temperature can be enhanced by increasing the amine or WSC concentrations. On the other hand, i t is well known, that an excess of carboxyl component favors the formation of symmetrical anhydrides (Schroder and Lubke, 1965). Delayed addition of the nucleophiles, in part icular, has been found to decrease the importance of O-acylisourea and increase the extent of anhydride's participation during peptide synthesis (Rebek and Fe i t l er , 1973). Symmetrical anhydrides, however, have been claimed to be undesirable intermediates from the racemization point of view (Bodanszky and Ondetti, 1966). Fig . 13 shows the effect of activation time on the rate of methionine binding to SPH. As i t can be seen, activation time of 0 min., that i s , direct condensation without pre-incubation of the SPH with EDC, resulted in higher methionine coupling than activation (condensation after pre-incubation of SPH with EDC) for 20 or 40 min. This finding may be ex-plained by recal l ing the note of Previero _et'al. (1973), who reported that direct condensation without preincubation of an acidic peptide with EDC resulted in the coupling of two times as many amino acid ester moles per mole of peptide as in the case of condensation after pre-incubation of the same acidic peptide with EDC for 90 min. at 40°C. On the other hand, the extent of condensation of amino acid esters with non-acidic peptides was constant and did not change during the incubation with EDC. These workers suggested that peptides containing a carboxyl group only at the C-terminal position rapidly reach a degree of activation Q. 6 B» A c t i v a t i o n t i m e ( m i n ) F i g . 13. E f f e c t of a c t i v a t i o n time on the rate of methionine binding to SPH, 51. which remains constant, while peptides containing a side-chain carboxyl group show an i n i t i a l maximum of activation which subsequently decreases to about one half of i t s value. Taking into consideration the fact that acidic amino acids (glutamic and aspartic) represent about 33% of the total amino acid composition of soy protein, i t might be expected that at least a few of the side-chain aspartic and glutamic carboxyl groups participated in the formation of peptide bonds in the cases at which direct condensation was used. These groups, however, did not contribute to coupling when condensation after pre-incubation was used. Another factor which may also explain the higher binding yields observed when no activation took place is the formation of the undesirable stable N-acylureas. As i t has already been mentioned, N-acylureas are formed during the carbodiimide condensation reaction, when no nucleophiles are present in the reaction mixture or the reaction of those present is slow or hindered (Kopple, 1966). The formation of N-acylurea, therefore, should be expected to be much higher when condensation after pre-incubation is used, and moreover, i t should be anticipated that the longer the pre-incubation time the higher the formation of N-acylurea would be. N-acylurea, however, is not an acylating agent and thus, i ts formation would lead to lower coupling yields . Figure 14 shows the effect of reaction time on the rate of methionine binding to SPH. As i t can be seen, the three assigned reaction time levels did not cause any significant difference in methionine binding to SPH. The decrease in methionine binding, however, corresponding to the 24 hr reaction time level could not be explained. Hoare and Koshland (1966) stated that the reaction between a WSC and a protein in the presence 5 2 . X CL 6 -(/> C 3 R e a c t i o n t i m e ( h r) Fig . 14. Effect of reaction time on the rate of methionine binding to SPH. 53. of nucleophiles is quite rapid. Moreover, Sheehan et a l . (1965) working with EDC demonstrated that during the synthesis of ol igo-peptide derivatives coupling was extremely rapid (about 80% after 1 hr) . However, the reaction time required for complete coupling varied from 24 to 66 hr. Therefore, i t seems from Fig . 14 that the optimum reaction time (time required for complete coupling) varies not only with the nature of carbodiimide used (Greenstein and Winitz, 1961), but also with the carboxyl component involved (starting protein or peptide). F ig . 15 represents the effect of temperature on the rate (efficiency) of tryptophan binding to SPH. As indicated in this figure, a reaction temperature of 50°C resulted in a signif icantly higher tryptophan binding than a temperature of 40° or 30°C. This finding can be easily explained by taking into account the fact that the synthetic reaction of a peptide-bond formation from a peptide and a free amino acid is a strongly endergonic (endothermic) process (Yamashita et a l . , 1970). Therefore, when a temperature of 50°C is used during the WSC condensation reaction, there w i l l be more energy available to the system to help overcome the energy barrier with con-comitant higher y ie ld of peptide bond formation. Holbek (1976) has also noted that a reaction temperature of 40°C resulted in higher lysine coupling to gluten than a temperature of 10° or 25°C. The carbodiimide-mediated peptide synthesis is usually carried out at room temperature, because i t is easier to handle and also to reduce the formation of undesirable N-acylurea derivatives. However,since the formation of N-acylureas would be expected to result in lower tryptophan coupling yields (for N-acylureas are not acylating agents), the fact that the tryphophan binding was more efficient at 50°C than at 40° or 30°C might indicate 54. 8 r R e a c t i o n t e m p e r a t u r e ( °C ) Fig . 15. Effect of reaction temperature on the rate of tryptophan, binding to SPH. 55. that the N-acylureas formation at 50°C was perhaps not significant in this reaction. The data presented in the above mentioned response curves are summarized in Table 7, which shows the factors affecting the binding of each amino acid to SPH, and their best levels of those tested. One problem encountered during this study was the low yie ld of the f inal product, namely, of the amino acid-bound SPH. Thus, in order to improve the f ina l product's y ie ld another 3-level factor ia l analysis of the carbodiimide reaction conditions was carried out. Table 8 shows the factors and their assigned levels analysed for their possible influence on the f ina l product's y ie ld . Footnote a of Table 8 indicates the reaction conditions (factors) which were held constant in a l l experiments of this factor ia l design. Tables 9 and 10, show the analysis of variance of this 3-level factor ia l experiment for methionine and tryptophan, respectively, used for selection of the factors influencing the f ina l product's y ie ld and optimization. As shown in Table 9, pH, SPH concen-trat ion, reaction temperature and reaction time signif icantly affected the methionine-bound SPH yields . On the other hand, Table 10 shows that only SPH concentration s ignif icantly influenced the tryptophan-bound