‘__.. I 'I‘H 3““ . min-III. ;;;;;; “fl.— _ mm. “-w‘ _ , ~- \_ ,\ _, w— —— --‘ -._, M- - ‘ - _ _., M 4 q - __ - fi- .. -‘ -— - _- w —-——~—~ 3". “v- ._. .v ———-—t—___ ' - s...‘ __ -. —__._._:< - 5*.— N —. . . 0-“ > A—:—_. N“... - ——__. 4- ~-—.. -.-—~___‘__ HR- _ “—‘_4 . 5 -km._.‘_. ——_. —— 5 .mm‘ m“ rm *W-__. oa~ m ___, ‘— —— '- . " - —_- - ~— ‘ I c 'm 9" g‘ __——:—=_— an... - “-m...‘ :———:_'.._'—‘-"—-. -- - -- “‘W w ‘ “fl“ .— “- . _ __.;_.- a“ .:Z-_——1—__.‘-_ -. , ‘1. ’ .-._,._ «M— 2m _——..4_I::_ -k. p (I III Man—o. ‘0. .—- w W -——- w .- ‘uz-‘TJ' neg-fun“, w—v II ,n-I I. " - . II I I II III {I I'-.I III I III; "H‘IIIII'IHII" II III FIIIII‘I-y'gg. I. “I III IHIIII ”W", IIII'I III‘I, IIII’IIIII‘I‘IIIII'I13.14:. '“ .flIIIIH IIIIHWI WIII . ’IIII' IIH III III: ' ‘ I. II I. III1 II“ III" IIIII IIIIIII ‘III ”1“?!”th II“ I II IIIII 'I'I ”'II I, IIII'III W lIIIIII IIII IIIII I I: II I ’E‘IHII IIII IIII {III IIgIPI III! I [I IHMNNM'TMKIIJ' ”I , I. . II' "I“ I'E'I'" 'III .I III IIIIIIIHI SHILI .III|‘I r“ 9.... _.——-_ -._.. W5 » -——_’ My.” —-.—..__:-_~ 4;: m flw*mrfl II IIIIII IIII III I:I, II I - II‘ “I “II II 'II HIIE' IIIII 1I|I IIIII II” ,IIII I‘iIHII I This is to certify that the ‘ ‘ thesis entitled CCthQV‘LScA £35 in u iTlCO enzfimrlflc 13166371020 AND aAr PJLOASSAV (pea) PfioCEDoQ53 Fofi Hwessmemv’ OF flu; MUT’RITION'IQL. 53 uAearx/ OF ”THEzmaLcy AND CH?m lC‘floLLY MOD! FIE/D CA) 5 IN presentedby Helfn C m (Cum; has been accepted towards fulfillment of the requirements for MS degree in F0 0 o! §CAJL-U 4!” HM w a yo Ma W /' Major professor Date ‘ (low 294%» 0-7639 COMPARISON OF IN VITRO ENZYMATIC DIGESTION AND RAT BIOASSAY (PER) PROCEDURES FOR ASSESSMENT OF THE NUTRITIONAL QUALITY OF THERMALLY AND CHEMICALLY MODIFIED CASEIN BY Helen C. McCune A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Food Science and Human Nutrition 1977 ABSTRACT COMPARISON OF IN VITRO ENZYMATIC DIGESTION AND RAT BIOASSAY (PER) PROCEDURES FOR ASSESSMENT OF THE NUTRITIONAL QUALITY OF THERMALLY AND CHEMICALLY MODIFIED CASEIN BY Helen C. McCune The purpose of this study was to compare protein quality results as measured by a rat bioassay (PER) with those obtained using an in vitro enzymatic digestion pro- cedure. Vitamin free casein was treated by four different thermal and chemical procedures in order to provide exper- imental samples varying in protein quality. Using these test proteins, three different parameters of an in vitro enzymatic (pepsin-pancreatin) digestion were investigated. These included: the initial rate of digestion, the rela- tive proportions of free amino acids and peptides in the products of digestion, and analyses of the amino acids released. . The most sensitive indicator of nutritional quality of the proteins was the profile of amino acids released during digestion. Various indices based on the essential amino acids released were calculated and compared to the PER results. The Enzyme Index correlated better than either the Pepsin Pancreatin Digest Index or the Enzyme Score with the PER results. ACKNOWLEDGMENTS The author wishes to express her appreciation to Dr. J. R. Brunner for his counsel, guidance, and inspir- ation during the course of this study. Appreciation and thanks are also extended to Drs. W. G. Bergen, B. R. Bennink, and C. Markakis for advice in the preparation of this manuscript. The author especially thanks Ursula Koch for the amino acid analyses and for her aid and counsel on numerous technical matters. And finally the author is especially grateful to her husband, Jess, for his constant encouragement and support during the course of this study. ii TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . . v LIST OF FIGURES o o o o o o o o o o o 0 Vi INTRODUCTION. . . . . . . . . . . . . . 1 LITERATURE REVIEW . . . . . . . . . . . . 3 Protein Quality Determination . . . . . . . 3 Bioassays. . . . . . . . . . . . . 4 Chemical Methods . . . . . . . . . 9 In Vitro Enzymatic Methods . . . . . . . 12 Protein Damage . . . . . . . . . . . . 16 Pure Protein. . . . . . . . . . . . 16 Protein and Carbohydrate. . . . . . . . 18 Protein and Lipids. . . . . . . . . . 21 EXPERIMENTAL. . . . . . . . . . . . . . 23 Chemicals and Materials . . . . . . . . . 23 Preparation of Protein Model Systems. . . . . 23 Model System Treatments . . . . . . . . . 24 Enzymatic Digestion . . . . . . . . . . 25 Chemical Methods . . . . . . . . . . . 26 Preparation of Digestion Samples for Analyses . . . . . . . . . . . 26 Nitrogen-Micro Kjeldhal . . . . . . . 26 Alpha Amino Nitrogen-Ninhydrin Test . . . . 27 Amino Acids--Acid Hydrolysates. . . . . . 28 Methionine and Cystine Analysis . . . . . 29 Free Amino Acids--Enzymatic Hydrolysate es . . 30 Tryptophan . . . . . . . . 30 Thiobarbituric Acid Test (TBA). . . . . . 31 Peroxide Value . . . . . . . . . . . 32 Moisture . . . . . . . . . . . . . 32 iii Bioassay--Protein Efficiency Ratio (PER) . . Preparation of Diets. . . . . . . . Bioassay Method . . . . . . . . . Statistical Analysis. . . . . . . . pH Stat Evaluation-~Initial Rate of Digestion Preparation of Protein Samples . . . . Preparation of Enzyme Solutions . . . . Initial Rate Determinations . . . . . Gel Filtration Chromatography. . . . . . RESULTS I O O O O O O O O O O O O O Lipid Oxidation of the Casein-Safflower Oil Model System. . . . . . . . . . . Color Changes During Treatment . . . . . Protein Efficiency Ratio Determination (PER). Amino Acid Content of Treated Proteins (Acid Hydrolysates) . . . . . . . . . . In Vitro Enzymatic Digestion . . . . . . Rate of Digestion. . . . . . . . . Gel Filtration of Digestion Products . . Free Amino Acid Profile. . . . . . . Free Amino Acid Indices of Protein Quality Initial Rate of Digestion: pH Stat Evalu- ations. . . . . . . . . . . . . DISCUSSION 0 O O O O O O O O O O O 0 Effects of Protein Treatments. . . . . . Mild Treatments--Autoclaving and Solvent Extraction of Unoxidized Oil . . . . Casein Exposed to Oxidizing Oil . . . . Casein Autoclaved with Glucose . . . . Protein Quality Indices. . . . . . . . CONCLUSION O O C O O O C O O O O O 0 APPENDIX . . . . . . . . . . . . . SELECTED BIBLIOGRAPHY . . . . . . . . . iv Page 32 32 34 34 34 34 35 35 36 37 37 37 39 39 44 44 46 46 52 55 61 61 61 62 65 67 73 74 79 LIST OF TABLES Table Page 1. Comparison of bioassay techniques to determine protein quality (from Chavez and Pellet, 1976) o o o I o o o o o o o o o o 10 2. Composition of diets used in PER test . . . . 33 3. Peroxide values of safflower oil and TBA values of casein-safflower oil model system. . . . 38 4. Average weight gain, food consumption, protein intake, and protein efficiency ratio for rats fed either treated or untreated caseins . 41 5. Amino acid composition of treated and untreated caseins (expressed as g residue/16 g N). . . 42 6. Free amino acids in pepsin-pancreatin digest of treated and untreated casein samples. . . . 50 7. Pepsin Pancreatin Digest Index of treated and untreated casein . . . . . . . . . . 54 8. Free amino acid composition (expressed as 9/16 9 N) of pepsin pancreatin digest, and Enzyme Score and Enzyme Index of treated and un- treated casein . . . . . . . . . . . 56 9. Initial rates of pancreatin digestion of treated and untreated casein . . . . . . 58 10. Initial rate of trypsin digestion of untreated and treated casein. . . . . . . . . . 60 11. Comparison of PER results with Protein Score, Essential Amino Acid Index (EAAI), and in vitro digestion indices for treated and untreated casein . . . . . . . . . . 68 LIST OF FIGURES Figure Page 1. Principle stages of the Maillard Reaction (from Adrian, 1974) . . . . . . . . . l9 2. Average weight gain of rats fed diets contain- ing either treated or untreated casein. . . 40 3. Rates of digestion of treated and untreated casein by pepsin-pancreatin-pancreatin. . . 45 4. Chromatograms from a Sephadex G-25 column of TCA-material from in vitro enzymatic digestion of treated and untreated caseins . 48 vi INTRODUCTION New and proposed government regulations concerning the nutritional value of protein foods has renewed interest in the investigation of new methods to determine protein quality. Currently, the method used to determine the protein quality for nutritional labeling information is the Protein Efficiency Ratio (PER). This method is time consuming and expensive. Moreover, there are numerous theoretical limitations associated with the PER method. Because of the recent use of proteins from uncon- ventional sources as food ingredients, new processing techniques have been developed. Thus, there is a need for a rapid method which correlates well with the official method, PER, to monitor any loss in protein quality dur- ing processing. In response to the above needs, various chemical and enzymatic methods have been developed to provide rapid, less expensive procedures for protein quality determination. These methods include: Chemical Score, Pepsin Digest Residue Index, and Pepsin Pancreatin Digest Index. However, more work is needed before these pro- cedures can be used as a replacement of protein quality evaluation. The object of this research was to compare a rat bioassay with in vitro enzymatic methods. Various aspects of the enzymatic digestion were investigated to determine if they reflected the nutritional value of the protein. These included: the initial rate and extent of digestion and the examination of the amino acids released by enzy- matic digestion. LITERATURE REVIEW Protein Quality Determination Since the advent of nutritional labeling and other government regulations concerning protein foods, the quantitative study of protein quality has received added impetus. The Protein Efficiency Ratio (PER) is the legal analytical basis for protein nutritional label- ing (Code of Federal Regulations, 1977). Nesheim (1977) has reviewed the use of PER in nutritional labeling and many proposed FDA regulations. The current regulation states that the USA Recommended Daily Allowance (RDA) of a protein is set at the level of 45 9 if the PER of the protein is equal to or greater than the PER of casein; and 65 9 if the PER of the protein is less than that of casein (Code of Federal Regulations, 1977). Furthermore, a protein with a PER which is less than 20% of the PER for casein must be labeled "not a significant source of protein." Other pr0posed regulations concerning fortifi— cation of foods, formulated meal replacements, fortified ready-to-eat or hot cereals, plant protein products, and the addition of amino acids to foods all are based on PER determination. The PER method is time consuming and has many disadvantages both in terms of cost and method— ology. Bodwell (1977) pointed out that proteins with a PER of less than 2.4 and greater than 0.5 will be treated as if they were nutritionally equivalent in these regu- lations; obviously they are not. Therefore, the food industry and concerned scientists are endeavoring to develop faster and more reliable methods of assessing protein quality. Bioassays The evaluation of protein quality by animal bio- assay has been the subject of several reviews (Hegsted, 1977; Hegsted, 1974; Hackler, 1977). In an ideal bio- assay, the amino acid content of the protein, the bio- availability of each amino acid, and the amino acid requirement of the test animal should be known qualities. At the same time, the relationship between an animal's response to a protein and a human's response should be recognized. These facts, however, have not been entirely elucidated. Moreover, there are a myriad of other factors which have been known to influence bioassays, including: age, sex, and weight of the animal, protein quality and quantity, food intake, other dietary components such as carbohydrate, minerals, vitamins, fat, and water, hus- bandry, and environmental conditions. Mitchell (1924) utilized the principle of nitro- gen balance to develop a method suitable for determining protein quality. His index, Biological