SPH yields . Table 11 indicates the factors, which were found to s ignif icantly affect the yields of methionine- or tryptophan-bound SPHs, and their best levels. The highest f ina l product's y ie ld obtained (in this factor ia l design) was 36%, whereas the lowest one was 17%. The methionine and tryptophan contents of the samples with the highest yie ld (36%) were 2.3% and 4.3% ( i . e . , 2.5 times and 4.6 times as much as those of SPI), respectively. Table 7 - Factors affecting amino acid binding and their best level Factor Amino acid Methionine Tryptophan pH 6.5 6.5 SPH concentration(mg) 160 160 Carbodiimide concentration(mg) 200 200 Amino acid concentration(mg) - 200 Activation time(min) 0 Reaction time(hr) 12 Reaction temperature(°C) - 50° Table 8 - Assigned levels for factors investigated by a 3-level factor ia l experiment for their possible influence on f ina l product's y ie ld . Factor 1 Level 2 3 SPI concentration(%) 2.4 3.6 4.8 Pepsin concentration(mg) 50 100 150 PH 5 6 7 SPH concentration(mg) 200 350 500 Reaction temperature (°C) 25 35 50 Reaction time(hr) 6 12 16 Constant factors: pepsin hydrolysis time (16 hr) , amino acid concentration (140 mg), carbodiimide concentration (140 mg), activation time (0 min). Table 9 - Analysis of variance of the 3-level factor ia l experiment for selection of the factors influencing the methionine-bound SPH yields and optimization. Source of Variation s.s. D.F. M.S. F-value SPI concentration(A) 8.06 2 4.03 0.94 Pepsin concentration(B) 2.32 2 1.16 0.27 PH (C) 268.90 2 134.45 31.49** SPH concentration(D) 192.93 2 96.47 22.59** Reaction temperature(E) 207.51 2 103.76 24.30** Reaction time(F) 60.02 2 30.01 7.03* A x B 97.85 4 24.46 5.73 A x C 27.45 4 6.86 1.60 Error 25.61 6 4.27 Total 890.65 26 F 0 > 0 5 ( 2 , 6 ) = 5.14, F 0 > 0 5 ( 4 , 6 ) = 4.53. F 0 > 0 1 ( 2 , 6 ) = 10.43, F 0 > 0 1 ( 4 , 6 ) = 9.15. Table 10 - Analysis of variance of the 3-level factor ia l experiment for selection of the factors influencing the tryptophan-bound SPH yields and optimization. Source of Variation S.S. D.F. MS. F-value SPI concentration(A) 4.25 2 2.13 0.33 Pepsin concentration(B) 17.89 2 8.95 1.37 pH (C) 29.05 2 14.53 2.23 SPH concentration(D) 71.92 2 35.96 5.52* Reaction temperature(E) 59.72 2 29.86 4.58 Reaction time(F) 18.46 2 9.23 1.42 A x B 169.67 4 42.42 6.51 A x C 7.75 4 1.94 0.30 Error 39.11 6 6.52 Total 417.82 26 * F 0 i 0 5 ( 2 , 6 ) = 5.14. ** F o . o i ( 2 ' 6 ) = 1 0 , 4 3 * Table 11 - Factors affecting f ina l product's y ie ld and their best levels . Factor Amino Acid Methionine Tryptophan pH 5 -SPH concentration(mg) 500 500 Reaction time(hr) 16 Reaction temperature(°C) 50° -61. Since the y ie ld of the f ina l product could not be markedly improved by the above 3-level factor ia l design, in which peptic SPH was used as the starting material in the carbodiimide reaction, another attempt to further improve the yie ld was done by using SPI i t s e l f ( i . e . , without preliminary pepsin hydrolysis) as the starting material and the same reaction conditions under which the amino acid-bound SPH samples with the highest coupling yields were obtained. The yield so obtained was 95% and 99% (on a weight basis) for the methionine- and tryptophan-bound SPI samples, respectively. Moreover, the efficiency of methionine and tryptophan binding to SPI was s t i l l excellent, as w i l l be demonstrated in the amino acid analysis section below. Amino acid analysis The amino acid composition of SPI, and methionine- and tryptophan-bound soy protein samples with the highest amino acid binding efficiency is reported in Table 12. These results show that the L-methionine content in the methionine-bound SPH sample was 7.22g/16gN, whereas i ts content in the SPI (control) was only 0.94g/16gN. Furthermore, i t can be seen that the L-tryptophan content in the tryptophan-bound SPH sample was 17.05g/16gN, whereas i t s content in the SPI was only 0.95g/16gN. On the other hand, Table 12 shows that the L-methionine content in themethionine-bound SPI sample was 5.92g/16gN, and the L-tryptophan content in the tryptophan-bound SPI sample was 10.74g/16gN. The above results , therefore, very clearly demonstrate that the L-methionine and L-tryptophan added have been very eff ic ient ly bound to SPHs and SPIs. The fact that the L-tryptophan has been more eff ic ient ly bound in both cases ( i . e . , when SPH or SPI was used as the starting material) than the L-methionine may be attributed to the higher reactivity of L-tryptophan (due to the presence of the indole ring in i t s molecul 62. Table 12 - Amino acid composition (g/16gN) of SPI and S amino acid-bound soy protein samples with the highest coupling yields _b Amino Control Met-bound Trp-bound Met-bound Trp-bound Acid SPI SPH SPH SPI SPI Aspartic acid 11.41 10.57 8.82 10.84 10.32 Threonine 3.67 3.66 3.09 3.53 3.39 Serine 5.31 5.75 4.25 5.00 4.73 Glutamic acid 19.40 18.21 15.37 18.34 17.50 Proline 5.10 5.54 5.52 4.84 4.68 Glycine 3.76 3.72 3.26 3.52 3.32 Alanine 3.75 3.16 2.83 3.56 3.46 Valine 3.67 3.54 3.09 3.23 2.96 Methionine 0.94 7.22 0.94 5.92 0.83 Isoleucine 4.27 3.61 3.68 3.97 3.69 Leucine 7.20 6.20 5.22 7.13 6.61 Tyrosine 3.58 3.34 3.07 3.30 3.15 Phenylalanine 5.30 4.50 4.37 5.00 4.72 Lysine 5.83 4.45 5.17 5.71 5.25 Histidine 2.53 2.77 2.63 2.43 2.25 Tryptophan 0.95 0.94 17.05 0.90 10.74 Arginine 7.38 6.82 5.64 6.93 6.56 Cystine 0.92 1.00 0.99 0.87 0.84 Each value is the average of duplicate analyses. Experimental conditions: (a) methionine binding: pH 6.5; SPH or SPI concentration, 160 mg; EDC concentration, 140 mg; activation time, 0 min; reaction time, 12 hr; pepsin hydrolysis time, 16 hr; methionine concentration, 120 mg; reaction temperature, 22°C; (b) tryptophan binding: pH 6.5; SPH or SPI concentration, 160 mg; EDC concentration, 200 mg; activation time, 0 min; tryptophan concentration, 140 mg; reaction temperature, 50°C; pepsin hydrolysis time, 16 hr; reaction time, 16 hr. since i t is known that the coupling reaction, in general, depends upon the reactivity of each amino acid (Tometsko, 1973). A select ivi ty in aqueous peptide synthesis mediated by EDC was also reported by Addy et a l . (1973B), which was suggested to be due not only to chemical react i -vi ty of the side-chains of the amino acids but also to size and geometry of their side-chains. However, the higher efficiency of tryptophan binding observed here is apparently due to the former reason ( i . e . , higher react iv i ty) . It is noteworthy that as indicated in Table 12, the methionine-and tryptophan-bound SPI samples possessed amino acid compositions (except for the bound amino acids) which were similar (or closely resembled) to that found for the SPI (control). Therefore, i t may be concluded that the reaction between SPI and EDC in the presence of added amino acid (methionine or tryptophan), at pH 6.5, did not involve direct reaction between the EDC and amino or imidazole groups in the protein, since the reaction between the EDC and a nucleophilic group (amino or imidazole group) in the protein should yield amino acid derivatives which would be stable to acid hydrolysis (Riehm and Scheraga, 1966). Although soybean protein contains about 33% acidic amino acids (aspartic and glutamic), the possible lesser involvement of side-chain carboxyl groups in the EDC-mediated reaction could be due to amidation of aspartic and glutamic acid residues in the soy protein in the forms of asparagine and glutamine residues. The amide groups of these residues, however, have been shown to be very resistant to the action of carbo-diimides (Liberek, 1962; Bodanszky'et a l . , 1976), since the formation of the obligatory active intermediate (O-acylisourea) with carbodiimides 64. requires that the carboxyl groups be free (Franzblau et^ a l . , 1963). Therefore, these groups would not participate in the EDC condensation reaction. Another factor which could also explain the lesser p a r t i -cipation of side-chain carboxyl groups to EDC reaction is the finding of Franzblau et a l . (1963), who reported that the unstable reactive intermediate O-acylisourea - in addition to generally accepted possi-b i l i t i e s of formation of a stable N-acylurea, a symmetrical anhydride or a peptide derivative - could cycl ize . Thus, in the case of an aspartic acid residue (in an a-peptide linkage) cycl ization would yie ld a succinimide r ing , and a glutamic acid residue (in a-peptide linkage) could form either a glutarimide or acylpyrrolidone r ing . If present, succinimide structures, on subsequent acid hydrolysis, would give aspartic acid and glutarimide or pyrrolidone rings would y ie ld glutamic acid. Therefore, the cyclized residues would again appear as unreacted amino acids. A third factor which could also partly explain the lesser involvement of side-chain carboxyl groups is the formation of undesirable N-acylurea as a by-product of EDC reaction, which (since N-acylurea is not an acylating agent) results in lower coupling yields . N-acylureas, however, have been suggested to be ac id- labi le (Riehm and Scheraga, 1966), thus, l iberating the amino acid involved on acid hydrolysis. Undoubtedly a certain amount of N-acylurea can accompany the formation of desired peptide derivatives in the EDC reaction. D. Enzymatic digestion test It is well known that the nutrit ive value of a protein depends not only on the pattern of i t s amino acid composition, but also on the physiological ava i lab i l i ty of i ts amino acids. Of the several characteristics which 65. determine the nutrit ional quality of food proteins, their d iges t ib i l i ty is of paramount importance. Although in the f ina l analysis the nutrit ive quality of a protein must be evaluated by feeding t r i a l s , in v i tro enzy-matic methods for protein quality evaluation have been ut i l i zed for preliminary quality evaluations (Stahmann and Woldegiorgis, 1975). Even though, the enzymatic hydrolysis may not be complete, the amount of the l imiting amino acid released by enzymatic digestion is a good indicator of the amino acid ava i lab i l i ty and hence protein quality. Table 13 reports the amount of each amino acid liberated by an In vi tro pepsin followed by pancreatin digestion from the SPI (control) and the amino acid-bound soy protein samples with the highest binding yields (efficiencies) . As i t can be seen, 2.18g/16gN of L-methionine were released from the methionine-bound SPH sample as compared to only 0.25g/16gN of L-methionine released from the SPI, whereas 4.55g/16gN of L-tryptophan were released from the tryptophan-bound SPH sample as compared to only 0.47g/16gN of L-tryptophan released from the SPI. On the other hand, Table 13 shows that 1.56g/16gN of L-methionine was released from the methionine-bound SPI sample and 2.06g/16gN of L-tryptophan was released from the tryptophan-bound SPI sample. These results , therefore, clearly indicate that the bound methionine and tryptophan can be eff ic ient ly released from both the SPH and SPI treated samples by enzymatic digestion in v i t r o . 66. Table 13 - Liberated amino acids (g/16gN) from SPI and amino acid-bound soy protein samples with the highest coupling yields^ by pepsin-pancreatin digestion jln v i t r o . Amino Control Met-bound Trp-bound Met-bound Trp-bound Acid SPI SPH SPH SPI SPI Threonine 0. .05 0. 20 0. ,20 0, .18 0. 19 Valine 1. .46 0. 59 0. ,50 0. .51 0. 44 Methionine 0. ,25 2. 18 0. 09 1. .56 0. 13 Isoleucine 0. .95 0. 73 0. ,63 0, .75 0. 56 Leucine 2. .74 1. 58 1. ,34 2. .62 2. 19 Tyrosine 1. .98 1. 29 1. ,09 1. .75 1. 16 Phenylalanine 2. .65 1. 58 1. ,40 2. .76 1. 68 Lysine 1. .51 1. 24 1. ,09 1. .38 0. 97 Histidine 0. .36 0. 57 0. ,53 0, .53 0. 49 Tryptophan 0. .47 0. 54 4. ,55 0, .49 2. 06 Arginine 2, .17 3. 36 2. ,28 3, .16 2. 64 Each value is the average of duplicate analyses. Nutrit ionally important amino acids only were shown. For experimental conditions see footnote b on Table 12. Gel f i l t r a t i o n chromatography Gel f i l t r a t i o n chromatography was used to study the effect, i f any, of the carbodiimide condensation reaction on the molecular weight distribution pattern of soy protein. In order to determine the molecular weights of the different protein fractions the Sephadex G-200 column was calibrated with the standard proteins l i s ted in Table 14. A linear standard curve was obtained by plotting the logarithm of the molecular weights of the standard proteins against the ratios of their elution volumes to the void volume of the column (Whitaker, 1963). Figure 16 shows the cal ibration curve of this column. Figure 17 represents the elution profi le of SPH solubilized with SDS. As i t can be seen, three protein fractions were separated, the molecular weights of which were estimated (from the calibration curve) to be 158,490, 70,000, and 4,924, respectively (in order of increasing elution volume). Figure 18 shows the elution pattern of tryptophan-bound SPH sample with the highest binding efficiency solubilized with SDS. This pattern was obtained af ter di lut ing (1:5 dilution) each collected fraction (5.2 ml) with eluting buffer. This di lut ion was necessary in order to obtain absor-bance values being within the spectrophotometer's absorbance range (0.0-2.0). The f irst-e luted fraction of the Figure 18 appeared at the void volume of the column and thus i t s molecular weight could not be estimated. It is assumed to be very high, since the peak of catalase curve, whose molecular weight is 244,000, was collected inthe fraction (tube) 35, whereas the peak of this protein fraction appeared at the tube 29, namely, was eluted 31.2 ml (6 tubes) before the catalase. The molecular weights of the two other protein fractions of Figure 18 were estimated to be 97,700 and 15,550. F ig . 