Value (BV), is defined as a ratio of retained nitrogen to absorbed nitrogen of a test animal. This method is technically cumbersome becauseof the difficulty in collecting feces and urine, and in accurately estimating the endogenous urinary and fecal nitrogen (Williams et al., 1974). Another drawback of this method is that there is no allowance for digestibility. The development of the Net Protein Utilization (NPU) (Bender and Miller, 1953) and Net Protein Ratio (NPR) (Bender and Doell, 1957) techniques eliminated some of the difficulties associated with the EV method. Both NPU, which is defined as car- cass nitrogen of a test group minus the carcass nitrogen of a protein-free group divided by the nitrogen consumed, and NPR, represented by the weight gain of a test group plus the weight loss of a protein-free group divided by nitrogen consumed, include protein digestibility as part of the assessment. Although NPU and NPR are tech- nically less rigorous than BV, the carcass nitrogen determination can present problems. In the calculation of NPR, the assumption that the body composition is constant may not be valid when widely different test and protein deficient diets are fed to the animals. The major drawback with the above methods is the dependence of the results on the level and quality of dietary pro- tein. For example, the EV of meat, oats, corn, and potatoes is lower for a diet containing 10% protein than it is for a diet containing only 5% protein. This dif- ference in apparent BV is smaller with the poorer quality proteins such as potatoes and oats (McLaughlin, 1972). Hegsted and co-worker (1965, 1969) developed an assay relating nitrogen balance or weight gain to nitrogen intake. In this procedure a protein is fed at different levels; the slope of the growth response curve is an indication of protein quality. However, the slope may not always be a valid index of protein quality. For example, if lysine is the limiting amino acid, a lower than expected slope is obtained: and if the limiting amino acid is threonine, a higher than expected slope is noted. In addition, it is sometimes difficult to select the linear portion of the response curve. The PER method (Osborn and Mendel, 1919) which is defined as the ratio of weight gain over protein consumed by a weanling rat for a 28-day test period, is probably the simplest bioassay. There are, however, some fundamental problems associated with PER as an index of protein quality. No allowance is made for amino acid maintenance requirements of the test animal. It is also difficult to measure complimentary effects of two or more proteins, when using PER as an indication of protein quality. In addition to BV, NPU, and NPR, McLaughlin (1972) found that PER, too, varied with the amount of .7 protein in the diet. For example, the highest PER for egg protein is obtained when the protein is fed at the 8% level. In contrast, for wheat protein the highest PER is observed when protein is fed at a 15% level. Since PER is the basis of nutritional labeling, various researchers have studied factors which might affect the results (Morrison and Campbell, 1960; Hurt et al., 1975; Hegarity, 1975; Steinke, 1977). The results of these studies indicated that: (1) younger rats (21-23 day old versus 29 day old) yield a higher PER; (2) different rat strains vary in their PER response, but standardization of test results with respect to the apparent PER of casein eliminates the problem; and (3) female rats grow slower than males thus producing dif- ferent PER values. Studying the effects of the nonprotein dietary components, Hurt et a1. (1975) determined that the mineral content and dietary fat composition (satur- ated versus unsaturated) did not affect the PER. How- ever, the level of dietary fat and fiber significantly influenced the results. Fiber and fat were found to affect the PER of casein to the same extent as the test protein. Therefore, if both the casein and test diets are balanced with respect to these ingredients, the results will not be biased. Steinke (1977) discussed the effect of different carbohydrates on PER. He found that rats prefer a sweeter tasting diet. Consumption of a sweet diet will exceed that of an unsweetened diet of similar composition. Hopkins and Steinke (1976) found that the hydration of a protein prior to mixing improved the PER. Interestingly, the hydration of soy isolate raised the PER to a greater extent than the hydration of casein. More recently bioassays have been developed to eliminate some of the problems associated with PER. McLaughlin and Keith (1975) developed a modification of the PER method which incorporated the maintenance requirement into the calculation. In this method, Nitrogen Utilization (NU) is defined as weight gain plus 0.1 times the initial weight plus final weight divided by the protein consumption. NU is often ex- pressed as Relative Nitrogen Utilization (RNU) when it is compared to a standard which is generally lactalbumin. Another new method which includes the maintenance requirement in the value for protein quality is the Weight Gain Coefficient (Canolty and Koong, 1977). The slope of the line which correlates weight gain, expressed as kg of body weight per metabolic body size, and feed con- sumption is defined as the Weight Gain Coefficient. More data on these methods are indicated before they could replace PER as the standard method. These assays are not rapid; requiring 3-4 weeks. Chevez and Pellet (1976) compared several bio- assays for twelve different food mixtures common to Latin America. These included Relative Protein Value (RPV) (slope assay calculated without zero protein data), Relative Nutritive Value (RNV) (lepe assay calculated with zero protein data), PER, and NPR. Regardless of the assay employed, all the dietary proteins were simi- larly ranked (see Table 1). In another comparison, McLaughlin and Keith (1975) determined the PER, RNU, slope assay, and NPR for twelve different proteins of varying protein quality. They found similar values for protein quality by each method, except that the PER values tended to underestimate the quality of the lower quality proteins. Chemical Methods Other methods used to evaluate the nutritional quality of proteins have been based on the chemical analysis of the amino acid content of the protein. Mitchell (1946) based his method on the fact that all amino acids required for the synthesis of proteins must be present at the site of synthesis; and that protein synthesis would be limited by the essential amino acid present in the least supply. His assessment of protein quality, Chemical Score, is represented by the smallest ratio of an essential amino acid of a test protein to the same amino acid in whole egg protein. Egg protein 10 Table 1. Comparison of bioassay techniques to determine protein quality (from Chavez and Pellet, 1976). First Set of Samples Food Relative RNV RPN PER NPR NPR Lactalbumin 1.00 1.00 4.00 4.60 1.00 Rice, Beans 0.85 0.95 3.00 3.70 0.81 Cassava, Beans, Plantain, & Cheese 0.85 0.99 2.80 3.70 0.81 Corn, Beans 0.70 0.74 2.10 3.10 0.68 Bread, Cheese, & Beans 0.70 0.77 2.00 3.10 0.66 Spaghetti, Cheese, & Tomato Sauce 0.64 0.68 1.80 2.90 0.62 Arepa & Margarine 0.42 0.28 0.50 1.90 0.42 Second Set of Samples Food Relative RNV RPN PER NPR NPR Lactalbumin 1.00 1.00 4.50 4.80 1.00 Arepa, Meat 1.04 1.14 4.50 4.80 1.00 Arepa, Sardines 0.87 0.94 3.90 4.20 0.86 Arepa, Cheese 0.86 0.92 3.80 4.10 0.83 Rice, Plantain, & Sardines 0.82 0.75 3.60 3.80 0.79 Rice, Beans, & Plantain, Meat 0.76 0.80 3.30 3.60 0.75 Arepa, Beans 0.69 0.76 2.90 3.30 0.67 11 has been found to have excessive amounts of certain amino acids relative to the requirements of growing rats. In particular, tryptophan, and to some extent, the sulfur amino acids are in excess; however, lysine is present at about the requirement level. Therefore, if the limiting amino acid is lysine, the Chemical Score and PER will be proportional (McLaughlin, 1972). Because of the excess of essential amino acids in the egg pattern, the FAQ/WHO (1973) proposed a pro- visional amino acid pattern based on the amino acid requirements of pre-school infants. The index of protein quality using this pattern of amino acids is referred to as the Amino Acid Score. Kaba and Pellet (1975) com- pared various amino acid scoring patterns (whole egg protein, human milk protein, whole rat carcass, FAO, 1957, and FAD/WHO, 1973). The FAO/WHO (1973) pattern was superior to the other patterns in predicting the limiting amino acids for test diets in which the first limiting amino acids were either lysine or methionine. In addition, the correlation between amino acid score, using the 1973 pattern, and NPU with rats was higher than when other patterns were used. Oser (1951) devised a criterion for protein quality, Essential Amino Acid Index, based on the come parison of the essential amino acids of a test protein to the essential amino acids of whole egg protein. In 12 contrast to the Chemical Score, which is merely the expression of the smallest ratio, the Essential Amino Acid Index is the geometric mean of all the ratios of the essential amino acids of the test protein to that amino acid of the reference protein. This index is generally higher than the Chemical Score. The major problem with chemical evaluations of protein quality is the assumption that all the amino acids are biologically available; an assumption which is not always valid. On the other hand, the advantages of these indices are: (1) the small sample size required, (2) the short time of analysis compared to bioassays, and (3) the information provided concerning the identity of the limiting amino acid. In Vitro Enzymatic Methods Development of enzymatic digestion methods to assess protein quality began when Melnick et a1. (1946) reported that different proteins, or proteins which had been processed differently, were hydrolyzed by pancreatin at different rates. Evans and Butt (1949) found that by autoclaving soya, the release of some of the amino acids by enzymatic digestion was retarded. Mauron et a1. (1955) examined the availabilities of lysine, methionine, and tryptophan in different milk products by analyzing the amino acids liberated by in 13 vitro enzymatic digestion. The milk products were digested with pepsin for 14 h followed by pancreatin for an additional 24 h. The digestion was run in a dialysis apparatus in which the end products were con- stantly being removed in order to avoid any possible end product inhibition. As an improvement over the evaluation methods which only measured digestion rate or amino availability, Sheffner et a1. (1956) employed an index based on an in vitro digestion with pepsin to measure protein quality. They used sufficient pepsin to obtain approximately 30% hydrolysis of the test protein. This index was based on the assumption that the initial products of digestion represented an indication of the nutritional value of the protein. After pepsin digestion, the samples were deproteinated and the released essential amino acids were determined by microbiological methods. A Pepsin Digest Residue Index (PDRI) was then calculated by comparing the released essential amino acids and those remaining undigested with the pattern of essential amino acids found when whole egg protein was digested in a similar manner. The calculation also involved cor- rection for the difference in degree of proteolysis between the test and reference protein. For the proteins tested, the EV using rats and PDRI/digestibility were highly correlated. 