19 represents a comparison of the elution profiles of SPH (Fig. 17, used as control) and tryptophan-bound SPH (Fig. 18). It can be clearly seen that the elution pattern from the peptic SPH shifted towards the high-molecular weight side as a Table 14 - Proteins used in cal ibration of Sephadex G-200 column eluted with 0.1M sodium phosphate buffer, pH 7.0, and containing 0.1% SDS . Proteins V /V Mol.Wt. (Ref.) e o 1. Catalase 1.10 244,000 (102) 2. Bovine serum albumin 1.52 70,000 (112) (Monomer) 3. a-Chymotrypsin 2.03 22,500 (112) 4. Sperm whale myoglobin 2.14 17,500 (*) 5. Ribonuclease A 2.17 13,600 (112) 6. L-tryptophan 2.38 204 (*) * Mol. Wt. provided by supplier. 69. Log. Molecular Weight Fig. 16. Calibration curve of Sephadex G-200. Fraction number Fig. 17. Gel f i l t r a t i o n pattern of SPH solubilized with SDS on Sephadex G-200. 71. Fraction number Fig. 18. Gel f i l t r a t i o n pattern of tryptophan-bound SPH solubilized with SDS on Sephadex G-200. The elution pattern was measured after di luting (1:5) each fraction with the eluant . 72. Fraction number Fig. 19.. Gel f i l t r a t i o n patterns of SPH (solid line) and tryptophan-bound SPH (broken line) on Sephadex G-200. 73. result of the EDC condensation reaction. Figure 20 shows the gel f i l t r a t i o n pattern of SPH solublized with NaOH (pH 12.0). Three fractions were, again, obtained, the molecular weights of which estimated from the calibration curve were 158,490, 70,000 and 6,300, respectively (in order of increasing elution volume). Figure 21 represents the elution pattern of the methionine-bound SPH sample with the highest binding efficiency, solubilized with NaOH (pH 12.0). As i t can be seen four fractions were separated. The f irst -e luted fraction appeared at the void volume of the column, and thus, the estimation of i t s molecular weight was not possible. The other fractions were estimated to have molecular weights of 91,200, 28,900, and 13,600, respectively (in order of increasing elution volume). Figure 22 shows a comparison of the elution profi les of SPH (Fig. 20, used as control) and methionine-bound SPH sample (Fig. 21). It can be seen, again, that the EDC conden-sation reaction caused an increase in the molecular weights of soy protein fractions. This finding is in accordance with the results of several papers dealing with the use of carbodiimides in peptide synthesis. Sheehan and Hlavka (1957), have reported that the reaction of a commercial gelatin with a WSC (in the absence of nucleophiles) resulted in very rapid gelation attributed to interchain cross-linking by amide bond formation occuring between side-chain carboxyl and amino groups. Blout and Des Roches (1959), then, reported that through the use of carbodiimides as condensing agents the molecular weights of polypeptides could be increased s ignif icantly , and moreover, that in several instances they observed four-fold or greater increases in molecular weights. These workers also suggested that this increase in molecular weight occurred because of terminal inter-1.0 . E c O CO 0 . 8 L CM I 3 0 4 0 5 0 6 0 7 0 8 0 Fraction n u m b e r Fig..20. Gel f i l t r a t i o n pattern of SPH s o l u b i l i z e d with NaOH (at pH 12.0) on Sephadex G-200. 75. Fi g . 21. Gel f i l t r a t i o n pattern of methionine-bound SPH s o l u b i l i z e d with NaOH (at pH 12.0) on Sephadex G-200. 3 0 4 0 5 0 6 0 7 0 8 0 Fraction number F i g . 22. Gel f i l t r a t i o n patterns of SPH ( s o l i d l i n e ) and methionine-bound SPH (broken l i n e ) on Sephadex G-200.. 77. molecular peptide bond formation, but definite proof was d i f f i c u l t to obtain. However, indirect evidence that linear peptide bond formation was involved could be deduced from their observation that polypeptides with one terminal amide group ( i . e . , the carboxyl end of the polypeptide chain was blocked by amide formation) and one terminal amine group did not undergo molecular weight increases when treated with carbodiimides, and also from the results obtained with poly-L-proline. With this poly-peptide no inter-molecular condensations other than terminal may occur and poly-L-proline did show a definite molecular weight increase when treated with carbodiimides. More recently, Mukerjee and S r i Ram (1965), by carrying out the reaction between bovine serum albumin (BSA) and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide in the absence of nucleo-philes, and based on subsequent sedimentation studies, reported that some aggregation of the protein had occurred. These workers also reported that, on reaction of EDC with BSA in the presence of diglycine, part of the peptide apparently was conjugated to the protein. In order to determine i f there was any select ivity regarding the binding of added amino acid to different soybean protein fractions during the EDC reaction, the elution profiles of tryptophan-bound SPH sample, solubil ized with NaOH (pH 12.0), at 280 nm3 (Fig. 23) and 210 nm'3 (Fig. 24) were obtained, and consequently, the ratio of the three separated fractions (areas under the eluted peaks) for each elution profi le was determined. The area ratio (%) of the three fractions of Absorption maximum of proteins due to the presence of aromatic amino acids (tryptophan, tyrosine, and phenylalanine). k Absorption of the peptide bond (Pomeranz and Meloan, 1971). Fraction number Fig . 23. Gel f i l t r a t i o n pattern (A280nm) ° f tryptophan-bound SPH solubilized with NaOH (at pH 12.0) on Sephadex G-200. The elution pattern was measured after di lut ing (1:5) each fraction with the eluant. 79. 20 30 40 50 60 70 80 Fraction number Fig. 24. Gel f i l t r a t i o n pattern (A2in n m) of tryptophan-bound SPH solubilized with NaOH (at pH 12.0) on Sephadex G-200. The elution pattern was obtained after di luting (1:20) each fraction with 0.1 M sodium phosphate buffer, pH 7.0. 80. F ig . 23 was (44.79):(36.46):(18.75), in order of descending molecular weight, as compared to the ratio of (45.91):(38.26):(15:83) of the corresponding fractions of F ig . 24. Since these two ratios are very close (the small differences between the corresponding areas % of Figs. 23 and 24 are negligible and can be attributed to the di lut ion effects), i t is concluded that there was no selective amino acid binding among the different soy protein fractions during the EDC reaction. GENERAL DISCUSSION From the results of Table 12, i t appeared that the L-methionine and L-tryptophan added could be very eff ic ient ly bound to both SPH and SPI by the EDC condensation reaction. Moreover, Table 13 indicated a high d iges t ib i l i ty in both cases ( i . e . , amino acid bound SPH and SPI). The use of SPI (instead of SPH) as the starting material in this reaction would be more advantageous for the following reasons: (1) elimination of the pepsin hydrolysis step required for the production of SPH. Thus, the covalent binding procedure would be simplified to one-step process which would permit direct binding of amino acids to soy protein; (2) prevention of the loss of some amino acids which is associated with the pepsin hydrolysis step (see Table 12); and (3) very high product yields (95-99%) as compared to the low yields (maximum yield 36%) obtained when SPH is used as the starting material. The results of Table 12 showed that the L-methionine content in the methionine-bound SPI sample was 5.92g/16gN; this value is 6.3 times the methionine leve l , 0.94g/16gN, of the unmodified SPI. On the other hand, the L-tryptophan content in the tryptophan-bound SPI sample (Table 12) was 10.74g/16gN, namely 11.3 times the tryptophan content, 0.95g/16gN, of the 81. SPI (control). Although such contents (5.92g/16gN methionine and 10.74g/16gN tryptophan) are considered to be a surplus for optimum amino acid nutr i t ion, i t is possible, however, to obtain a protein product with a satisfactory methionine or tryptophan content i f the factors affecting the binding of each amino acid to soy protein (e.g. , amino acid concentration, carbodiimide concentration, pH, etc.) are properly controlled upon the application of EDC reaction. A more pract ical way to decrease the content of these amino acids is to mix the products with SPI. A 12:20:1 mixture, for example, of methionine-bound SPI sample: SPI:tryptophan-bound SPI sample w i l l contain 2.7g/16gN methionine, 0.9g/16gN cystine, and l.lg/16gN tryptophan. Such a mixture w i l l resemble the ideal protein proposed by the FAO/WHO Expert Committee (1973), in respect to the levels of sulfur-containing amino acids (3.5g/ 16gN) and tryptophan (1.0g/16gN). It is well known that, i f not properly considered, the addition of an excess of the f i r s t l imit ing amino acid may result in a secondary deficiency and, thus, there w i l l be no improve-ment in the overall quality of the diet (Bressani, 1975). Consequently, the f ina l mixture should include both the f i r s t and second l imiting amino acids in the amounts needed to maintain the proper ratio according to requirements of the reference pattern (FAO/WHO Expert Committee - suggested pattern). A major obstacle in the application of chemical modification to food proteins is the potential health hazards. Although enzymatic procedures (Whitaker, 1977) offer advantages with regard to potential health hazards, chemical procedures are frequently more advantageous for different products due to a low cost (Feeney, 1977). The major advantage of EDC used in this study is the fac i le removal of excess reagent and the corresponding urea 82. by washing with d i l u t e acid or water. Moreover, i t has been reported (Aumuller, 1963) that the mammalian t o x i c i t y of carbodiimides i s low ( L D ^ Q of dicyclohexylcarbodiimide was reported to be 2.6 g/kg i n r a t s ) . However, the p o s s i b i l i t y that some E D C and urea derivatives were s t i l l remained i n the f i n a l product, after i t s p u r i f i c a t i o n , could not be excluded. I t i s obvious, therefore, that a thorough b i o l o g i c a l evaluation i s needed to assess the effect of chemical modification of soy proteins by E D C on their i n vivo d i g e s t i b i l i t y as well as any effect that modification might have on the t o x i c i t y of their derivatives (methionine- or tryptophan-bound soy proteins). CONCLUSION According to the data presented i n t h i s study, the carbodiimide condensation reaction appeared to be a very e f f e c t i v e means for covalently binding methionine and tryptophan to soy protein and thus improving i t s n u t r i t i o n a l q u a l i t y . It i s f e a s i b l e to obtain a product with a s p e c i f i e d methionine or tryptophan content by simply c o n t r o l l i n g the factors a f f e c t i n g the binding of each of these amino acids to soy protein during the carbodiimide reaction. It was found that the methionine and tryptophan added could be very e f f i c i e n t l y bound to both soy protein hydrolysate and soyprotein i s o l a t e by the carbodiimide reaction. Moreover, a pepsin-pancreatin digestion test demonstrated a high d i g e s t i b i l i t y i n both cases. Soy protein i s o l a t e , however, was found to be more advantageous as s t a r t i n g material i n the carbodiimide (EDC) reaction than soy protein hydrolysate, because of the s i m p l i f i c a t i o n of the covalent binding procedure to only one-step process, prevention of the loss of some amino acids associated with the pepsin hydrolysis step, and above a l l , because of the very high product y i e l d s obtained i n t h i s case. The n u t r i t i o n a l q u a l i t y of the product obtained, as determined by amino acid composition, was found to be superior to that of soyprotein i s o l a t e ( c o n t r o l ) . Gel f i l t r a t i o n chromatography elucidated the formation of higher-molecular weight substances from peptic hydrolysates of soy protein as a r e s u l t of the EDC reaction. Gel f i l t r a t i o n chromatography, moreover, revealed that there was not any s e l e c t i v i t y concerning the binding of added amino acid to the d i f f e r e n t soyprotein f r a c t i o n s during the EDC reaction. The carbodiimide reaction may be of special importance from the viewpoint of food processing i n practice. Further research i s required to determine the commercial and economic f e a s i b i l i t y of this method as a means for improving the n u t r i t i o n a l and functional properties of food proteins. Moreover, i t should be emphasized that the safety and acceptability of the product as a human food must be carefully evaluated. LITERATURE CITED 85. 1. Addy, M. E . , Steinman, G . , and Mallette, M. F. 1973A. A problem in the mechanism of carbodiimide-mediated synthesis of peptides in aqueous medium. Biochem. Biophys. Res. Com. 52:1034. 2. Addy, M. E . , Steinman, G . , and Mallette, M. F. 1973B. Selectivity in aqueous peptide synthesis by l-Ethyl-3-(3-dimethylaminopropyl) carbodiimide. Biochim. Biophys. Acta 295:385. 3. Adibi , S. A. 1971. Intestinal transport of dipeptides in man: relative importance of hydrolysis and intact absorption. J . C l i n . Invest. 50:2266. 4. Akeson, W. R. , and Stahmann, M. A. 1964. A pepsin pancreatin digest index of protein quality evaluation. J . Nutr. 83:257. 5. Altschul , A. M. 1974. Fort i f icat ion of foods with amino acids. Nature 248:643. 6. Anderson, G. W., and Callahan, F. M. 1958. Racemization by the dicyclohexylcarbodiimide method of peptide synthesis. J . Amer. Chem. Soc. 80:2902. 7. A r a i , S. 1974. "Proceedings of the Symposium on New Techniques for Use of Enzymes in Food. Technology". Japanese Assoc. Food Technol., Tokyo, Japan. Cited by Fujimaki et a l . , 1977. 8. Aumuller, W. 1963. Angew. Chem. 75:873. Cited by Kurzer, F . , and Douraghi-Zadeh, K . , 1967. Advances in the chemistry of carbodiimides. Chem. Rev. 67:107. 9. Ballance, P. E . 1961. Production of vo lat i l e compounds related to the flavor of foods from the Strecker degradation of DL-methionine. J . Sc i . Food Agric . 12:532. 