14 The microbiological determination of the amino acids for the PDRI was slow and laborious. In 1964, Akeson and Stahmann used an amino acid analyzer to determine the essential amino acids released during digestion with pepsin and pancreatin. They used the same calculations as the PDRI to obtain a Pepsin Pan- creatin Digest Index (PPDI). The results of the two indices were similar. The suggested reason for the similarity in the results of the two indices was that during the microbiological determination of amino acids some peptides were utilized by the microbes; thus, increasing the apparent amount of free amino acids. In the PPDI, the use of the pancreatin released an equivalent amount of free amino acids as the microbial utilization (Mauron, 1970). Ford and Salter (1966), like Mauron, were con- cerned about end product inhibition. Therefore, they digested samples on a column of Sephadex G-10 to provide for chromatographic removal of small digestion products (less than 700 daltons). After digestion and deprotein- ation, the samples were fractionated on a Sephadex G-25 column. Amino acid analysis was done of the fraction with a molecular weight of less than 250. Comparison of the amino acid results of the static digests of Akeson and Stahmann (1964) and the dynamic digestion methods of Ford and Salter (1966) and Mauron (1970) revealed 15 no apparent differences. These data indicated that end product inhibition was inconsequential to these assays (Stahmann and Woldegiorgis, 1975). PPDI, PDRI, and NPU were used to monitor the effect of heat processing casein (Stahmann and Wolde- giorgis, 1975). It was observed that all three methods predicted similar values for protein quality. An Enzyme Score, which is calculated like the Chemical Score, com- paring the essential amino acids released from the test protein and those released from egg protein, was also determined on the heat processed casein. The Chemical Score and Enzyme Score compared well with the PPDI. However, the Chemical Score tended to underestimate the quality of the most severely heated protein. The major problem with the use of in vitro digestion methods as an evaluation of protein quality concerns the incomplete digestion of the test protein. Therefore, a reference protein is always assayed in con- junction with the test protein. The advantages of using the in vitro enzymatic digestion methods for deriving protein quality indices include: (1) lower cost and shorter analysis time, (2) less variation in in vitro results than found with rat bioassays, (3) the gener- ation of information on amino acid availability and relative nutritional adequacy (Stahmann and Woldegiorgis, 1975). 16 More recently, Satterlee et a1. (1977) employed a simplified in vitro digestion to develop a value for predicting PER. These authors have used a recording pH meter to determinethe extent of digestion by trypsin, chymotrypsin, and petidase. The pH after 10 min was found to correlate with apparent digestibility deter- mined in vivo. A combination of the digestibility and the essential amino acids content of the protein, expressed as a percentage of the FAO/WHO-l973 pro- visional pattern, was used to derive an apparent PER. The actual rat PER and the apparent PER were compared for 45 different food samples. The average difference between the estimates was 0.12 PER units. Protein Damage The effect of processing on protein quality has been reviewed by Bender (1972). He classified protein damage into four categories: (1) amino acid destruction by oxidation, (2) loss of palatability, (3) modification of some linkages, or (4) formation of enzyme resistant linkages. The extent of damage is usually dependent on food composition and processing conditions. Pure Protein In the preparation of protein isolates and the texturization of vegetable protein, the temperature of processing often exceeds 100 C. Bjarnason and Carpenter 17 (1970) studied the mechanism of damage of pure proteins. After heating bovine plasma albumin 27 h,at 115 C, the cysteine and lysine contents were reduced. At 145 C for 27 h, all the amino acids except glutamic acid and those with paraffin side chains were reduced. These authors suggested that the loss of lysine was due to the reaction of its e—amino group with the amide group of asparagine and glutamine. In addition, the e-amino group of lysine appeared to react with the decomposition products of cysteine. A number of researchers have investigated the nutritional significance of the lysine-containing pep- tides formed during heat treatment in the absence of carbohydrate or fat (Ford and Shorrock, 1971; Warbel and Carpenter, 1972; Hurrell and Carpenter, 1976). Their results indicated that the formation of the lysine-con- taining peptides do not totally explain the low biologi- cal value of the heated protein. Ford (1973) suggested that the formation of lysine-glutamine bonds may hinder the access of peptidases to adjacent bonds in the protein. Severe heat treatment of casein (1-8 h at 120- 130- C) induced a decrease in aspartic acid, threonine, serine, cysteine, histidine, and lysine. In contrast, glutamic acid and alanine were found to increase after the heat treatment (Osner and Johnson, 1974). The BV and NPU of casein, which had been heated under the same 18 conditions as above, decreased significantly (Osner and Johnson, 1975). The microbiological availability of valine, methionine, isoleucine, leucine, histidine, arginine, and tryptophan decreased after the heat treat- ment. In addition, the in vitro pepsin digestibility of the heated casein was found to decrease after severe heat processing (Osner and Johnson, 1968). Protein and Carbohydrate The Maillard reaction, or nonenzymatic browning, has been studied extensively (Bender, 1972; Janicek, 1973; Feeney et al., 1975; Adrian, 1974). In food products, generally the amino groups from proteins and aldehyde groups from either reducing sugars or carbonyl compounds formed during lipid oxidation are the reactive species. The series of reactions results in brown pigments and a destruction of certain amino acids (see Figure l). The loss of amino acids occurs long before melandoidin pig- ments are formed. The first reaction product, an aldo- samine, is not considered to be nutritionally available. Although it is hydrolyzed with concentrated acid, the aldosamine is not hydrolyzed during digestion (Feeney et al., 1975). For example, in over-heated milk, un- available lysine was found to be in either the aldosamine, Schiff base, or deoxy-ketosyl form. After acid hydroly- sis, lysine in the aldosamine or Schiff base was regen- erated, but only 49.5% of the lysine in the deoxy-ketosyl 19 SUGAR + AMINO ACID NITROGEN-SUBSTITUTED GLYCOSAMINE AMADORI REARRANGEMENT l-AMINO-l-DEOXY-Z-KETOSE MILD FISSION STRONG DEHYDRATION DEHYDRATION PYRUVALDEHYDE FURFURALS DIACETYL ECT. DEHYDROFURFURALS, HMF AND BANAL ARMA- AND BANAL AROMATIC TIC SUBSTANCES SUBSTANCES REDUCTONES DEHYDRORE- DUCTONES STRECKER DEGRADATION ALDELYDES + CO2 POLYMERIZATION AND INSOLUBILIZATION MELANDOIDINS Figure 1. Principle stages of the Maillard Reaction (from Adrian, 1974) 20 form was recovered (Finot, 1973). Tanaka et a1. (1975) examined the metabolism of fructose-L-tryptOphan, formed after Amadori rearrangement, using rat feeding studies. While a portion of the fructose-L-tryptophan could be absorbed intestinally, most of the compound was excreted in the urine without being metabolized. The nutritional effects of the Maillard reaction were well illustrated by the results of Baldwin et a1. (1951) who observed that the PER of a casein-glucose model system, which had under- gone some browning during autoclaving for only 3 min, decreased by 25%. The Maillard reaction can occur during storage, and certain conditions increase the rate of the reaction. Browning increases with temperature, is favored by alka- line conditions, and is strongly dependent on water activity (Bender, 1972). The type of sugar influences the rate of the reaction; pentoses being more reactive than aldoses (Tu and Eskin, 1973). The amino acids most affected by the Maillard reaction are generally the N-terminal chain amino acids, followed by the basic amino acids, the sulfur amino acids, and sometimes tryptophan (Feeney et al., 1975). The reaction between the e-amino group of lysine and reducing sugars has been extensively examined by Car- penter (1973). In addition to the amino acids mentioned above, histidine, and threonine are subject to heat 21 degradation in the presence of carbohydrate (Bender, 1972; Baldwin et al., 1951; Ford and Salter, 1966). Protein and Lipids The presence of oxidizing lipid can also cause damage to proteins. The destruction of sulfhydryl groups in either amino acids or SH enzymes occurred more rapidly in the presence of oxidized than unoxidized linoleic acid (Lewis and Wills, 1962). Methionine, histidine, cysteine, and lysine were found to be particularly sensitive to free radicals produced by peroxidizing lipids (Roubal and Tappel, 1966). Andrews et a1. (1965) investigated the reaction between insulin and oxidizing methyl lino- leate. They found that the intermediates of fatty acid oxidation react with the e-amino groups of lysine and the N-terminal amino acids, phenylalanine and glycine. Porkony (1973, 1975) examined the browning of proteins which were exposed to oxidizing lipids. Tan- nenbaum et a1. (1969) found that the methionine content of casein, which had been mixed with methyl linoleate and incubated for 67 days, decreased as the browning increased. Malonaldehyde (MA), a dialdehyde intermediate of fatty acid oxidation, appeared to react with amino acids and form fluorescing compounds (Karel, 1973). When reacting the protein myosin, MA was found to selectively attack histidine, arginine, tyrosine, and 22 methionine (Buttkus, 1967). Crawford et a1. (1967) studied the mechanism and kinetics of the reaction between bovine serum albumin and MA. The N-terminal amino acids and the amino group of lysine were the reactive sites on the protein. When ribonuclease was incubated with oxidizing lipid, MA was implicated in the inactivation of the enzyme (Chio and Tappel, 1969). The nutritional consequences of the protein lipid interaction have been studied. For example, the PER for herring meal was observed to decrease from 2.94 to 2.53 after lipid oxidation had occurred (Carpenter et al., 1963). Other researchers have studied the effects of oxidized ethyl linoleate on casein nutritional quality (Yangita et al., 1973). The in vitro digestibility of casein decreased after the reaction with oxidized ethyl linoleate. The available lysine of casein decreased after incubation with an oxidized fatty acid. Lysine and methionine were destroyed to the greatest extent (Horigome et al., 1974; Horigome and Muira, 1974). When egg albumin and oxidized ethyl linoleate were combined, BV and true digestibility decreased. Histidine, lysine, arginine, and methionine were reduced to the greatest extent after exposure of the egg albumin to oxidized ethyl linoleate (Yanagita and Sugano, 1974). EXPERIMENTAL Chemicals and Materials Pepsin, hog stomach mucosa, was 1-10,000X pur- ified. Pancreatin, hog pancreas, was 5X crystallized. Trypsin was 2X crystallized and salt free. The three enzymes were obtained from ICN Pharmaceuticals. Other chemicals from various commercial sources were employed and will be mentioned in the description of the method where used. Preparation of Protein Model Systems Vitamin-free casein (INC Pharmaceuticals) was used in all the following model systems: 1. Casein-Glucose Model System A 25% (w/v) glucose solution was prepared using deionized water. When the glucose was completely dis- solved, sufficient casein was added to produce a slurry which was approximately 50% total solids, containing equal parts of glucose and protein. The mixture was mixed vigorously in a Warring Blender and freeze dried. 23 24 2. Casein-Safflower 011 Model System A 20% (w/v) safflower oil (PVO International Inc.) mixture was prepared with deionized water and homogenized at 2000 psi with a Chase-Logeman Laboratory Homogenizer. Employing a Warring Blender, sufficient casein was blended into the safflower-oil-water mixture to yield a 40% total solids slurry, containing equal quantities of safflower oil and protein. In order to insure rapid oxidation of the oil, 0.2 ppm of copper in the form of cupric sulfate was added. The mixture was blended vigorously and freeze dried. A copper-free casein-safflower oil model system was prepared as above. Model System Treatments 1. Casein-Safflower Oil Model System (minus copper) Immediately after freeze drying, the casein- safflower oil mixture was extracted twice with hexane (1 part to 5 parts hexane), and one time with diethyl ether (1 part to 5 parts diethyl ether). After extract— ing the oil from the system, the residual casein remained under the hood for approximately 12 h to exhaust residual ether and placed in vacuum oven (30 mm Hg pressure) for 5 h at ambient temperature to insure that all traces of solvent were removed. 25 2. Casein-Safflower Oil Model System (with copper) After freeze drying, the casein-safflower oil combination was incubated 10 days at 55 C to promote oxidation of the oil. Following the incubation period, the oil was extracted as described above. 3. Casein Casein was spread approximately 1-2 in. thick in Pyrex pans which were covered with aluminum foil. The casein was then autoclaved for 5 min at 121 C (15 psi). 4. Casein-Glucose Model System The casein-glucose mixture, after freeze drying, was autoclaved in the same manner described for casein. Enzymatic Digestion Both treated and untreated proteins were par- tially dissolved in deionized water by slowly adjusting the pH to 1.8 with l N HCl. Sodium azide was added at the 0.02% level to prevent microbial growth. The volume was adjusted with deionized water to obtain a 1% protein solution. Thirty m1 of the 1% protein solutions were added to 125 m1 flasks along with 15 mg pepsin. After the pepsin was dissolved, the flasks were stoppered and placed in a 37 C incubator for 3 h. Peptic digestion was terminated by the addition of 1N NaOH until pH 8.3 was reached. Deionized water was added to make the total addition of liquid equal to 5 ml. Then 15 mg of pancreatin was added to each flask. 26 The contents of the flasks were mixed and reincubated at 37 C. After 3 h of pancreatin digestion, a second 15 mg portion of pancreatin was added and the digestion con- tinued for an additional 21 h. Enzyme blanks were treated identically except for the absence of the protein sample. Chemical Methods Preparation of Digestion Samples for Analyses The digestion of the samples was stopped by adding sufficient trichloroacetic acid (TCA) to obtain a final concentration of 15%. After sitting overnight at 4 C, the samples were centrifuged in a clinical cen- trifuge at 1000x g for 5 min to remove the precipitated material. The clear supernatant was analyzed for nitro- gen, alpha amino nitrogen, and free amino acids. Nitrogen-Micro Kjeldhal Approximately 15 mg of dried protein were digested in duplicate with 4 ml of digestion mixture over a gas flame for 1 h. The digestion mixture con- tained 5.0 g CuSO :8H 0 and 5.0 g SeO in 500 m1 of 4 2 2 concentrated H2S04. After cooling the flasks, 1 ml of 30% H202 was added, and digestion was continued for an additional hour. Each digestion flask was then cooled, and the sides of the flask were rinsed with 10 ml of deionized water. The digestion mixture was neutralized 27 with 25 ml of a 40% NaOH solution. The released ammonia was steam-distilled into 15 ml of a 15% boric acid solution containing 5 dr0ps of bromocresol green, methyl red indicator, containing 400 mg bromocresol green and 40 mg methyl red in 100 ml 95% ethanol. The distillation was continued until the volume of the boric acid receiver reached 60 ml. The ammonium borate complex was titrated with 0.02 N HCl which had been standardized with a stan- dardized solution of NaOH. The recovery of tryptophan N served as a control. Nitrogen was calculated as follows: (m1 HCl - ml blank) (Normality of HCl)x100 %N = mg of sampIe Alpha Aminngitrogen- Nifihydrin Test The method of Clark (1964) for the determination of alpha amino nitrogen was followed. Aliquots (0.5 ml) of deproteinated diluted samples were pipetted into test tubes. The dilution factors ranged from 10, 25, to 100 for the samples from the enzyme blank, peptic digest, and second pancreatin digest, respectively. A reagent blank was run with 0.5 m1 of deionized water. Then 1.5 m1 of ninhydrin solution (see Appendix for prepar- ation) was added to each tube. The contents of the tubes were mixed well, and placed in a boiling water bath for 20 min. After heating, the tubes were cooled and 8 ml of a 50% (w/v) aqueous n-propanol solution was 28 added to each tube. The contents were mixed vigorously. After 10 min, the absorbance was read at 570 nm against the reagent blank on a Beckman D K 2 A Double Beam Spec- trophotometer. A standard curve was prepared using glycine. Amino Acids-~Acid Hydrolysates Amino acid analyses were performed on HCl hydroly- sates of protein using a Beckman Amino Acid Analyzer, Model 120 C, according to the procedures of Moore et a1. (1958). Samples consisting of approximately 4 mg of protein were weighed into 10 m1 ampoules. Five m1 of 6 N HCl were added to the ampoules. The contents were frozen in a dry-ice-ethanol bath. The ampoules were evacuated with a high vacuum pump. As the contents slowly melted, the gases were removed. The contents were then refrozen and the ampoules were sealed using an air-prOpane flame. The sealed ampoules were placed in an oil bath in a forced draft, recirculating oven regulated at 110 i 2 C for 24 or 48 h. After hydrolysis, the ampoules were opened, and 1 ml of norleucine solution (2.5 umoles/ml) was added as an internal standard. The hydrolysate was then quanti- tatively transferred from the ampoule to a 25 m1 pear- shaped flask. The hydrolysate was evaporated to dryness on a rotary evaporator. The dried sample was washed with a small amount of deionized water and again taken 29 to dryness. In all, three washings were performed to remove residual HCl. The washed and dried hydrolysate was dissolved in 0.067 M citrate HCl buffer (pH 2.2), and diluted to a volume of 5 ml. The solution was then filtered with a 0.22 pm Millipore Filter, and 0.2 m1 aliquots were used for analysis. The Chromatograms were quantitated by peak integration using a Spectra Physics Autolab System AA. Standard amino acid mixtures were analyzed using the same ninhydrin solution within a four-day period. Methionine and Cystine Analysis Since methionine and cystine undergo a variable amount of oxidation during acid hydrolysis, they must be analyzed separately. The methods of Schram et a1. (1964) and Lewis (1966) were used. These methods involve per- formic acid oxidation of methionine and cystine to methionine sulfone and cysteic acid, respectively. Approximately 10-20 mg of sample, representing 5-8 mg of protein, was weighed into a 25 m1 pear-shaped flask. The protein was oxidized for 24 h with 10 m1 of per- formic acid at 4 C. After oxidation, 1 ml of norleucine (2.5 umoles/ml) was added. The performic acid was removed on a rotary evaporator. The dried sample was quantitatively transferred to a 10 m1 ampoule with 5 ml of 6 N redistilled HCl. Hydrolysis and amino acid analyses were performed as previously discussed. 30 Free Amino Acids--Enzymatic Hydrolysates The 27 h peptic-pancreatin digestions were depro- teinated with 15% TCA. After centrifugation to remove the precipitated protein, the supernatant was filtered with 0.8 pm Millipore Filter. Then, 2 ml of the filtrate plus 1 m1 of norleucine solution (2.5 umoles/ml) were diluted to 5 ml with 0.67 M citrate HCl buffer (pH 2.2). Because the enzyme blank contained much less nitrogen, a 12 m1 sample plus 1 m1 of the norleucine solution was evaporated to dryness on a rotary evaporator. The dried sample was then diluted to 5 ml with the same citrate buffer. The diluted sample was filtered with a 0.22 pm Millipore Filter. Amino acid analyses were performed as previously described. Since acid hydrolysis was not employed, the sulfur amino acids were not destroyed and could be quantitated directly. Tryptophan The method of Spies (1967) was followed. This method involved a 24 h pronase (Sigma Chemical Co.) digestion, followed by the production of a chromaphore resulting from the action of free tryptophan with p- dimethylaminobenaldehyde (Matheson, Coleman, Bell Co.). Absorbance of the sample was determined against a reagent blank at 590 nm on a Beckman D K 2 A Double Beam Spec- trophotometer. A pronase blank was also run so that 31 samples could be corrected for the tryptophan content of the pronase. A standard curve was prepared with trypto- phan which had been dried over P205. In the case of the free amino acids, the same procedure was followed except that the pronase digestion step was eliminated. Thiobarbituric Acid Test (TBA) The extent of lipid oxidation in the casein- safflower oil model system with copper was monitored with a modified TBA test procedure of King (1962). A 1.00% protein solution was prepared with deionized water. The mixture was blended in a Warring Blender on high speed for 3 min. Aliquots (15 ml) were placed in small flasks. Two ml of aldehyde-free redistilled ethanol and 2 m1 of 50% TCA were added, followed by vigorous shaking. The samples were held for 30 min and filtered through Whatman no. 1 and no. 42 filter paper, consecu- tively. Four ml of each filtrate was pipetted into a screw-cap test tube. One ml of TBA reagent was added to each, followed by mixing and incubation in a 60 C water bath for l h. After cooling, the absorbance of the sample was read versus distilled water at 532 nm on a Beckman D K 2 A Double Beam Spectrophotometer. See Appendix for description of reagent preparation. 32 Peroxide Value The peroxide value was determined according to the AOAC (1975) method. To approximately 5 g of oil sample, 30 m1 of an acetic-acid-chloroform (3:2) mixture was added. Then, 0.5 m1 of a saturated KI solution was added, followed by 30 m1 of deionized water. The solution was titrated with a standardized 0.1 N Na28203 solution. The peroxide value, expressed as meq. peroxide/kg sample was calculated as follows: (ml of NaZSZOB) (N of NaZSZOB) x 1000 g sampIe Moisture Approximately 1 9 samples were dried in a vacuum oven (30 mm Hg pressure) at 90-95 C for 20 h. The water loss was determined gravimetrically. Bioassa --Protein Efficiency Rat1o (PERYG The PER method of AOAC (1975) was followed. Preparation of Diets Five diets were prepared using either the un- treated or treated casein samples. The composition of the diets is shown in Table 2. Each diet was thoroughly mixed in a Hobart Blender. 33 Table 2. Composition of diets used in PER test. . 1 Ingredient D1et 1 2 3 4 5 Treated casein2 Untrt 11.03 Exunox 11.23 Exox 12.90 Auto 10.60 Auto Glu 19.42 CotEBHEaaa‘bi1 8.00 8.00 8.00 8.00 8.00 Salt mix3 5.00 5.00 5.00 5.00 5.00 Vitamin mix 1.00 1.00 1.00 1.00 1.00 Cellulose (A1- phace11)5 1.00 1.00 1.00 1.00 1.00 Water 4.22 3.82 4.10 4.49 3.96 Corn starch 69.75 69.95 68.00 69.91 61.62 1 protein and 5% water. contents of casein samples. 2 ‘Expressed in percent. Each diet contains 10% See Appendix for N and moisture Vitamin free casein from ICN Pharmaceuticals. Untrt, Untreated casein; Exunox, casein after extraction of unoxidized safflower 01I; Exox, casein after extrac- tion of oxidized safflower oiI; Auto, casein autoclaved 5 min 121 C; Auto Glu, casein and qucose autoclaved 5 min 121 C. 3 USP XVII, 4“Vitamin Diet Fortification Mixture, ceuticals. 