10. Beigler, M. A. 1969. Pract ical problems of amino acid for t i f i ca t ion . In "Protein-Enriched Cereal Foods for World Needs". Ed. Milner, M . , p.200. American Association of Cereal Chemists, St. Paul, Minn. 11. Benoiton, L . , and Rydon, H. N. 1960. Polypeptides. Part V. The Synthesis of Some Acetylserine Peptides Related to the Active Centre of Chymotrypsin. J . Chem. Soc. p. 3328. 12. Berry, T. H . , Becker, D. E . , Rasmussen, 0. G . , Jensen, A. H . , and Norton, H. W. 1962. The l imit ing amino acids in soybean protein. J . Animal Sc i . 21:558. 86. 13. Berry, T. H . , Combs, G. E . , Wallace, H. D . , a n d Robbins, R. C. 1966. Responses of the growing pig to alternations in the amino acid pattern of isolated soybean protein. J . Animal Sc i . 25:722. 14. Blout, E. R. , and Des Roches, M. E. 1959. The Preparation of High Molecular Weight Polypeptides. J . Amer. Chem. Soc. 81:370. 15. Bodanszky, M . , Klausner, Y. S., and Ondetti, M. A. 1976. Peptide Synthesis. Wiley-Interscience Publ. N. York. 16. Bodanszky, M . , and Ondetti, M. A. 1966. Peptide Synthesis. Interscience Publ. N. York. 17. Bookwalter, G. N . , Kwolek, W. F . , Black, L . T . , and G r i f f i n , E . L . J r . 1971 Corn meal/soy flour blends: Characteristics and food applications. J . Food Sc i . 36:1025. 18. Bressani, R. 1974. Complementary amino acid patterns. In "Nutrients in processed foods". Proteins. Eds. White, P. L . , and Fletcher, D. C . , p.149. American Medical Association, Mass. 19. Bressani, R. 1975. Addition of Amino Acids to Foods. In "Nutrional Evaluation of .Food Processing, Eds. Harris , R. S., and Karmas, E . , p.568. Avi Publishing Co. , Inc. , Westport, Conn. 20. Burston, D . , Addison, J . M . , and Matthews, D. M. 1972. Uptake of dipeptides containing basic and acidic amino acids by rat small intestine in Vi tro . C l i n . Sc i . 43:823. 21. Catsimpoolas, N. 1969. Isolation of glycinin subunits by isoelectric focusing in urea-mercaptoethanol. FEBS Lett . 4:259. 22. Catsimpoolas, N . , Leuthner, E . , and Meyer, E . W. 1968. Studies on t h e c h a r a c t e r i z a t i o n of .soybean proteins by I m m u n o e l e c t r o -phoresis. Arch. Biochem. Biophys. 127:338. 23. Chem. Engng News 1973. P. 18, Dec. Cited by Altschul , A. M. 1974. 24. Cheng, B . , Navab, F . , L i s , M. T . , M i l l e r , T. N . , and Matthews, D. M. 1971 Mechanisms of dipeptide uptake by rat small intestine in Vi tro . C l i n . Sc i . 40:247. 25. Concon, J . M . , and Soltess, D. 1973. Rapid Micro Kjeldahl Digestion of Cereals Grains and other Biological Materials. Anal. Biochem. 53:35. 87. 26. Coppock, J . 1974. Soy Proteins in Foods - Retrospect and Prospect. J . Amer. O i l Chem. Soc. 51:59A. 27. Craft , I. L., Geddes, D . , Hyde, C. W., Wise, I. J . , and Matthews, D. M. 1968. Absorption and malabsorption of glycine and glycine peptides in man. Gut 9:425. Cited by Burston et a l . , 1972. 28. DeTar, D. F . , and Si lverstein, R. 1966A. Reactions of Carbodiimides, I. The Mechanisms of the Reactions of Acetic Acid with Dicyclo-hexylcarbodiimide. J . Amer. Chem. Soc. 88:1013. 29. DeTar, D. F . , and Si lverstein, R. 1966B. Reactions of Carbodiimides. II. The Reactions of Dicyclohexylcarbodiimides with Carboxylic Acids in the presence of Amines and Phenols. J . Amer. Chem. Soc. 88:1020. 30. DeTar, D. F . , Si lverstein, R. , and Rogers, F. F. J r . 1966C. Reactions of Carbodiimides. III . The Reactions of Carbodiimides with Peptide Acids. J . Amer. Chem. Soc. 88:1024. 31. Diehl, J . F. 1957. Dissertation, Ruprecht-Karl Universitat Heidelberg. Cited by Schroder and Lubke, 1965. 32. Doleschall, G . , and Lempert, K. 1963. On the mechanism of carboxyl condensations by carbodiimides. Tetrahedron Letters 18:1195. 33. Du Vigneaud, V . , Bartlett , M. F . , and Johl , A.1957. The Synthesis of Lysine-vasopressin. J . Amer. Chem. Soc. 79:5572. 34. Evans, R. J . , and McGinnis, J . 1946. The influence of autoclaving soybean oilmeal on the ava i lab i l i ty of cystine and methionine for the chick. J . Nutr. 31:449. 35. FAO/WHO Ad Hoc Expert Committee. 1973. Energy and Protein Require-ments. WHO technical report series No. 522, p.63, Geneva. 36. Feeney, R. E. 1977. Chemical Modification of Food Proteins. In "Food Proteins. Improvement through Chemical and Enzymatic Modification. Adv. Chem. Ser. 160, Amer. Chem. S o c , Wash., D.C. 37. Fomon, S. J . , Thomas, L . N . , F i l e r , L . J . J r . , Anderson, T. A . , and Bergman, K. E. 1973. Requirements for protein and essential amino acids in early infancy. Studies with a Soy-Isolate formula. Acta Paediatr. Scand. 62:33. 88. 38. Franzblau, C . , Gallop, P. M . , and Seifter, S. 1963. The presence in Collagen of y-glutamyl peptide linkages. Biopolymers 1:79. 39. Fujimaki, M . , A r a i , S., and Yamashita, M. 1977. Enzymatic Protein Degradation and Resynthesis for Protein Improvement. In "Food Proteins. Improvement through Chemical and Enzymatic Modification". Eds. Feeney, R. E . , and Whitaker, J . R. Adv. Chem. Ser. 160, American Chem. S o c , Wash., D.C. 40. Greenstein, J . P . , and Winitz, M. 1961. Chemistry of the Amino Acids, Vol . 2:1016. Wiley J . and Sons, Inc. , N.Y. 41. Hasegawa, K . , Kusano, T . , and Mitsuda, H. 1963. Fractionation of Soybean Proteins by Gel F i l t r a t i o n . Agr. B i o l . Chem. (Tokyo) 27:878. 42. Helferich, B . ,„and Boshagen, H. 1959 Chem. Ber. 92:2813. Cited by Schroder and Lubke, 1965. 43. Hippe, R. L . , and Warthesen, J . J . 1978. Effect of Methionine and N-Acetylmethionine Fort i f icat ion on the Flavor of Soy Bread and Soy Milk. J . Food Sc i . 43:793. 44. Hoare, D. G . , and Koshland, D. E. J r . 1966. A procedure for the Selective Modification of Carboxyl Groups in Proteins. J . Amer. Chem. Soc. 88:2057. 45. Hofmann, K . , Woolner, M. E . , Spiihler, G . , and Schwartz, E . T. 1958. Studies on polypeptides. X. The Synthesis of a pentapeptide corresponding to an amino acid ^sequence present in corticotropin and in the melanocyte stimulating hormones. J . Amer. Chem. Soc. 80:1486. 46. Holbek, E. K. 1976. The Chemical Binding of Lysine to Wheat Gluten (B.Sc. Thesis). Dept. of Food Science, U.B.C. 47. Horan, F. E. 1973. Wheat-Soy blends - high quality protein products. Cereal. Sci . Today 18:11. 48. Horan, F. E. 1974A. Soybeans - Their contribution to the welfare of mankind. Food Techn. in Australia 26(11):492. 49. Horan, F. E . 1974B. Meat Analogs. In "New Protein Foods". Ed.' Altschul , A . M . , P.366. Academic Press, N. York. 89. 50. Horinishi , H. , Nakaya, K . , and Tani, A. 1968. States of Amino Acid Residues in Proteins. XV. Ethyl-morpholinylpropyl-carbodiimide for Modification of Carboxyl Groups in Proteins. The J . of Biochem. (Japan) 63:41. 51. Howe, E. E . , G i l f i l l a n , E. W., and Milner, M. 1965A. Amino acid supplementation of protein concentrates as related to the world protein supply. Amer. J . C l i n . Nutr. 16:321. 52. Howe, E. E . , Jansen, G. R. , and Anson, M. L . 1967. An approach toward the solution of the world food problem with special emphasis on protein supply. Amer. J . C l i n . Nutr. 20:1134. 