5 ICN Pharmaceuticals. ICN Pharmaceuticals. ICN Pharma- 34 Bioassay Method After a 4-day acclimatization period on a standard casein diet, male, weanling, Spraque-Dawley rats were randomly assigned to 5 groups of 10 rats each. Rats were housed individually in metal cages with raised wire mesh floors. Water and diets were fed ad libum. Food intake and body weight were recorded at 3- and 4-day intervals, respectively, for the 28-day test period. PER (weight gain/protein intake) was calculated from the weight gain and protein consumption of each rat. Statistical Analysis The Analysis of Variance with a completely ran- domized block design and the Tukey procedure were used to analyze the data (Neter and Wasserman, 1974). pH Stat Evaluation--Initial Rate of Digestion Initial rates of pancreatin and trypsin diges- tions were determined on either intact proteins or pepsin pretreated proteins using a Sargent pH Stat. Preparation of Protein Samples A 1% protein solution, using either the treated or untreated casein samples, was prepared with 0.005 M TRIS, 0.04M NaCl buffer (pH8.3). For the pepsin pre- treated proteins, a 1% protein solution was adjusted to pH 1.8. Pepsin was then added in the same enzyme to 35 substrate ratio (1:60) as that employed in the in vitro enzymatic digestion described above. After 3 h of digestion at 37 C, the pH was raised to 8.3 with l N NaOH, and the sample was ready for initial rate determinations. Preparation of Enzyme Solutions The pancreatin solution was prepared with 0.01 M HCl, 0.01 M CaCl2 buffer (pH 3.8) to a concentration of 7 mg/ml. This solution was kept in an ice bath until ready for use. Sufficient pancreatin solution for l h of experimentation was adjusted to pH8.3 with concentrated NaOH and placed in an ice bath until needed. The trypsin solution (5.6 mg/ml) was prepared with the same buffer that was employed for the pancreatin solution. Since a much smaller amount of trypsin was used in the assay, no pH adjustment was required prior to use. Initial Rate Determinations Seven ml of protein solution, either intact or pepsin pretreated samples, were added to the reaction vessel. When the temperature reached 37 C and a stable base line was achieved, the enzyme was injected into the vessel. One ml of the pancreatin solution was used for the determination of the pancreatin initial rate. For the tryptic digestion, the amount of trypsin solution used was experimentally determined so that the initial rate of the tryptic digestion of the untreated casein 36 equalled that rate obtained when pancreatin was used. This amount was 0.75 ul of trypsin solution. A mag- netic stirring bar maintained continuous mixing while the titrant, 0.1 M TRIS, was added. The initial rates, expressed as ml of base delivered/min, were determined by calculating the slope of the tangent to the reaction curve at zero time. The data were analyzed using Analysis of Variance with a completely randomized block design and the Tukey procedure (Neter and Wasserman, 1974). Gel Filtration Chromatography A Sephadex G-25 column (2.6 cm x 36 cm) was pre- pared according to Pharmacia Fine Chemical (1974). The column was equilibrated with 0.1 M ammonium acetate buffer (pH 7.0) with 0.1% sodium azide. A void volume of 75 ml was determined using Blue Dextran. RESULTS Lipid Oxidation of the Casein-Safflower Oil Model System When the casein-safflower oil model system with copper was incubated for 10 days at 55 C, the oil became quite oxidized as indicated by increased peroxide and TBA values (Table 3). The peroxide value rose by a factor of 100. The TBA values, which measures the fatty acid oxidation intermediate, malonaldehyde (MA), doubled. After the oxidized oil had been extracted, the TBA value of the extracted casein was still slightly higher than that of either the casein-safflower oil model system before incubation or the untreated casein. This indi- cates that possibly some MA was still present after sol- vent extraction. Color Changes During Treatments During the oxidation of the oil, the model system turned a light amber color which remained following sol- vent extraction. The autoclaved casein-glucose mixture turned a light tan color. However, autoclaving casein alone caused no color change. 37 38 Table 3. Peroxide values of safflower oil and TBA values of casein-safflower oil model system. Peroxide Value TBA Value Sample (meq. peroxide/ (Absorbance kg sample) 532 nm) Safflower oil before incubation 1.36 -- Safflower oil ex- tracted from casein- safflower oil model system after incu- bation 10 days 55 C 159 -- Casein-safflower oil model system before incubation -- 0.055 Casein-safflower 011 model system after incubation 10 days 55 C -- 0.133 Casein after extraction of oxidized oil -- 0.064 Untreated casein -- 0.044 39 Protein Efficiency Ratio Determi- natiBn (PER) The growth curves of the 5 groups of rats fed diets containing either untreated or the various treated caseins are presented in Figure 2. Those fed diets containing casein exposed to the oxidizing oil (Exox) grew at the slowest rate. Rats fed either an untreated casein (Untrt) diet, or the diet containing casein which had the unoxidized oil removed (Exunox) grew at approxi- mately the same rate. When the diets with the autoclaved casein (Auto or Auto Glu) were fed, rats grew at a rate slightly less than that obtained when rats were fed the 22355 casein diet. The PER results are presented in Table 4. Although the Exox casein exhibited the only PER which was signifi- cantly lower than the PER of the gntgt casein, the PERs for the auto-samples were lower than the Untrt, suggest- ing that heat treatment had a slight effect. Amino Acid Content of Treated Proteins (ACId HydrOIysates) Table 5 represents the total amino acid content of the 5 casein samples determined after acid hydrolysis for 24 or 48 h. These data represent the higher of the two values (24 or 48 h hydrolysis) for each amino acid. The various treatments had only a slight effect on the amino acid content. Because the precision of the amino 4O .cwommo concouucs Ho powwow» Honufim mcwcwmucoo mumflp pom much Mo :Hmm uzmww3 oomum>¢ .N musmflm w> va.o mam mm.~m Ho.H mm.mm Hm.H hm.mm mm.o mm.vn mm.a mm.mh mm.a mus mm.mv HH.N vo.om Ho.~ m>.ov mm.a mm.mv m~.~ nv.mv vm.m mam Hm.mv mm.m mn.om hm.m mm.mv mm.m ma.om Hm.m Ha.mv hm.~ was Ha.aa ~m.m ha.am Am.a aa.mm ms.m No.3m ma.a H~.Hm 46.6 son mm.ma mm.o Hm.mH mm.o Ho.NH Hw.o mm.ma Hm.o om.mH mn.o mHH mm.ma mN.H mm.m~ mm.H mm.ma HH.H m~.- oa.a mn.a~ mm.a Hm> mh.mv mo.a mv.Hn om.a mn.hm mw.o hm.~m mm.a ~m.~m hm.a um: oo.o oo.o mm.ha no.o hm.ma wo.o mm.ha no.0 oo.o oo.o mwo ~\H mo.ha no.0 nn.ma mw.o m~.HH ~¢.o mv.wa mm.o mm.ha mw.o use om.hv mm.m ma.nm mh.v mm.Hm mm.m hm.mm mm.v mm.om mo.v mhq mlmc muuooao mlav mouooao mlac mouooao mlmv muuooao mlmv mouooao owwmm H.9aw .ous< H.0p54 H.xo .xm H.xoca .xm H.uuucD comm monEMm Gammon poumouucs cam cmumonu mo ummmap aflumouodmmlcflmmom 2H mpwom ocHEm comm .o magma 51 .mEcHnomm Hoomcnocm on one 30H cmo mcoao>v .mmmxmomcxc pmom an cccmEmcucc pmoo osmso Houou mo uncomcm mo pcmmcmmxc mosao>m .umcmmp mo 2 m ma\m mo pcmmcmmxc mcDHo>m .O HNH CHE m ©c>oaoou5o cmooaam mac cmcmmo ..5Hu ovum no HNH cmE m po>cao Iomdo smcmmu ..oms¢ ammo mc3oammom nonwpmxo mo coauocmmxc mcuwo cmcmcu ..xo .xm “Hmo mc3onmom pcnmpmxoco mo comuoommxc momma amcmou ..xoca .xm uuccfimocmu oz ..umucD m om.- ~m.a~ ~s.m~ ~m.am ss.m~ Hmuoa oo.o oo.o -.o oo.o oo.o osommsm 0oz oo.o mm.o oo.o Ha.o om.o oomxoumsm no: om.om os.o as.ma om.o as.o~ om.o m~.oa oe.o ma.o~ oa.o mmz om.a a~.o oo.mm om.o sm.o om.o am.om s~.o mm.om o~.o mam ma.a so.o oa.om om.o no.3 so.o ~o.o m.o ma.m oo.o mac oo.o oo.o oo.o oo.o oo.o oo.o oo.o oo.o oo.o oo.o oum mm.o mm.m no.5 mo.m aa.a ~o.m ao.o am.m ao.o oa.m amo name Numoomo mama Numoomo mlwv mouoomo name Nuuommo mlmv. muuoomo omom ocmfid H.5Hw .oun4 H.0mdd H.xo .xm H.xo:D .xm H.9mucs comm pcscmucou .m canoe 52 approximately the same extent as the 22355 control. However, the release of two nonessential amino acids, glycine and alanine, from the two milder treated pro- teins was greater than from that of the 22555 casein. All of the liberated essential amino acids of the Exgx sample were decreased when compared to those of the Untrt casein. The largest decreases of amino acids were ob- served for threonine, 36%; valine, 29%; sulfur amino acids, 35%; isoleucine, 23%; and tryptophan, 33%. The concentration of released lysine, leucine, tyrosine, and phenylalanine were reduced by less than 20%. With the exception of aspartic acid, all the nonessential amino acids released from the Exgx casein were reduced. Auto- claving the casein-glucose mixture caused a reduction in the release of all essential amino acids, but only in the case of lysine was the reduction greater than that noted for the Exox sample. The hydrolysis of the non- essential amino acids of the A232.§12 sample were also reduced. The release of serine was decreased by 39%; while the liberation of other nonessential amino acids was reduced by less than 20%. Free Amino Acid Indices of Protein Quality Various indices have been calculated from the concentrations of essential amino acids liberated by digestion. The Pepsin Pancreatin Digest Index (PPDI; 53 Akeson and Stahmann, 1964) was calculated (see Appendix for enzymatic digestion of whole egg protein). The PPDI method involves identical digestion of a reference pro- tein, whole egg, and a test protein. The calculation of PPDI includes a comparison of both the essential amino acids released by digestion and those amino acids remain- ing undigested between the reference and test protein. Correction factors are also included in the calculation to account for differences in the extent of proteolysis between the reference and test protein. Because egg protein has been shown to contain an excessive amount of essential amino acids relative to the amino acid requirement of the rat, QEEEE casein was selected as a reference. The PPDI results using the Untgt casein and egg protein as references are shown in Table 7. When egg protein served as the reference, the PPDI of the Untrt, Exunox, and 5252 casein were similar; 77, 78, and 76, respectively. However, the more severely treated proteins, E525 and 5222.§l21 yielded higher values; 80 and 83, respectively. In contrast, when Untrt casein was taken as the reference, the PPDI varied from 91 for the E525 casein to 100 for the Untgt casein. An Enzyme Score (Stahmann and Woldegiorgis, 1975) and Enzyme Index were calculated from the essential amino acids released during digestion. The Enzyme Score, which is calculated similarly to the Protein Score, is the 54 Table 7. Pepsin Pancreatin Digest Index1 of treated and untreated casein. 2 PPDI Treatments Untrt casein Egg Reference Reference Untrt 77 100 Exunox 78 96 Exox 80 91 Auto 76 92 Auto Glu 83 94 1 Akeson and Stahmann (1964). 2Untrt, No treatment; Exunox, Casein after ex- traction of unoxidized safflower oiI; Exox, Casein after extraction of oxidized safflower oil; Auto, Casein auto- claved 5 min 121 C; Auto Glu, Casein and glucose auto- claved 5 min 121 C. -——' 55 smallest ratio of the essential amino acids released from the test protein to that essential amino acid re- leased from the reference protein. The Enzyme Index, derived similarly to the Essential Amino Acid Index, is the geometric mean of all the ratios of essential amino acids released from a test protein to these amino acids released from a reference protein. The Enzyme Score and Enzyme Index for the five proteins are presented in Table 8. When the released essential amino acids of egg protein were used as the reference, the sulfur amino acids were the limiting amino acids in all proteins. However, when the Untrt casein was used as the reference, the limiting amino acids were; valine for Exunox, threo— nine for Exgx, and lysine for 52E2.§l§° For the 5352 sample, the amino acids released were either the same or greater than those released from the Untrt casein. The Enzyme Score was consistently lower than the Enzyme Index. In addition, the values for the Enzyme Index and Enzyme Score, using egg protein as the reference, were lower than when Untrt casein was the reference protein. Initial Rate of Digestion: pH Stat Evaluatians The purpose of this experiment was to investigate the effects of treatments on the initial rates of protein 56 .O Hhhlcme m po>oaoouso cmoooam can cmomoo .SHO ousc no HNH GAE m oc>maoousc mc3 coma cognac c cums pccmamcpcs pmoo ocmsd omoom cahucm cuoasoaco om pom: mo3 ccma pmaom sums pccmHmoccd pmoo ocmE¢ .xOCdxm «ucoEuccmm oz .umuco cmcmmo .ousd ammo mc3oammom ccnmpmxo mo c0muoomuxc momma Gmomoo .xoxm “Hmo mc3on Imam ccnmpmxoco mo comuocmmxc momma :Homco N .coccmcmcm cam mc3 cmcmoo umuco cos: cmoom cahucm on» cuoadoaoo on com: .coccmcmcm cam mo3 amouomm mmc co£3 H mm ooa mm mm com coccmcmom :mcmmo amps: om am on mm mm coscuo cm mom xcpcm c Nam mm com mm mm ooa cocomcmcm cmcmco ummca mm mo ma as so oosououou mom cmoom cahncm Hm.a mm.o mmhn mm.H mm.a oa.a ama hm.o mn.o ~v.o wwnw mm.o om.o mas mm.a mm.a HH.H vv.H mm.a om.H Hm> mo.o mm.o Hm.o Hm.o mn.o om.o cam mm.m hm.v mn.m mv.v v¢.v om.v mood on.” moum nn.~ mm.n .Hpvm om.m use .osm man moqm am.o mm.m a~.m oa.m use «\m .ooz om.~ mh.¢ mm.m mm.w mo.¢ o>.~ can can ous< Quad xoxm xonsxm umucs mom wand Nucceuoome cmcmomm ocmE¢ Hcmcmmo Ucumcmmcs pom couccmu mo xcccH cahucm cam cmoom cahucm can .umommp ammocmocom cmmmcm mo Az m mH\m mo pcmmcmmxcv comummomeoo pmoo ocmec comm .m canoe 57 digestion by either pancreatin or trypsin. In addition to determining initial rates of digestion on intact pro- teins, similar values were also measured on proteins which had been pre-digested with pepsin. The pepsin pretreatment was performed to determine if the protein treatments affected initial pancreatin or trypsin diges- tion rates after the protein had been partially hydrolyzed. The initial rates of pancreatin digestion are shown in Table 9. For the intact protein, the initial rate of the Exgx casein was significantly lower than the rate for the UnE£E_or Exunox samples (P < 0.05). The initial rates for the autoclaved samples were slightly lower than those of the Untrt control; however, the dif- ferences were not significant at the 95% confidence level. The initial rate results for the pepsin-pretreated proteins were different from those obtained for the intact proteins. The initial rates for the autoclaved samples were significantly lower (P < 0.05) than that for the Untrt casein, while the rate for the Exgx sample was essentially the same as the Untrt control. The initial rates of digestion for the Exunox, for both the intact and pepsin-pretreated proteins, were slightly greater than those of the Untrt control. In addition to pancreatin, trypsin was also employed to determine the initial rates of digestion for the five casein samples. The initial rate results 58 Table 9. Initial rates of pancreatin digestion of treated and untreated casein. Initial Rate (ml base/min.) Treatments Intact Proteinl Peptic Dig.1 Untrt. 0.151 t 0.001ac 0.138 : 0.001ab Ex. Unox. 0.171 s 0.064a 0.156 f 0.025a Ex. Ox. 0.107 : 0.020b 0.131 f 0.011bc Auto. 0.135 r 0.017c 0.112 : 0.014c Auto. Glu. 0.144 1 0.014c 0.112 t 0.010c 1 Mean of 5 Replicates 1 SEM. NOTE: Means not showing common superscript are significantly different (P < 0.05). 59 for tryptic hydrolysis appear in Table 10. For the intact proteins, the initial rates were quite similar except for the rate of the Ante 913 casein which was significantly higher (P < 0.05) than the others. The initial rates of the tryptic digestion for the pepsin- pretreated proteins were approximately equal. 60 Table 10. Initial rate of trypsin digestion of untreated and treated casein. Initial Rate (ml base/min.) Treatments Intact Proteinl Peptic Dig.l Untrt. 0.156 1 0.010ab 0.157 1 0.003a Ex. Unox. 0.145 1 0.006a 0.165 1 0.009a Ex. Ox. 0.137 1 0.013a 0.155 1 0.055a Auto. 0.141 1 0.009a 0.160 1 0.000a Auto. Glu. 0.176 1 0.018b 0.150 1 0.023a 1 NOTE: Mean of 5 Replicates 1 SEM. Means not showing a common superscript are significantly different (P < 0.05). DISCUSSION Effects of Protein Treatments Mild Treatments--Autoc1aving and Solvent ExtractIOn of’Unoxidized Oil The two milder casein treatments, solvent extrac- tion of unoxidized oil and autoclaving, caused little change from the Untrt casein in the amino acid composition of either the intact proteins or the enzymatic superna- tants. In addition, the PER of these two treated proteins was not significantly different from the 22235 sample at the 95% confidence level. Although the gel filtration Chromatograms did not indicate any significant differences in the proportion of free amino acids between the treated and untreated proteins, the total in vitro enzymatic release of free amino acids from the Ante and Exunox samples was slightly greater than that of the gntgt casein. This difference in free amino acid release was greater for the Agtg_sample than for the Exunox sample. Some researchers have suggested that mild treatment increases the susceptibility of proteins to enzymatic digestion because it induces a slight denaturation (Osner and Johnson, 1975). However, since the initial 61 62 rate of digestion with either pancreatin or trypsin did not increase after autoclaving, and the rate of digestion monitored by alpha amino nitrogen of Ante casein did not exceed that of the Untrt, the increased release of amino acids might be due to the slight denaturation of casein, caused from autoclaving, which releases some indiginous TCA-soluble material. The increased TCA-soluble alpha amino nitrogen of the 5222.§12 sample compared to the Untrt control at zero time of digestion supports this hypothesis. Casein Exposed to Oxidizing Oil Amino acid analyses of acid hydrolysates of casein which had been exposed to oxidizing oil demonstrated sig- nificant changes from the Untrt casein only with respect to the decreases of sulfur amino acids and tyrosine. Several researchers have found that the major amino acids destroyed after a protein was exposed to oxidizing oil were lysine, arginine, histidine, and methionine (Braddock and Dugan, 1973; Yanagita and Sugano, 1974; Roubal and Tappel, 1967). These authors did not deter- mine tryptophan in the amino acid analyses. In vitro enzymatic digestion of the Exgx casein released a lower amount of amino acids than that released from the Untrt casein. In addition, the protein quality of the Exgx sample as measured by PER was significantly lower (P < 0.05) than the Untrt casein. The similarity 63 in the total amino acid content determined after acid hydrolysis of the §§g§_and Untrt casein, along with the dramatic decrease in amino acids released during enzymatic digestion of the Exgx_sample compared to the 22552, suggest that the decrease in protein quality is due to a decrease in amino acid availability. Probably more than one mechanism was involved in the decrease of availability of the amino acids. As evi- denced by the browning that occurred during oxidation of the oil, the reaction of the carbonyl compounds formed during fatty acid oxidation with amino groups of the protein could have been the major mechanism. The basic amino acids were likely the most affected amino acids. Pokorny et a1. (1973, 1975) studied the browning reaction that occurs in protein-lipid systems, conclud- ing that various aldehydes, formed from hydroperoxide breakdown, were primary substrates for the nonenzymatic browning. Yanagita et a1. (1973) observed a decrease in available lysine which paralleled nonenzymatic brown- ing in a casein-oxidized ethyl linoleate mixture. Thus, this reaction might account for some of the decreases observed in available lysine and arginine. As fats undergo oxidation in a protein-lipid system, the sulfur amino acids are concomitantly oxi- dized (Tannenbaum et al., 1969). The presence of meth- ionine sulfone in the enzymatically released amino acids 64 suggests that the methionine was oxidized in the Exgx casein. The products of sulfur amino acid oxidation do not have the same nutritive potency as methionine and cystine. For example, methionine sulfoxide has only 60% of the biological value of methionine; while meth- ionine sulfone and cysteic acid have no biological activity (Andrews et al., 1976). Cuq et a1. (1973) observed that methionine sulfoxide and sulfone were not enzymatically released from casein after treatment with H202, indicating that the peptide bonds involving oxidation products of methionine are resistant to enzy- matic hydrolysis. It was suggested that the free radicals produced during lipid oxidation react with proteins (Karel, 1973). Certain aromatic amino acids, such as tryptophan and histidine, have been demonstrated to possess antioxida- tive properties (Mitsuda, 1967). It was suggested that amino acids act as antioxidants by reacting with peroxide radicals to form inactive peroxyanions. Such free radical attack of reactive aromatic amino acid residues could therefore induce losses of protein nutritional quality. This reaction could possible explain the 12% decrease of tryptophan in the acid hydrolysate of the Exgx sample. In addition to a reduction in the enzymatic release of amino acids, the overall rate of digestion as monitored by the alpha amino nitrogen content of the TCA supernatants of Exox casein was slower than that of 65 the Untrt sample. A decreased initial rate of pancreatin digestion of the intact Exgx_casein was also noted. These results are in agreement with those of Yanagita et a1. (1973) who demonstrated a decrease in in vitro digesti- bility with pepsin and trypsin when casein and ethyl linoleate were incubated until browning occurred. The decrease in digestion rate could be due to the modifi- cation of side chains of amino acid residues normally susceptible to enzymatic cleavage, such as those involv- ing carbonyl compounds reacting with amino groups. The increased TBA values of E325 sample indicate that the dialdehyde compound, malonaldehyde, was present after the oxidized oil was removed. In addition as previously discussed, the oxidation products of sulfur amino acids might contribute to the inhibition of enzymatic hydroly- sis. Therefore, the combination of blocked amino acids by carbonyl compounds and the presence of oxidation pro- ducts of the sulfur amino acids could explain partially the reduced digestion rate. Casein Autoclaved with Glucose Although the PER of the 5239 Gig casein was not significantly lower than that of the Untrt casein (P < 0.05), the amino acid profile of the acid and enzymatic hydrolysates was altered. The lysine and sulfur amino acids of the acid hydrolysate of Agtg_§12 casein were lower than those of the Untrt hydrolysate. All other 66 amino acids in the acid hydrolysate of the Agtg4§lg appeared to be unaffected by the treatment. However, all the amino acids released from the A2£2_§lu_casein during enzymatic digestion were reduced relative to those liberated from the Untrt sample. Autoclaving a casein-glucose mixture results in a series of reactions known as nonenzymatic browning or the Maillard Reaction. These reactions have been studied extensively (Adrian, 1974; Bender, 1972; Feeney et al., 1975). These authors have found that the basic amino acids, the sulfur amino acids, and tryptophan were the most reactive amino acids in the Maillard Reaction. Rao and McLaughlin (1967) found that in a casein-glucose mixture, the available lysine decreased faster than the available methionine. These results support the data of the present study. The lysine released after in vitro digestion was reduced 27%, while the sulfur amino acids were reduced 16%. The total amount of amino acids released from Auto Glu casein during in vitro enzymatic digestion was less than that released from the Untrt sample; 22.50 g/16 g N versus 25.77 g/16 g N. The other experiments which examined the rate and extent of hydrolysis during enzymatic digestion were not sufficiently sensitive to detect the difference in the amount of amino acids released. The extent of hydrolysis of the 5222.912. was not significantly different from that of Untrt 67 casein at the 95% confidence level. When the initial rates of digestion were investigated, only in the case of the pancreatin initial rate of the pepsin-pretreated Auto Glu sample was there a significant decrease (P < 0.05) from the rate of the Untrt casein. Therefore, only by examining the released amino acids could this difference be detected. Protein Quality Indices No significant differences were observed in the Sephadex G-25 Chromatograms of the TCA-soluble material released by in vitro enzymatic digestion of the test pro- teins. Therefore, the molecular size distribution of the digestion products could not be employed as a quali- tative indicator of protein quality. Although slight differences between the initial rates of the treated and untreated samples were observed, the most sensitive indicator of protein quality was derived from analysis of the amino acids released during in vitro digestion. To quantitate the differences in nutritional quality among the proteins, various quality indices involving the essential amino acids liberated during in vitro enzymatic digestion were used. A com- parison of these indices with PER results is presented in Table 11. Linear correlation coefficients were calcu- lated to provide a criterion for comparison of the 68 .cocomcmcm can no cows mo3 :Hcmoo pcuocmucs mo GHcmomm moo mccuHMm .coccmcmcm can no cHomoo pcumommcs mo :chomm moo mocuHc mchs AmhmH .mHmmOHmcUHoz cam ccofincmmv cmoom cahucmv .AvOmH .cccanouw paw coucxdv cocomcmcm on» no cHommo pcmccmuc: mo chmomm mmc mccuHc OGHms xcch mmcmHa cHuocmoccmchmmcm m .mmmH omB\oomm OCHms pcmmHSOHoo xcch UHU< OCHE< HcHuccmmMN .Aman .om3\o«mv cmoom :chommH O0.0 m0.0 mm.O O0.0 m>.O NO.OI hh.O O0.0 ucoHOHmmooo COHmchmmoO mm om mm mm mm mm mm HO HO mh.m .SHO .ous¢ OOH om OOH mm mm mm mm Hm mm mm.m .omsd mp Oh no mo Hm om mm mm mm Om.