53. Howe, E. E . , Jansen, G. R. , and G i l f i l l a n , E . W. 1965B. Amino acid supplementation of cereal grains as related to the world food supply. Amer. J . C l i n . Nutr. 16:315. 54. Kato, J . and Muramatsu, N. 1971. Amino acid supplementation of grain. J . Amer. O i l Chem. Soc. 48:415. 55. Katsoyannis, P. G . , and Suzuki, K. 1962. Insulin peptides. III . Synthesis of a Protected Nonapeptide Containing the C-Terminal Sequence of the B-Chain of Insulin. J . Amer. Chem. Soc. 84:1420. 56. Khorana, H. G. 1952. Peptides. III . Selective degradation from the carboxyl end. The use of carbodiimides. J . Chem. Soc. 2081. 57. Khorana, H. G. 1953. The Chemistry of carbodiimides. Chem. Rev. 53:145. 58. Khorana, H. G. 1955. Observations on the use of dicyclohexyl-carbodiimide in the synthesis of peptides. Chem. and Ind. 1087. 59. Klaui , H. 1974. The technological aspects of the addition of nutrients to foods. In "Proc. IV Inter. Congress Food Sc i . and Technology. Vol . I , p. 740, Madrid, Spain. 60. Klausner, Y. S., and Bodansky, M. 1972. Coupling reagents in peptide synthesis. Synthesis 9:453. 61. Kopple, K. D. 1966. Peptides andAmino Acids. Benjamin, W. A . , Inc. , N.Y. 62. Koshiyama, I. 1968A. Factors influencing conformation changes in a 7S protein of soybean globulins by ultracentrifugal investigations. Agr. B i o l . Chem. (Tokyo) 32:879. 90. 63. Koshiyama, I. 1968B. Chemical and physical properties of a 7S protein in soybean globulins. Cereal chem. 45:394. 64. Koshiyama, I. 1969. Isolation of a glycopeptide from a 7S protein in soybean globulins. Arch. Biochem. Biophys. 130:370. 65. Koshiyama, I. 1971. Some aspects of sub-unit structure of a 7S protein in soybean globulins. Agr. B i o l . Chem. (Tokyo) 35:385. 66. Liberek, B. 1962. B u l l . Acad. Polon. Sc i . 10:227. Cited by Bodanszky and Ondetti, 1966. 67. Liner, I. E . 1972. Nutritional value of food protein products. In "Soybeans: Chemistry and Technology, Vol . 1, Proteins. Ed. Smith,A.K. and C i r c l e , S.J.P.203. Avi Publishing Co. , Inc. , Westport, Conn. 68. L i s , M. T . , Crampton, R. F . , and Matthews, D. M. 1972. Effect of dietary changes on intest inal absorption of L-methionine and L-methionyl-L-methionine in the rat . Br. J . Nutr. 27:159. 69. L i u , T . - Y . , and Chang, Y. H. 1971. Hydrolysis of proteins with p-toluenesulfonic acid. J . B i o l . Chem. 246:2842. 70. Lunder, T. L . 1974/75. An improved method for the colorimetric determination of methionine in acid hydrolysates of biological products. Nestle Research News, p.106. 71. Matthews, D. M . , Craft, I. L . , Geddes, D. M . , Wise, I. J . , and Hyde, C. W. 1968. Absorption of glycine and glycine peptides from the small intestine of the rat . C l i n i c a l Sc i . 35:415. 72. Means, G. E . , and Feeney, R . ' E . 1971. Chemical modification of proteins. Holden-Day, Inc. 73. Merr i f i e ld , R. B. 1958. Competitive inhibit ion of a strepogenin-active peptide by related peptides. J . B i o l . Chem. 232:43. 74. Merr i f ie ld , R. B . , and Woolley, D. W. 1956. rThe Synthesis of L -. Seryl-L-hlst idyl-L-leucyl-L-vanyl-L-glutamic acid, a peptide with Strepogenin Act iv i ty . J . Amer. Chem. Soc. 78:4646. 75. Merr i f i e ld , R. B . , and Woolley, D. W. 1958. The Synthesis of peptides with strepogenin act iv i ty . J . Amer. Chem. Soc. 80:6635. 91. 76. Mukerjee, H. , and Sr i Ram, J . 1965. On the reaction of a carbodiimide with proteins and amino acids. Fed. proc. 24:418. 77. Naismith, W. E. F. 1955. Ultracentrifuge studies on soya bean protein. Biochim. Biophys. Acta 16:203. 78. Nash, A. M . , Kwolek, W. F . , and Wolf, W. J . 1971. Denaturation of soybean proteins by isoelectric precipitation. Cereal Chem. 48:360. 79. Nash, A. M . , and Wolf, W. J . 1967. Solubil i ty and Ultracentrifugal studies on soybean globulins. Cereal Chem. 44:183. 80. Obara, T . , and Kimura, M. 1967. Gel f i l t ra t ion fractionation of the whole water-extractable soybean proteins. J . Food Sc i . 32: 531. 81. Patton, S., and Josephson, D.V. 1957. A method for determining significance of vo lat i l e flavor compounds in foods. Food Research 22:316. 82. Pel let , P. L . 1974. Supplementation and complementation. In "Nutrients in processed foods." Proteins. Eds. White, P . L . , and Fletcher, D. C , p.167. American Medical Association. Mass. 83. Pomeranz, Y . , and Meloan, C. E. 1971. Food Analysis: Theory and Practice. Avi Publ. Co. , Inc. , West. Conn. 84. Previero, A . , Derancourt, J . , and Coletti-Previerp:, M-A. 1973. Solid Phase Sequential Analysis: specific l inking of acidic peptides by their carboxyl ends to insoluble resins. FEBS Letters. 33:135. 85. Puski, G . , and Melnychyn, P. 1968. Starch gel electrophoresis of soybean globulins. Cereal Chem. 45:192. 86. Rebek, J . , and Fe i t l er , D. 1973. An Improved Method for the Study of Reaction Intermediates. The Mechanism of Peptide Synthesis Mediated by Carbodiimides. J . Amer. Chem. Soc. 95:4052. 87. Rebek, J . , and Fe i t l e r , D. 1975. Peptide synthesis with carbo-diimides. III . Racemization. Int. J . Peptide Protein Res. 7:167. 88. Riehm, J . P . , and Scheraga, H. A. 1966. Structural Studies of Ribonuclease. XXI. The Reaction between Ribonuclease and a Water-Soluble Carbodiimide. Biochemistry 5:99. 92 . 89. Rubino, A . , and Auricchio, S. 1971. J . B i o l . Chem. 246:3542. Cited by Burston et a l . , 1972. 90. Schneider, F. 1960. Z. Physiol. Chem. 320:82. Cited by Schroder and Lubke, 1965. 91. Schroder, E. , and Liibke, K. 1965. The peptides. Vol . I. Academic Press, N.Y. 92. Shankman, S., and Schvo, Y. 1958. Synthesis of the Optically Active Tripeptides of Valine. J . Amer. Chem. Soc. 80:1164. 93. Sheehan, J . C , Cruick Shank, P . A . , and Boshart, G. L . 1961. A Convenient Synthesis of Water-Soluble Carbodiimides. J . Org. Chem. 26:2525. 94. Sheehan, J . C , Goodman, M. , and Hess, G. P. 1956. Peptide derivatives containing hydroxyamino acids. J . Amer. Chem. Soc. 78:1367. 95. Sheehan, J . C , and Hess, G. P. 1955. A New Method of forming peptide bonds. J . Amer. Chem. Soc. 77:1067. 96. Sheehan, J . C , and Hlavka, J . J . 1956. The use of Water-Soluble and Basic Carbodiimides in Peptide Synthesis. J . Org. Chem. 21:439. 97. Sheehan, J . C . , and Hlavka, J . J . 1957. The Cross-linking of gelatin using a water-soluble carbodiimide. J . Amer. Chem. Soc. 79:4528. 98. Sheehan, J . C . , Preston, J . , and Cruick Shank, P. A. 1965. A Rapid Synthesis of Oligopeptide Derivatives without Isolation of Intermediates. J . Amer. Chem. Soc. 87:2492. 99. Sheehan, J . C . , and Yang, D-D. H. 1958. The use of N-formylamino acids in peptide synthesis. J . Amer. Chem. Soc. 80:1154. 100. Shibasaki, K . , and Okubo, K. 1966. Tohoku J . Agr. Res. 16:317. Cited by Wolf and Cowan, 1975. 101. Smith, A. K . , and Rackis, J . J . 1957. Phytin elimination in soybean protein isolat ion. J . Amer. Chem. Soc. 79:633. 93. 