~ .xo .xm mm mm mm Hm mm Oh hm HO «OH vH.m .xocD .xm OOH mm OOH no OOH nu pm Oh OOH 50.m .umucs .ummco mmm .umuca omm .ummcn mom cmoom mmm xc a cause cmoo cahuc ~H< CH can .AH<¢mO xcch OH0¢ OCHE¢ Honccmmm .cmoom chuomm nuH3 muHSmcm mmm mo GOmHmcmEou .HH cHQoB 69 correlation of these in vitro indices of nutritional quality with bioassay results (PER). The Protein Score and Essential Amino Acid Index did not reflect the nutritional quality in the treated proteins as determined by the PER method. Because the Protein Score and Essential Amino Acid Index are based on the total rather than the biologically available amino acids, these methods are not accurate indicators of the nutritional quality of the proteins. The results demonstrate that, in proteins in which amino acid availability is the predominant limit- ing factor in the nutritional quality of the protein, an in vitro enzymatic digestion provides a better indication of the protein quality than the total amino acid indices. The PPD Index (Akeson and Stahmann, 1964), which utilizes amino acid data from an in vitro digestion procedure, was calculated for the five casein samples. For the two most severely treated proteins, E525 and 5232 Gig, the PPD Index was observed to be greater than that of the Untrt casein, contrary to results obtained by the PER method. These results were supported by those of Stahmann and Woldegiorgis (1975), demonstrating that the PPD Index overestimates the quality of moderately damaged proteins. Since egg protein contains an excess of certain amino acids relative to the amino acid requirement of the rat, Untrt casein was tested as a 70 reference protein for the in vitro nutritional quality estimates. Even when Untrt casein was the reference, the PPD Index overestimated the protein quality of Exgx and ABE2_§12_samples. The method of calculation employed in the PPD Index determination presumably is responsible for the overestimation of protein quality. The major problem in the PPD Index calculation concerns the use of a factor which corrects for differences in the extent of proteoly- sis during in vitro digestion between the test and ref- erence protein. Using this correction factor, one assumes that the amino acids of the test protein are released to the same extent as those of the reference protein. In the case of the §§g§_and A252 G12 samples, which have undergone chemical modification, some of the bonds which are normally susceptible to digestive enzymes are resis- tant to enzymatic attack. Therefore, because the amount of amino acids released from these moderately processed proteins will not be equal to the amount released from the reference protein, the use of the correction factor is not valid. A reduction in the PPD Index is only observed when there is a large decrease in the initial release of an amino acid, such as in the case of severely processed samples. Another index of protein quality, Enzyme Score (Stahmann and Woldegiorgis, 1975), which is based only 71 on those essential amino acids released during in vitro digestion, was tested as an alternate indication of pro- tein quality. By using a reference of amino acids released from either egg protein or Untrt casein, this evaluation indicated that protein quality was lower than the relative protein quality as determined by bioassay (PER). However, the UnE££_casein reference did provide a better correlation with PER than was observed with the egg protein reference. A limiting feature of the Enzyme Score is the fact that the concentration of the most limiting released amino acid used for calculation might be lower than that found if digestion were complete. In the Exgx casein, the initial rate and overall digestion were lower than those of the Untrt sample. Therefore, at the end of the 27 h digestion period, fewer amino acids were released from the Exgx. However, in vivo, the length of the intestine allows for more of the slower digested material to be absorbed (Rerat, 1976). Thus, the value for the limiting amino acid after 27 h of digestion is probably lower than it would have been if digestion were complete. Another expression of protein quality, Enzyme Index, was calculated by computing the geometric mean of all the ratios of the released essential amino acids of the test protein to those released in the reference pro- tein. By taking a weighted mean of the ratios, instead 72 of using only the smallest ratio as in the Enzyme Score, potential biases associated with incomplete digestion were minimized. The values for the Enzyme Index, using either the Untrt casein or egg protein as the reference, gave the best correlation to the PER results of the treated proteins. CONCLUSION The difference in the nutritional quality of the four treated caseins was reflected in their varying PERs. Various aspects of an in vitro enzymatic digestion were evaluated as indices of protein quality by comparing them with the PER results for the casein samples. Although slight differences in the rates of the in vitro digestion were observed, the best indicator of protein quality was the profile of amino acids released during pepsin-pan- creatin sequential digestion. Various indices using the essential amino acids released during digestion were calculated. When the untreated casein was used as the reference protein, Enzyme Score and Enzyme Index cor- related best with the bioassay results. Although further investigation is needed to verify the correlation between the Enzyme Score or Enzyme Index and PER, the in vitro enzymatic digestion appears to be a useful method for protein quality determination. The in vitro digestion procedure is relatively rapid and provides information concerning amino acid availa- bility. 73 APPENDIX APPENDIX Preparation of Ninhydrin Solution l. 0.2 M citrate buffer (pH 5.0) was prepared using 4.3 g citic acid and 8.7 g Na3citrate'2H20 in 250 ml water. 2. To the above solution 400 mg of reagent grade SnC12°2 H20 was added. 3. 250 ml of methyl cellusolve containing 10 g dissolved ninhydrin was added to the citrate buffer containing the SnClz. This solution was aged two days before using and was prepared fresh every 7-8 days. Reagents for the TBA Test Aldehyde-free redistilled absolute ethanol--was prepared by dissolving 1.5 g AgNO3 in 3 ml distilled water and mixing with l l of absolute ethanol. Three grams KOH were dissolved in 15 m1 of warm absolute ethanol. When cool, the two solutions were mixed thoroughly. The resulting precipitate was allowed to settle, then filtered. The filtrate was distilled and stored in an amber glass container. 74 75 TBA reagent--was prepared by transferring 1.4 g TBA (East- man Organic Chemicals) into a volumetric flask and dilut- ing to 100 ml with aldehyde—free redistilled absolute ethanol. The mixture was sonicated l h, and then fil- tered through Whatman no. 41 filter paper. The reagent was prepared immediately before use. Moisture and Nitrogen Content of Casein Samples Used in PER Diets - Percentage . Case1n Treatment Nitrogen'7%) M01sture (%) No treatment 14.5 7.11 Casein after exposure to unoxidized oil 14.3 10.00 Casein after exposure to oxidized oil 12.4 6.97 Autoclaved 5 min. 121 C 15.1 4.84 Casein glucose mixture autoclaved 5 min. 121 C 8.24 5.36 Enzymatic Digestion of the Dried Whole Egg Solid SampIel Total2 Digest3 Lys 7.1 2.7 His 2.8 0.8 Arg 6.6 3.1 Asp 9.8 0.2 Thr 4.6 0.64 Ser 7.4 0.54 Glu 12.6 0.5 (continued) Total2 Digest3 Pro 4.7 0.00 Gly 3.1 2.8 Ala 5.5 5.0 1/2 Cys 2.3 1.9 Val 6.1 6.8 Met 3.3 2.4 Ile 8.2 9.0 Leu 8.2 9.0 Tyr 4.0 4.1 Phe 4.8 5.3 Tyr 1.9 1.9 Total 97.4 25.2 1 Values are 2Values are 16 g N of protein. 3Values are digest. 4Values are from Dimler (1975). expressed as g amino acid residue/ expressed as g amino acid/16 g N of estimated from Dimler (1975). 77 Calculation of In Vitro Digestion Indices l. Pepsin-Pancreatin-Digest-Index A. The total essential amino acids of the egg and test protein are listed along with the essential amino acids which were released during digestion (digest) and those remaining undigested (residue). B. The quantity of each digested amino acid was expressed as a percentage of the total digested amino acids for both proteins. The same pro- cedure was followed for the residue amino acids for both proteins. C. Ratios of each amino acid of the test protein to the egg protein expressed as a percentage were calculated for both the digest and residue fractions. These ratios are referred to as egg ratios. D. Geometric means for both egg ratio fractions were calculated. These mean values were then cor- rected for the degree of proteolysis of the test protein relative to the egg protein. These cor- rected means are weighted in accordance with the percentage each represents of the total egg protein, and averaged geometrically to obtain PPDI. 78 2. Enzyme Score Enzyme score is calculated like the Protein Score. It is the smallest ratio of the essential amino acid released during digestion of the test protein to that amino acid released from the reference protein. 3. Enzyme Index Enzyme Index is the geometric mean of the ratios of the essential amino acids released during digestion of the test protein to those amino acids released from the reference protein. SELECTED BIBLIOGRAPHY SELECTED BIBLIOGRAPHY Adrian, J. 1974. Nutritional and physiological conse- quences of the Maillard reaction. World Review of Nutrition and Dietetics. 12:71. Akeson, W., and M. A. Stahmann. 1964. A pepsin pancrea- tin digest index of protein quality evaluation. J. of Nutr. 83:257. Anderson, G. H., D. V. M. Ashley, and J. D. Jones. 1976. Utilization of L-methionine sulfoxide, L-methio- nine sulfone, and cysteic acid by weanling rat. J. of Nutr. 126:1108. Andrews, F., J. Bjorksten, F. B. Trenk, A. S. Henick, and R. B. Rock. 1965. The reaction of an autoxi- dized lipid with protein. Am. Oil Chem. Soc. J. 42:779. Association of Official Analytical Chemists. 1975. Official Methods of Analysis. Ed. W. Horwitz. AOAC, Washington, D.C. Baldwin, R. R., J. R. Lowry, and R. Thiessen Jr. 1951. Some effects of processing of the nutritive pro- perties of proteins. Food Research. l§:107. Bender, A. E. 1972. Processing damage to protein food; A review. J. Food Tech. 1:239. Bender, A. E., and B. H. Doell. 1957. Biological evalu- ation of proteins. A new aspect. Br. J. of Nutr. 11:140. Bender, A. E., and D. S. Miller. 1953. A new brief method of estimating net protein value. Biochem. J. Egzvii. Bjarnason, J., and K. J. Carpenter. 1970. Mechanisms of heat damage in proteins. 2. Chemical changes in pure proteins. Br. J. of Nutr. 24g313. 79 80 Block, R. J., and H. H. Mitchell. 1946. The correlation of the amino acid composition of proteins with their nutritive value. Nutr. Rev. 33:249. Bodwell, C. E. 1977. Problems in the development and application of rapid methods of assessing protein quality. Food Tech. 33:73. Braddock, R. J., and L. R. Dugan. 1973. Reaction of autoxidizing linoleate with Coho Salmon. Am. Oil Chem. Soc. J. 33:343. Buttkus, H. 1967. The reaction of myosin with malonalde- hyde. J. Food Sci. 33:432. Canolty, N. L., and L. J. Koong. 1977. Evaluating pro- tein quality by estimating weight gain coeffi- cients. J. Food Sci. 23:1130. Carpenter, K. J. (with Booth, V. H.). 1973. Damage to lysine in food processing: Its measurement and its significance. Nutr. Abst. Rev. 33:424. Carpenter, K. J., C. H. Lea, and L. J. Parr. 1963. Chemical and nutritional changes in stored herring meal 4. Nutritional significance of oxidation of the oil. Br. J. of Nutr. 31:151. Chavez, J. F., and P. L. Pellett. 1976. Protein quality of some representative Latin American diets by rat bioassay. J. of Nutr. 106:792. Chio, S., and A. L. Tappel. 1969. Inactivation of ribo- nuclease and other enzymes by peroxidizing lipids and malonaldehyde. Biochem. 3:2827. Clark, J. M. Jr. 1964. Protein and amino acids. In: .Experimental Biochemistry, lst. ed. W. W. Free- man and Co. San Franc1sco. Code of Federal Regulations. 1977. Nutritional labeling of foods. New Section 21 C. FR 101.9. Crawford, D. L., T. C. Yu, and R. O. Sinnhuber. 1967. Reaction of malonaldehyde with protein. J. Food Sci. 33:332. Cuq, J. L., M. Provansal, F. Guilleux, and C. Cheftel. 1973. Oxidation of methionine residues of casein by hydrogen peroxide. Effects of in vitro diges- tion. J. of Food Sci. 33:11. 81 Dimler, S. R. 1975. In vitro enzymatic digestion of milk proteins. M.S. Thesis. Michigan State University, East Lansing. Evans, R. J., and H. A. Butts. 1949. Studies on the heat inactivation of methionine in soybean oil meal. J. Biol. Chem. 178:543. FAO/WHO Ad Hoc Expert Committee. 1973. Energy and pro- tein requirements. FAO Nutritional Meeting Report Series. No. 52. FAQ/WHO, Rome. Feeney, R. E., G. Blankenhorn, and H. B. Dixon. 1975. Carbonyl-amine reactions in protein chemistry. Adv. Protein Che. 33:135. Finot, P. A. 1973. Nonenzymatic browning. In: Proteins in Humgn Nutrition, Ed. J. W. G. Porter, an B. A. Rolls. Academic Press, New York. Ford, J. E. 1973. Some effects of processing on nutri- tive value. In: Proteins in Human Nutrition, Ed. J. W. G. Porter, and B. A. R0115. Academic Press, New York. Ford, J. E., and D. N. Salter. 1966. Analysis of enzyma- tically digested food proteins by Sephadex-gel filtration. Br. J. of Nutr. 33:843. Ford, J. E., and C. Shorrock. 1971. Metabolism of heat damaged proteins in the rat: Influence of heat damage on the excretion of amino acids and pep- tides in the urine. Br. J. of Nutr. 33:311. Hackler, L. R. 1977. Methods of measuring protein quality: a review of bioassay procedures. Cer. Chem. 33:984. Hegarty, P. V. J. 1975. Some biological considerations in the nutritional evaluation of foods. Food Tech. 33:52. Hegsted, D. M. 1974. Assessment of protein quality. In: Improvement of Protein Nutriture. Ed. Committee on Amino Acids, Food and NutritIOn Board. Academy of Science, Washington, D.C. Hegsted, D. M. 1977. Protein quality and its determi- nation. In: Food Proteins, Ed. J. Whitaker and S. R. Tannenbaum. Avi, Westport, Conn. 82 Hopkins, D. T., and F. H. Steinke. 1976. Effect of water of hydration on the measurement of the protein efficiency ratio of casein and soybean protein with rats. J. of Nutr. 333:1438. Horigome, T., and M. Miura. 1974. Interaction of protein and oxidized ethyl linoleate in the dry state and the nutritive value of the reacted protein. J. Agr. Chem. Soc. Japan 33:437. Horigome, T., T. Yanagida, and M. Miura. 1974. Nutritive value of proteins prepared by the reaction with oxidized ethyl linoleate in aqueous medium. J. Agr. Che. Soc. Japan 33:195. Hurrell, R. F., and K. J. Carpenter. 1976. Mechanism of heat damage in proteins. 7. Significance of lysine containing isopeptides and of lanthionine in heated proteins. Br. J. of Nutr. 33:383. Hurt, H. D., R. H. Forsythe, and C. H. Krieger. 1975. Factors which influence the biological evaluation of protein quality by the protein efficiency ratio method. In: Eggtein Nutritional Qualit of Food and Feeds. Part I AssaeretHOds—Bio- IegicaI, BIOEhefiICal, afid‘Chemical,Ed. M. Fried- man. Marcel DeEker Inc. New York. Janicek, G. 1973. Maillard's and related reactions and nutritive value of foodstuffs. Scripta Medica 46:107. Kaba, H., and P. L. Pellett. 1975. Prediction of the true limiting amino acids using available pro- tein scoring systems. Ecol. of Food and Nutr. 4:109. Karel, M. 1973. Symposium: Protein interactions in bio- systems. Protein-lipid interaction. J. Food Sci. 33:756. King, R. L. 1962. Oxidation of milk fat globule mem- brane material. I The thiobarbituric acid test as a measure of oxidized flavor in milk and model system. J. Dairy Sci. 33:1165. Lewis, 0. A. M. 1966. Short ion-exchange column method for the estimation of cystine and methionine. Nature 209:1239. 83 Lewis, S. E., and E. D. Wills. 1962. The destruction of -SH groups of proteins and amino acids by peroxides of unsaturated fatty acids. Biochem. Phar. 33:901. Mauron, J. 1970. Nutritional evaluation of proteins by enzymatic methods. In: Evaluation of Novel Erotein Products, Ed. A. ET Bender, R. KifilBerg, B. Lofqvist, aid L. Munck. Pergamon Press, New York. Mauron, J., F. Mottu, E. Bujard, and R. H. Egli. 1955. The availability of lysine, methionine, and tryp- tophan in condensed milk and milk powder. In vitro digestion studies. Arch. Biochem. Biophys. 59:433. McLaughlan, J. M. 1972. Effects of protein quality and quantity on protein utilization. In: Newer Methods of Nutritional Biochemistry Vol. 5. Ed. A. A. Albanese. Academic Press, New York. McLaughlan, J. M., and M. 0. Keith. 1975. Bioassays for protein quality. In: Protein Nutritional Quality of Foods and Feeds. Part I Assay Methods--Bio- IEgICaITBiocgemIEa33afidChemicaI} Ed. M. Fried- man. Marcel Dekker Inc., New Yofk. Melnick, D., B. Osner, and S. Weiss. 1946. Rate of enzymatic digestion of proteins as a factor in nutrition. Science 103:326. Mitchell, H. H. 1924. A method of determining the bio- logical value of protein. J. Biol. Chem. ‘33:873. Mitusuda, H., K. Yasumoto, and K. Iwami. 1967. Mechanism of antioxidation of amino acids in lipid oxida- tion. Memoirs of the College of Agr. Kyoto Univer. No. 92. (Chemical Series no. 32). Moore, 8., D. H. Spackman, and W. H. Stein. 1958. Chro- matography of amino acids on sulfonated polysty- rene resin. Anal. Chem. 33:1185. Morrison, A. B., and J. A. Campbell. 1960. Evaluation of protein in foods. V Factors influencing the protein efficiency ratio in foods. J. of Nutr. 70:112. Nesheim, R. O. 1977. Food industry requirements for rapid methods of protein quality assessment. Food Tech. 33:71. 84 Neter, J., and W. Wasserman. 1974. Applied Linear Sta- tistical Models. Richard D. IrwIfi, Inc., Home- wood, IlI. Osborne, T. B., L. B. Mendel, and E. L. Ferry. 1919. A method of expressing numerically the growth-pro- moting value of proteins. J. Biol. Chem. 33:223. Osner, B. L. 1951. Method for integrating essential amino acid content in the nutritional evaluation of protein. J. Am. Dietetics Assoc. 31:396. Osner, R. C., and R. M. Johnson. 1968. Nutritional changes in proteins during heat processing. J. Food Tech. 3:81. Osner, R. C., and R. M. 1974. Nutritional and chemical changes in heated casein. I Preliminary study of solubility, gel filtration pattern, and amino acid pattern. J. Food Tech. 3:301. Osner, R. C., and R. M. Johnson. 1975. Nutritional and chemical changes in heated casein. II Net protein utilization, peptic digestibility and available amino acid content. J. Food Tech. 33:133. Pharmacia Fine Chemicals. 1974. Sephadex Gel Filtration in Theory and Practise. Pharmacia Fine Chemicals A B, Uppsala, Sweden. Pokorny, J., B. A. El-Zeany, and G. Janicek. 1973. Non- enzymatic browning. VI Browning produced by oxi- dized polyunsaturated lipids on storage with protein in presence of water. Z. Lebensm. Unters-Forsh. 333:323. Pokorny, J., A. Kolakowska, B. A. El-Zeany, and G. Janicek. 1975. Nonenzymatic browning. XI Effect of free amino groups on browning reactions in lipid-pro- tein mixtures/Z. Lebensm. Unters-Forsch. 331:323. Rao, M. N., and J. M. McLaughlan. 1967. Lysine and meth- ionine availability in heated casein-glucose mix- tures. J. of AOAC. 33:704. Rerat, A., T. Corring, and J. P. Laplace. 1976. Protein digestion and Absorption. In: Protein Metabolism and Nutrition, Ed. D. J. A. Cole, K. N. Boorman, P. J. Buttery, D. Lewis, R. J. Neale, H. Swan. Butterworths, Boston. 85 Roubal, W. T., and A. L. Tappel. 1966. Damage to pro— teins, enzymes, and amino acids by peroxidizing lipids. Arch. Biochem. BiOphys. 113:5. Said, A. K., and D. M. Hegsted. 1969. Evaluation of dietary protein quality in adult rats. J. of Nutr. 33:474. Satterlee, L. D., J. G. Kendrick, and G. A. Miller. 1977. Rapid in vitro assays for estimating protein quality. Food Tech. 33:78. Schram, E., S. Moore, and E. J. Bigwood. 1954. Chromato- graphic determination of cystine as cysteic acid. Biochem. J. 31:33. Sheffner, A. L. 1967. In vitro protein evaluation. In: gewer Methods of Nutritional Biochemistry Vol III, Ed. A. A. Albanese. Academic Press, New York. Sheffner, A. L., G. A. Eckfelt, and H. Spector. 1956. The pepsin-digest-residue (PDR) amino acid index of net protein utilization. J. of Nutr. 33:105. Spies, J. R. 1967. Determination of tryptophan in pro- teins. Anal. Chem. 33:1412. Stahmann, M. A., and G. Woldegiorgis. 1975. Enzymatic methods for protein quality determination. In: Protein Nutritional Quality of Foods and Feeds. Piit I Assa ‘Methods-:Biological, BiochemicaIT and Cfiemicai, E3. M. Friedman. Marcel Dekker, New Yor . Steinke, F. H. 1977. Protein efficiency ratio: pitfalls and causes of variability; A review. Cer. Chem. 54:949. Tanaka, M., T. C. Lee, and C. O. Chichester. 1975. Nutritional consequences of the Maillard reaction. The absorption of fructose-L-tryptophan in the large intestine of the rat. J. of Nutr. 333:989. Tannenbaum, S. R., H. Earth, and J. P. LeRoux. 1969. Loss of methionine in casein during storage with autoxidizing methyl linoleate. Agr. Food Chem. ‘33:1353. Tu, A., and N. A. M. Eskin. 1973. The inhibitory effect of reducing sugars on the hydrolysis of casein by trypsin. Can. Inst. Food Sci. Technol. 3:50. 86 Waibel, P. E., and K. J. Carpenter. 1972. Mechanism of heat damage in proteins. 3. Studies with e(a-L-glutamyl)-L-lysine. Br. J. of Nutr. 31:509. Williams, H. H., A. E. Harper, D. M. Hegsted, and G. Arro- yave. 1974. Nitrogen and amino acid requirements. In: Improvement of Protein Nutriture, Ed. Com- mittee on Amino Acids, Food'and Nutrition Board. Academy of Science, Washington, D.C. Yanagita, T., and M. Sugano. 1973. Effect of casein reacted with oxidized ethyl linoleate on biologi- cal value and liver components in rat. (Effect of oxidized fat on nutritive value of proteins Part II). J. Agr. Chem. Soc. Japan 33:577. Yanagita, T., and M. Sugano. 1974. Changes in biological value and available lysine of egg albumin accom- panied with oxidation of lipid. Nutr. and Food. 27:275. Yanagita, T., M. Sugano, S. Cho, and M. Wada. 1973. Changes in available lysine and in vitro diges- tibility of casein accompanied with oxidation of ethyl linoleate (Effect oxidized fats on nutritive value of proteins Part I). J. of Agr. Chem. Soc. Japan 31:73.