102. Smith, M. H. 1968. Peptides and proteins. Molecular weights of proteins and some other materials including sedimentation, diffusion and f r i c t iona l coefficients and part ia l specific volumes. In "Handbook of Biochemistry - Selected Data for Molecular Biology". Eds. Sober, H. A . , and Harte, R. A . , CRC, Ohio. 103. Smith, M . , Moffatt, J , G . , and Khorana, H. G. 1958. Carbodiimides. VIII. Observations on the reactions of carbodiimides with acids and some new applications in the synthesis of phosphoric acid esters. J . Amer. Chem. Soc. 80:6204. 104. Spies, J . R. 1967. Determination of Tryptophan in Proteins. Anal. Chem. 39:1412. 105. Stahmann, M. A . , and Woldegiorgis, G. 1975. Enzymatic methods for protein quality determination. In "Protein nutrit ional quality of foods and feeds". Part I. Ed. Friedman, M . , Marcel Dekker, Inc. N.Y. 106. Steinman, G . , Kenyon, D. H . , and Calvin, M. 1966. The Mechanism and protobiochemical relevance of dicyanamide-mediated peptide synthesis. Biochim. Biophys. Acta 124:339. 107. Taguchi, G. 1957. Experimental Design. Maruzen Publishing Co. Tokyo. 108. Theodoropoulos, D . , and Folsch, G. 1958. Studies on h i s t idy l peptides. I. Imidazole protected hist idine in the synthesis of C-terminal and N-terminal h i s t idy l peptides. Acta Chem. Scand. 12:1955. 109. Tometsko, A. M. 1973. Determining the ava i lab i l i ty of activated amino acids for sol id phase peptide synthesis. Biochem. Biophys. Res. Com. 50:886. 110. Waddell, J . 1958. Supplementation of plant proteins with amino acids. In "processed plant protein foodstuffs". Ed. Altschul , A.M., p. 307. Academic Press, N.Y. 111. West, E . S., and Todd, W. R. 1961. Textbook of Biochemistry. The MacMillan Comp. N.Y. 112. Whitaker, J . R. 1963. Determination of Molecular Weights of Proteins by Gel F i l t ra t ion on Sephadex. Anal. Chem. 35:1950. 94. 113. Whitaker, J . R. 1977. Enzymatic modification of proteins applicable to foods. Adv. Chem. Ser. 160, Amer. Chem. S o c , Wash., D.C. 114. Wilding, M. D . , Alden, D. E . , and Rice, E . E . 1968. Nutritive value and dietary properties of soyprotein concentrates. Cereal Chem. 45:254. 115. Wolf, W. J . 1970. Soybean proteins: their functional, chemical and physical properties. J . Agr. Food Chem. 18:969. 116. Wolf, W. J . 1972A. What is soy protein? Food Technol. 26 (No. 5):44. 117. Wolf, W. J . 1972B. Purif icat ion and properties of the proteins. In "Soybeans: Chemistry and Technology". Vol . 1, Proteins, Eds. Smith, A. K . , and Circ l e , S. J . Avi Publishing Co. , Inc. , Westport. Conn. 118. Wolf, W. J . 1975. Soy proteins for fabricated foods. In "Fabricated Foods." Ed. Inglett, G. E. Avi Publishing Co. , I n c , Westport. Conn. 119. Wolf, W. J . , and Briggs, D. R. 1958. Studies on the cold-insoluble fraction of the water-extractable soybean proteins. II . Factors influencing conformation changes in the 11S component. Arch. Biochem. Biophys. 76:377. 120. Wolf, W. J . , and Cowan, J . C. 1975. Soybeans as a food source. CRC Press. Cleveland, Ohio. 121. Wolf, W. J . , and Sly, D. A. 1965. Chromatography of Soybean Proteins on Hydroxylapatite. Arch. Biochem. Biophys. 110:47. 122. Wolf, W. J . , and Sly, D. A. 1967. Cryoprecipitation of Soybean 11S protein. Cereal Chem. 44:653. 123. Yamashita, H . , A r a i , S., Tsa i , S . - J . , and Fujimaki, M. 1970. Supplementing S-containing amino acids by plastein reaction. Agr. B i o l . Chem. (Tokyo) 34:1593. 95. APPENDIX 96. Fractional factor ia l experiment (Taguchi, 1957) A 2-level experiment for selection of the factors influencing the amino acid binding to SPH. Fractional factor ia l design Run No. Factor l e v e l 3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 2 2 2 2 3 1 1 2 2 2 1 1 2 2 4 1 1 2 2 2 2 2 1 1 5 1 2 1 1 2 1 .2 1 1 6 1 2 1 1 2 2 1 2 2 7 1 2 2 2 1 1 2 2 2 8 1 2 2 2 1 2 1 1 1 9 2 1 1 2 1 1 2 1 2 10 2 1 1 2 1 2 1 2 1 11 2 1 2 1 2 1 2 2 1 12 2 1 2 1 2 2 1 1 ' 2 13 2 2 1 2 2 1 1 1 2 14 2 2 1 2 2 2 2 2 1 15 2 2 2 1 1 1 1 2 1 16 2 2 2 1 1 2 2 1 2 3 See Table 1. 97. A scheme to detect the interactions 1. Assign the nine factors (to be tested for their possible influence on amino acid binding to SPH) to columns 1,2,4,5,6,8,11,12,and 13: Column 1 PH • 2 reaction temperature 4 pepsin hydrolysis time 5 amino acid concentration 6 SPH concentration 8 reaction time 11 carbodiimide concentration 12 activation time 13 polarity The. columns shown on the connecting lines of the scheme are left blank for detection of interactions. The effect of interactions w i l l appear in those columns, i . e . : Column 3 pH x reaction temperature 7 pH x SPH concentration 9 pH x reaction time 10 reaction temperature x reaction time 14 carbodiimide concentration x amino acid concentration The unused column(s) w i l l be used for the calculation of the error term. 98. 2. Assign 2 d i f f e r e n t l e ve l s to each f a c to r . 3. Carry out runs 1 to 16 i n random order. 4. Determine i n f l u e n t i a l factors and in te rac t ions by carry ing out the ana lys i s of var iance (see Tables 2 and 3). 99. Fractional factor ia l experiment (Taguchi, 1957) A 3-level experiment for optimization of amino acid binding to SPH. Fractional factor ia l design Run Factor level No. 1 2 3 4 5 6 7 8 9 10 11 12 13 1 1 1 1 1 1 2 1 2 2 2 2 3 1 3 3 3 3 4 1 1 2 2 3 . 5 1 2 3 3 1 6 1 3;. 1 1 2 7 1 1 3 3 2 8 1 2 1 1 3 9 1 3 2 2 1 10 2 1 2 3 2 11 2 2 3 1 3 12 2 3 1 2 1 13 2 1 3 1 1 14 2 2 1 2 2 15 2 3 2 3 3 16 2 1 1 2 3 17 2 2 2 3 1 18 2 3 3 1 2 19 3 1 3 2 3 20 3 2 1 3 1 21 3 3 2 1 2 22 3 1 1 3 2 23 3 2 2 1 3 24 3 3 3 2 1 25 3 1 2 1 1 26 3 2 3 2 2 27 3 3 1 3 3 a See Table 4. 100. A scheme to detect the interactions 9 10 12 13 e • e • 1. Assign the f i v e influencing f a c t o r s , which are to be optimized, to columns 1,5,9,10 and 12: (a) methionine binding Column 1 pH 5 SPH concentration 9 carbodiimide concentration 10 a c t i v a t i o n time v .12 reaction time (b) tryptophan binding Column 1 pH 5 SPH 9 carbodiimide concentration 10 amino acid concentration 12 reaction temperature A l l of the unused columns w i l l be used for the c a l c u l a t i o n of the error term. 2. Assign 3 d i f f e r e n t l e v e l s to each factor. 3. Carry out runs 1 to 27 i n random order. 4. Carry out the analysis of variance (see Tables 5 and 6). 5. Construct response curves by p l o t t i n g theaverage of the data calculated for each l e v e l of the f a c t o r s .
UBC Theses and Dissertations
Covalent binding of methionine and tryptophan to soy protein Voutsinas, Leandros Panagis 1978
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.
- 831-UBC_1979_A6_7 V68.pdf [ 3.99MB ]
- JSON: 831-1.0094620.json
- JSON-LD: 831-1.0094620-ld.json
- RDF/XML (Pretty): 831-1.0094620-rdf.xml
- RDF/JSON: 831-1.0094620-rdf.json
- Turtle: 831-1.0094620-turtle.txt
- N-Triples: 831-1.0094620-rdf-ntriples.txt
- Original Record: 831-1.0094620-source.json
- Full Text
Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url: