ABSTRACT BIOLOGICAL VALUE OF PEA PROTEINS AS INFLUENCED BY GENETIC VARIATION BY Satinder Bajaj The biological value of 32 varieties of peas (EEEEE. sativum L.) grown under similar field conditions was deter- mined using weanling rats by means of net protein utiliza- tion (NPU) and protein efficiency (PER). Different varieties of peas, when fed as the sole source of protein at a 10% level in an otherwise adequate diet, varied from 0.18-0.78, in comparison to casein, in their ability to support growth and nitrogen retention in 3-week-old Sprague Dawley rats. From the analysis of rat growth curves, the pea varieties were separated into those that produced fairly good growth and those that barely maintained the initial weight of rats. Carcass protein expressed as a percentage of body weight was higher in pea fed (20-22%) than in casein fed rats (18-19%). This was probably associated with a dif- ference in the body fat content. PER and NPU were fairly constant when the experiments were repeated, indicating Satinder Bajaj the value of these biological techniques in evaluating protein quality. The total nitrogen content of different varieties of peas provides a poor indication of the biolog- ical value of their nitrogen containing compounds. Nitrogen distribution in protein fractions was studied in varieties of peas selected to represent high, low, and mediocre biological values. Peas with a higher PER contained more albumin nitrogen than those of lower PER. For data from 21 varieties of peas, a quadratic func- tion of the albumin content had a high coefficient of cor- relation (R = 0.949) with PER. The equation: Y = 24.7x - 13.6x2 - 8.8 provides a quick, simple and accurate method for evaluating protein quality of peas. The nitrogen in the other frac- tions did not correlate with the protein quality. The protein quality of the peas was evaluated by a microbiological method using Streptococcus zymogenes. The microbiological values were expressed as a percentage of response produced by casein. These values of pea meals were similar to the PER and NPU when expressed as a per- centage of casein PER and NPU. When the major extractable protein fractions were assayed by the microbiological method, albumin had a higher biological value (117-128) than either globulin (64-100) or the residual fraction (28- 56). The biological value of the albumin fraction from different varieties of peas was not significantly different Satinder Bajaj (P<0.05), whereas differences in the values of the globulin and the residual fraction were significant. The extractable protein from all varieties of peas, when subjected to starch gel electrophoresis was represented by 3-4 major bands. The two slower moving bands were asso- ciated with the globulin and the others with the albumin fraction. In the pea extract the combined globulins moved as a single streak with two bands faintly discernible. The addition of reducing agents (7M urea and Z-Mercaptoethanol) dissociated the globulin (3-bands vicilin; 4-bands legumin) but not the major albumin band. The study demonstrates that, total nitrogen is a poor index of protein quality of peas, both quantity and quality of peas vary in different varieties, and that the quantity of albumin correlates highly with the PER of the peas, an observation which can be utilized to predict the PER of peas. The albumin fraction which contains one major band is of better quality than the other protein fractions studied in the pea extract. BIOLOGICAL VALUE OF PEA PROTEINS AS INFLUENCED BY GENETIC VARIATION BY Satinder Bajaj A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Foods and Nutrition 1969 ACKNOWLEDGMENTS The author expresses her gratitude to Dr. Olaf Mickelsen for making her stay at the Michigan State Univer- sity a learning experience. His encouragement towards independent research, and guidance during the preparation of this dissertation are greatly appreciated. Sincere thanks are expressed to Dr. Deran Markarian for initiation into this project and to Dr. Larry R. Baker for his continued interest. Extreme gratitude is expressed to Dr. Hans A. Lillevik for providing laboratory facilities and guidance so essential for the work on proteins. The author expresses thanks to the members of her committee: Dr. Dena C. Cederquist, Dr. Stanley K. Ries, Dr. Hans A. Lillevik, Dr. Joseph E. Varner, Dr. Larry R. Baker and Dr. Olaf Mickelsen, for discussions, comments and criticisms, and a liberal use of their time. The author wishes to acknowledge many kindnesses extended to her by professors from various departments, especially Dr. John L. Gill for discussions on the statis- tics of this work. A special thanks to Dr. Yashpal S. Bajaj for his advice, criticisms, and comments during the preparation of the manuscript. ii TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . . . . LIST OF FIGURES AND PLATES . . . . . . . . INTRODUCTION . . . . . . . . . . . . . . . REVIEW OF LITERATURE . . . . . . . . . . . Proteins in peas and other legumes . . . Location of pea globulins . . . . . Extraction procedures and homogeneity of pea proteins . . . . . . . . . . . . . Physical properties of pea globulins Chemical composition of pea proteins Residue proteins . . . . . . . . . Non-protein nitrogen of pea seeds . . Phylogenetic relationship of globulins seeds . . . . . . . . . . . . . Physiological functions of pea proteins Function of the globulins . . . . . . in legume Quantitative variation in protein as influenced by genetic variation . . . . . . . . . . . . . Biological value . . . . . . . . . . . . Factors affecting the nutritive value of plant proteins . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . PART I. BIOLOGICAL VALUE OF THE PROTEIN IN DIFFERENT VARIETIES OF PEAS Abstract . . . . . . . . . Introduction . . . . . . . Experimental . . . . . . . Results and Discussion . . References . . . . . . . . iii Page vi com 11 12 14 15 16 16 17 18 21 25 29 36 37 38 40 47 PART II. NITROGEN DISTRIBUTION AND BIOLOGICAL VALUE OF PROTEIN FRACTIONS IN DIFFERENT VARIETIES OF PEAS Abstract . . . . . . . Introduction . . . . . Materials and Method Microbiological Method Results and Discussion References . . . . . . PART III. PREDICTION OF PROTEIN QUALITY OF PEAS FROM ALBUMIN CONTENT Abstract . . . . . . . . . Introduction . . . . . . . Experimental . . . . . . . Results and Discussion . . References . . . . . . . . PART IV. A STARCH GEL ELECTROPHORETIC ANALYSIS OF THE NUTRITIONALLY SIGNIFICANT AND SALT-EXTRACTABLE OF PEAS Abstract . . . . . . . . . . . . . . . . . . . . . Introduction . . . . . . Experimental . . . . . . Materials and Reagents Procedure . . . . . . Results and Discussion . References . . . . . . . SUMMARY AND CONCLUSIONS . . . . . . . . . . . . . . APPENDIX A. Amino Acid Contents of Selected Varieties of Peas O O O O O O O O O O O O O O O O I O O I O O B. Moisture Content of Peas . . . . . . . . . . . . iv Page 48 49 52 57 58 65 67 67 68 71 75 76 76 78 78 79 80 92 94 98 103 Table 1. 10. 11. LIST OF TABLES Page Composition of legumin, vicilin and legumelin according to Osborne and Campbell (1896) . . 13 Amino acid content (% of weight) of some seed proteins . . . . . . . . . . . . . . . . . . 14 Percent protein in peas, NPU, PER, percent protein in carcass and the weight gain on pea diets in comparison to casein diet . . . 41 Variation in PER and NPU between experiments . 44 Nitrogen distribution in pea proteins - % pea meal 0 O O O O I O I I O O O O O I O O O I O 59 A comparison of the biological values of pea meals as estimated by the microbiological assay and the values obtained by rat assay . 61 Microbiological assay values of protein fractions isolated from pea meals . . . . . . 63 Effect of extraction time and procedure on the albumin nitrogen extracted . . . . . . . . . 69 Protein efficiency ratio and albumin content of pea varieties . . . . . . . . . . . . . . 72 Amino acid composition of pea meals g/lOOg nitrogen O I O O O O O O O I I O O O O O O O 101 Moisture content of peas . . . . . . . . . . . 103 LIST OF FIGURES Figure 1. Rat growth curves, on pea diets in comparison to casein . . . . . . . . . . . . . . . . . 2. Scheme for fractionation of pea proteins . . 3. Relation of the albumin content to the microbiological values of pea meals . . . . 4. Starch gel electrophoresis of pea proteins with no dissociating agents . . . . . . . . 5. Starch gel electrophoresis of pea albumins in continuous buffer . . . . . . . . . . . 6. Starch gel electrophoresis of pea proteins with 7M urea and 2-Mercaptoethanol . . . . Plates 13-6 Starch gel electrophoretograms of salt- extractable proteins from different varieties of peas under varying conditions of pH, dissociating and reducing agents . . vi Page 46 54 62 81 83. 85 86-91 INTRODUCTION Plant geneticists have for a long time bred plants to alter specific characters such as color, size, shape, disease resistance and yield. Biologists studying plants and animals have shown that specific proteins are charac- teristic of certain families (Dalby and Lillevik, 1969; Boulter and Thruman, 1968). The nutritional implications of this type of work are obvious, namely that it is pos- sible that quantitative as well as qualitative variation in protein occurs with genetic variation, and can be fur- ther enhanced by making suitable crosses. With the increasing world population and the en— suning food shortage which threatens to affect even the ecxanomically developed countries, scientists have begun tc>.look seriously at many hitherto unexplored areas for protein rich foods. Algae and leaves are being examined as gnossible sources of protein. Although such foods can be used in animal feeds or make interesting academic studies, acceptability is an important factor in human nutrition. Most of these unusual sources of food need either to be processed suitably or a tremendous effort has to be put in toward educating people to accept the un- usual foods where they may be required. These might be the countries that are not industrially developed and where communication is a problem. In any search for sources of human protein foods several important prOperties need attention: 1. The foodstuff should have a high content of protein and if an increase in the protein is made, the contribu- tion of the food-stuff to the protein in the diet should be significant. 2. It should not have any toxic factors, especially if they are difficult to remove or destroy. 3. It should be easy to grow and the yield must be suffi- cient to make its production economical. 4. If possible, it should be easy to store and transport. This becomes especially important in tropical countries where there is very little refrigeration. Considering these requirements cereals and legumes appear ideally suited to provide an increasing portion of the world's protein needs. Although cereals are consumed in larger quantities as staple foods, legumes contain more protein (20-30% vs 10-20% in cereals). According to an FAO estimate (Food balance sheet, FAO, 1957) the world production of legume seeds amounted to 28.3 million metric tons. India, the largest producer of legumes had an estimated production of 10.617 million metric tons; on a per capita basis this amounted to 60.8g/ caput/day. The United States had the second highest pro- duction, 0.914 million metric tons, which amounted to 10.6g/caput/day. A great deal of effort is currently being focused upon increasing the lysine content of cereals. This in- terest stems from the work of Mertz and his collaborators (Mertz, 1966; Dimler, 1966; Mossé, 1966), who showed that the lysine content of corn was controlled primarily by a single gene (Opaque-2) which is the locus of synthesis of glutelin (alcohol insoluble protein). Most of the lysine and tryptophan is associated with this protein. By alter- ing' the quantity of glutelin the lysine content of corn could be altered. Furthermore the high lysine corn had a biological value similar to that of casein. Mattern gt 31. (1968) from the University of Nebraska, under a project financed by the U.S. Agency for International Development have launched a world wide proj- ect for screening 9,500 varieties of wheat from the USDA collection, for protein and lysine. A similar project has been undertaken by Swaminathan and Austin (1968), at New Delhi, India. In most countries where populations depend heavily upon vegetable sources of protein, the cereal diets are usually supplemented with legumes whenever people can afford them. The legumes contain sufficient lysine to supplement cereals (Patwardhan, 1956). The improvement of the biological value of protein in legumes therefore deserves more attention than is currently being given to it. The governments of some countries are seriously considering fortifying cereals with lysine, the validity of such supplementation is questionable in view of the fact that amino acids can produce toxicities and imbalances as demonstrated by experiments with rats (Harper, 1964). One solution to this problem would be to increase the production of legumes to make them available to most peo- ple, especially those varieties which have been selectively bred to increase their protein content and their biological quality. Before an enthusiastic use of legume proteins is made, toxic factors which are associated with many legume seeds need attention. Very few toxic factors have been reported in peas (Pisum sativum L.), in fact, peas are considered such a good source of food that they are incor- porated into baby foods. Peas are consumed in both the dry and green form; in the United States the green pea is used more frequently. The protein fractions of peas are better charac- terized than most other legumes, yet more information concerning the distribution of amino acids in the protein fractions, the quantity of each fraction present in dif- ferent varieties as it relates to the biological quality, and simple assays for these proteins are required to aid the geneticist in making suitable crosses to improve the quality and quantity of protein. REVIEW OF LITERATURE PROTEINS IN PEAS AND OTHER LEGUMES Several reviews have appeared on plant proteins, seed proteins and nitrogenous components of plant material (Steward and Street, 1947; McCalla, 1949; Danielsson, 1956; Stahman, 1963; Altschule_t_:_ g” 1966). These frequently refer to pea proteins particularly in reference to compar- ison with other legume seeds. Most of the original work on the classification of pea proteins is credited to Osborne and Campbell (1896, 1898). According to their definition which, due to its simplicity, is the most universally accepted, pea proteins are defined as follows: Albumins: are soluble in water and coagulable by heat. Globulins: are insoluble or sparingly soluble in water, but their solubility is greatly enhanced by the addition of neutral salts like sodium chloride. Many globulins are easily prepared from plant and animal tissue since they are readilerxtracted by salt solution and precipitated from saline by dilution with water. The fundamental work on seed proteins was done by Osborne who started in 1891 and during the following 30 years published more than 100 papers on the subject (for references see Osborne, 1924). He and his coworkers used mild methods of preparing their proteins and many of their results are still valid. According to Osborne, pea proteins contained a third unidentified protein, legumelin in addi- tion to an albumin and a globulin fraction. Osborne showed that the seed proteins from dif- ferent members of the Leguminoseae family could be separated into a few simple fractions; each probably containing one protein. Osborne thought that the different proteins were homogenous and consisted of individual proteins. From the seeds of §i§2m_sativum, Eryum_lgn§ and Vigia‘fgba_he isolated two globulins, legumin and vicilin, and a third unidentified protein legumelin. The elementary composition (C,H,N.O,S) of the individual proteins from all the three plants was identical, but small distinct differences between proteins existed even when extracted from the same plant. For example the sulfur content of legumelin, legumin and vicilin was 1.06, 0.5 and 0.16% respectively. This finding is of importance since, in purified proteins, the sulfur might be expected to be present as sulfur containing amino acids, which are limiting in most legume proteins. It is inter- esting to note that other workers have not been able to isolate legumelin. Since the methods used by Osborne were crude, par- ticularly for determining homogeneity and purity of pro— teins, many recent experiments utilize Osborne's work as a base for the study of proteins but check the homogeneity by more modern methods. Location of Pea Globulins The globulins of the pea are reported to be local- ized in distinct subcellular particles called protein bodies, which can be isolated and probably contain globulins only (Varner and Schidlovsky, 1963a). There is little or no globulin outside these bodies. The protein bodies are Spherical in shape and distinctly delineated by a membrane similar in structure to the plasma membrane (Varner and Schidlovsky, 1963b). With the introduction of electron microsc0py there has been renewed interest in protein bodies. Protein bodies or aleurone grains were discovered by Hartig (1856) and later studied in detail by Pfeffer (see Sachs, 1874). Pfeffer concluded that: 1. Protein bodies occur widely in both starch bearing and oil bearing seeds. 2. Some protein bodies contain crystalline inclusions of salts. 3. They are probably surrounded by membranes. 4. They contain most of the cellular protein but none of the oil. 5. Their formation commences only during the later stages of ripening of seeds (that is the time when the storage proteins or the globulins are synthesized). 6 - They swell, coalsce and disappear early in germination having fulfilled their function in the nourishment of the embryo . Extraction Procedures and Homogeneity of Pea Proteins Agreement on the homogeneity of pea proteins is not universal due to various procedures employed for extraction and purification. Osborne's method (1896) consisted of extracting the seeds with naptha to free them of oils, and then extracting with 10% sodium chloride to remove the proteins; the globulins were precipitated by aqueous dilu- tion of the salt solution. The precipitate was washed with alcohol and then dried at 110C over sulfuric acid. Danielsson (1950) modified Osborne's method omitting the naptha extraction, and the alcohol washing which might denature the proteins. The globulins were precipitated from the extract with ammonium sulfate, and further resolved into their component proteins by the isoelectric precipita- tion of legumin at pH 4.7. Wetter and McCalla (1949), working at about the same time as Danielsson, were unable to detect single pro- teins in their purified fractions when these were observed on free boundary electrophoresis. More recent experiments suggest two well defined globulin fractions which can be separated under carefully controlled conditions. Wetter 10 and McCalla used long extraction procedures for the removal of lipids which might have altered the chemical composition of the proteins. In addition they might not have carried the separation of the protein to its conclusion, where they would appear as single proteins on electrophoresis. Workers studying pea proteins from a physiological rather than a chemical standpoint, need to quantitatively fractionate proteins, under such conditions that precise, repeatable results are secured. Accordingly Raacke (1957) modified Danielsson's method after a careful study of results obtained on altering the experimental conditions. In her study, the globulins were separated from the albumins by lowering the salt concentration by dialysis. For the quantitative estimation of legumin and vicilin, both Raacke and Danielsson used an ultracentrifuge. Chromatographic methods have been used to separate pea proteins. Varner and Schidlovsky (1963a) using DEAE cellulose to separate pea proteins observed four peaks instead of the two due to globulins. Grant and Lawrence (1964) similarly separated pea globulins on DEAE cellulose, and they could observe only two peaks corresponding to legumin and vicilin. Grant and Lawrence (1964) studied the chromatographically separated proteins on polyacrylamide gels under the influence of various dissociating agents. They observed two distinct bands corresponding to the two globulins when no dissociating agents were added. Addition 11 of urea gave six bands, four of which were attributed to legumin and two to vicilin. Under the influence of sodium dodecylsulfate 12 bands were obtained, 6 of which were attributed to legumin, 4 to vicilin. Two bands were not identified. Grant and Lawrence (1964) concluded that they might be due to random reassociation of the component fractions. Goffman and Vaintraub (1960) evolved a simple technique for monitoring the homogeneity of seed proteins. This method, based on paper electrophoresis, emphasizes the correct application of the sample to the paper so that the proteins to be separated migrate in the opposite directions. From the observations on the extraction procedures it appears that the homogeneity of proteins, the quantity of protein extracted, and other properties depend to a large extent on the extraction procedure employed, for this reason it is necessary for workers to report the exact experimental details and employ the gentlest methods pos- sible for the extraction of proteins. Physical Properties of Pea Globulins l. Danielsson (1950) investigated pea globulins on free boundary electrophoresis in the pH range 3.7-9.3, and found both vicilin and legumin to be homogenous giving single peaks. The iso-electric pH of legumin was found to be 4.8, a factor utilized in the separation of the two globulins. 12 2. Ultracentrifuge study of pea globulins (Danielsson, 1949b) showed the globulins to be homogenous in the ultracentrifuge. The sedimentation constants were determined to be 12.48 for legumin and 8.18 for vicilin. 3. The molecular weight was 186,000 for vicilin and 331,000 for legumin (Danielsson, 1949b). Chemical Composition of Pea Proteins The elemental composition of pea proteins was studied by Osborne and Campbell (1896). Nitrogen content of proteins extracted by Danielsson and Lis (1952) were similar to those obtained by Osborne. These results are reported in Table 1. From the values reported it appears that legumin contains 18% nitrogen and vicilin contains 17.4% nitrogen which would require the use of conversion factors of 5.55 and 5.75 respectively. The albumin fraction however does contain 16% nitrogen. The sulfur content of the albumin fraction is twice that of the globulins which is significant since sulfur containing amino acids are limiting in peas. Danielsson and Lis (1952) studied the amino acid composition of pea proteins. Their results are reported in Table 2. From their report, the lysine and trypt0phan content of the albumin fraction appears to be 2-3 times higher than that of the globulins. 13 Table l.--Composition of legumin, vicilin, and legumelin according to Osborne and Campbell (1896). Element Pea Lentil Horse Vetch Average bean 12991112 Carbon 51.74 51.73 51.72 51.69 51.72 Hydrogen 6.90 6.89 7.01 6.99 6.95 Nitrogen 18.04 18.06 18.06 18.02 18.04 Sulfur 0.42 0.40 0.39 0.43 0.41 Oxygen 22.90 22.92 22.82 22.82 22.88 Vicilin Carbon 52.36 52.13 52.38 — 52.29 Hydrogen 7.03 7.02 7.04 - 7.03 Nitrogen 17.40 17.38 17.52 - 17.43 Siilfur 0.18 0.17 0.15 - 0.17 Oxygen 23.03 23.30 22.91 - 23.08 Legumelin Carbon 53.6 - - 53.55 Hydrogen 6 . 93 - - 6 . 99 Nitrogen 16 . l4 - - 16. 46 Sulfur 1.00 — - 1.02 OXygen 22.55 - - 22.27 \ 14 Table 2.--Amino acid content (% of weight) of some seed proteins. (From Danielsson and Lis, 1952.) Amino Acid Legumin Vicilin Albumin Tryptophan 0.4 1.3 2.3 Tyrosine 4.2 4.2 5.5 Arginine 11.5 13.1 5.1 Histidine 2.4 3.0 3.2 Lysine 4.6 3.5 10.3 Glutamic 22.7 30.1 10.3 Aspartic 15.7 16.3 14.5 Grant and Lawrence (1964) studied the amino acid composition of legumin and vicilin on the Stein and Moore automatic amino acid analyzer. Their observation indicates that one of the vicilin fractions contains a large portion of the essential amino acids. The results obtained by Danielsson and those obtained by Grant and Lawrence (1964) are in poor agreement, possibly due to differences in the purity of the proteins, and the methods used for the esti- mation of amino acids. Residue Proteins Van Etten 32,31. (1961) determined the amino acid composition of seed meals obtained from 27 genera of 13 botanical families. Biological values as estimated by rat 15 and chick growth showed a majority of the seeds were defi- cient in methionine-- cystine, lysine, or both. Three genera of legume families contained canavanine. The same workers found hydroxy proline in 63 out of 99 solvent extracted acid hydrolysed seed meals. Solubility studies indicated that the compound is a part of the seed coat and the pericarp. Seed meals derived from the kernel alone did not contain any hydroxy proline. Lamport (1965) has shown hydroxy proline to be a part of the primary cell wall protein. Non-Protein Nitrogen of Pea Seed Most seeds contain large amounts of amide nitro- gen and are noted for a variety of non-protein amino acids. The latter may sometimes amount to as much as 2% of the dry weight of the seed (Fowden, 1964). Some examples of unusual amino acids in legume seeds are pipe- colic acid in beans and homoserine in germinating pea seeds. Phylogenetic Relationship of Globulins in Legume Seeds Studies of the families leguminoseae and Malvaceae (Shadmanov, 1964) showed that phylogenetically old species 16 were characterized by low quantities of nitrogen (and hence protein) in comparison to higher forms. A relationship was suggested between the position of a plant on the evolution scale and the nature of the Storage protein. In older forms of bean plants there was a predominance of the more difficultly soluble (legumin with higher molecular weight) protein over the more easily soluble form (vicilin with lower molecular weight). That is in the phylogenetically younger plants, vicilin predominates. Fox gt_gl. (1964) electrophoresed proteins to establish such relationship within the family leguminoseae. Physiological Function of Pea Proteins The genetic selection of certain varieties would be greatly facilitated if the time of synthesis and the physiological role of the component proteins were known. Albumins The albumin fraction is heterogenous in composi- tion and the role ascribed to it is mainly enzymic. Among the enzymes isolated from it are: a proteolytic enzyme (Young and Varner, 1959), a peroxidase, a catalase, a phosphatase, an amylase and an aldolase (Hatz and Leuthardt, 1967). Danielsson (1952) worked on protein synthesis in ripening pea seeds and showed that the total protein increased 17 on ripening. In unripe seeds all of the protein (100%) was present as albumin. On ripening of the seeds, al- though the absolute amount of albumin increased, there was a great drop in the percentage of protein present as albumin (17%). Since the albumin fraction is heterogenous it is possible that more than one function might be allocated to it. Function of the Globulins While the albumins appear to be enzymic in func- tion, globulins are considered storage proteins. During synthesis, the lower molecular weight vicilin appears first. The appearance of globulins coincides with the appearance of protein bodies. 18 Quantitative Variation in Protein as Influencediby Genetic Variation It appears possible to make a precise science of the study of peas or other legumes based on the quantity of protein in these seeds. If no toxic factors are pres- ent in the seed, the increment in the quantity of protein in the absence of improvement in quality is desirable. Hegsted (1962) calculated that human diets based on wheat alone could provide sufficient amino acids for growth. Bolourchi (1967) showed the ability of wheat diets to main- tain adult human subjects in nitrogen equilibrium. Camp- bell (1963), showed that with casein, maximum growth per unit protein intake was obtained when it was fed at a 7% level. With plant proteins, maximum growth was obtained when they were fed at a 15% level in the diet. These ob— servations indicate the usefulness of increasing the quantity of protein in legumes or other seeds. Pesola (1955) studied the variation in protein quantity of peas as influenced by genetic variation and climatic conditions. He concluded that the variation in protein content of peas is a varietal character, and showed that the protein content could be increased by the selection of high protein lines. Pesola suggested that the protein content may vary with the varying intensity of the symbiotic action of different pea varieties and root nodule bacteria. Climatic conditions were likewise shown to influence the protein content. 19 Evans gt 31. (1947) studied the effect of fertilizer treatment, environmental conditions and genetic variation, on protein, cystine and methionine content of peas. They showed that the protein content was determined by the loca- tion where the plants were grown and by breeding. The methionine content could not be altered by any of the conditions studied, however, the cystine content could be increased by the use of sulfur containing fertilizers. Since methionine is required partly to synthesize cystine in the mammalian system, am: increase in cystine content by the use of sulfur containing fertilizers might be important. Esh eE_al. (1959) studied the effect of environ- mental conditions and genetic strain on the protein content of legume seeds. They found variation in protein content (nitrogen X 6.25) when seeds of different strains were grown in the same locality. A variation of as much as 60% was sometimes noted. Similarly, location influenced the protein content, a variation of 13-34% was observed with changes in locality where the peas were raised. The quantity of protein is influenced by genetic variation. This has been shown for several major seed crops; such as, wheat (Mattern 33 31., 1968) and sorghum (Swaminathan and Austin, 1968). These workers are using this information to cross high protein lines and obtain nutritionally superior varieties. 20 In addition to the effect of genes in changing protein quantity, herbicides such as simazine (2-chloro-4, 6-bis ethyl amino-s-trizaine) have been shown to increase the protein content of peas and oats (Ries 23.313! 1967). In addition to increasing the protein content, this herbi- cide was reported to increase the yield (Schweizer and Ries, 1969). 21 Biological Value Although many studies have been done on the in- fluence of genes on the quantity of protein, very little work has been done to show the effect of genes on protein quality. The quality of protein in foods depends on several factors, such as, the digestability of protein, the amino acid composition, the availability of the amino acids and the presence of toxic factors. Systematic studies to ob- serve differences in the digestability of protein, the amino acid composition, the availability of amino acids and the toxic factors as influenced by genetic variation have only been initiated in the past few years. Mertz gp‘gl. (1965) showed that the protein quality of high lysine corn was very similar to that of casein. This improvement in the biological value of corn resulted from an increase in the lysine content. Most of the lysine in the corn was reported to be associated with the alcohol insoluble protein fraction called glutelin (Dimler, 1966). Although chemical scores were obtained for the alcohol soluble proteins (prolamines) to show that they were of poor nutritional quality (Mossé, 1966) the same was not done for the glutelin fraction. A limited number of studies on the nutritional quality of proteins in legume seeds (Esh, 1958; Niyogi gE_§EL., 1932) as influenced by genetic variation showed that.the protein content of these seeds varied from 20-30%, 22 the digestability from 75-95%; and the biological value from 45-70%. Evans and Bandemer (1967) studied the nutritional value of legume seeds. The two varieties of peas studied (Alaska and First and Best) by these authors had different biological value as estimated by rat growth. The results secured by these authors were not corrected for food intake, but expressed as the ratio of growth produced when the experimental diet was fed compared to that produced by the casein ration. Such results are reported to be highly variable (Campbell, 1963). According to the amino acid content, First and Best had a higher quantity of lysine, methionine and cystine; despite this the biological value of Alaska pea was higher (51 Vs 41 for First and Best). Supplementation of the pea ration with methionine at 0.5% level improved the biological value of Alaska pea from 51 to 124%. Methionine has been shown to be a limiting amino acid of peas and other legumes. Tryptophan is occasionally reported to be a limiting amino acid of legume seeds, how- ever no studies with supplementation of tryptophan were conducted. Similar studies on supplementation of methionine to pea.diets at 0.1% level resulted in a dramatic improvement of the PER of peas from 0.5 to 2.35 (Russel 33 21" 1946). Mitchell and Block (1946) postulated a relationship Ibetween a limiting amino acid in a dietary protein and its 23 biological value. It appears possible that PER of cereal proteins should be influenced by lysine and legume proteins by methionine and possibly tryptOphan which are limiting amino acids in the respective foodstuffs. Suggestions are not wanting that nutritive value of cereal proteins should be improved by the addition of the amino acid which is the limiting factor. This suggestion is not capable of imple- mentation on a large scale in the realm of human nutrition. It seems far more practicable to attempt to improve the nutritional value of cereal proteins by incorporation in the diet of other foodstuffs (such as legumes) which will supply the deficient amino acids. In fact Patwardhan (1956) observed that in Indian dietaries, a combination of legumes and cereals provide sufficient lysine for daily adult requirements. Phansalkar and Patwardhan (1956) observed that the supplementation of cereal protein with legume protein im- proved the biological value and egg replacement value of the mixture in human subjects.~ Phansalkar gt_§1. (1957) studied the effect of supplementation of cereals with legumes by the rat growth method. The PER of wheat, rice and millets was improved when these were supplemented with legumes. Red gram (Cajanus ggjgn) was more effective than other legumes (Cicer arietinum, Phaseolus mungo and Phaseolus radiatus) as a supplementary source. 24 In an extensive review, Swaminathan (1967) tabulated the supplementary effect of plant proteins. The supplemen- tation of peas with barley in the ratio of 1:1 gave a PER of 1.98 when either of these alone had PER values below 1.5. Several attempts have been made to prepare nutri— tious milk substitutes based on vegetable protein. Soy bean and peanut milk was reported to be successful in feed- ing infants in China. Dean (1953) reported that highly nutritious spray dried food could be prepared from a blend of barley malt and soya bean, and that about half of the milk in the diet of infants up to 1 year of age could be replaced by soy bean foods without affecting growth and nutritional status. Several brands of proprietary foods based on soy bean are manufactured and used in the U.S.A. for feeding infants who are allergic to cow's milk (Meyer, 1960). Chick pea (Cicer arietinum) has been used in protein foods based on oil seed meals (Parpia gE_al., 1964). This food was reported to be effective in treating protein mal- nutrition in children. Scrimshaw e; 31. (1960) found that a mixture of corn and black bean protein was not effective in curing protein malnutrition, but the addition of lysine and tryptophan to the diet significantly increased the nitrogen retention in malnourished children. Hansen 22.21: (1960) reported that a mixture of corn and pea meal did not initiate cure in three cases of Kwashiorkor. 25 Factors Affecting the Nutritive Value of Plant Proteins l. Toxic Factors.--These substances are present in a large number of foods and exert a deletrious affect on the utili- zation of proteins and the growth of the animal. A number of toxic factors have been reported in legume seeds (Mickelsen and Yang, 1966), among those reported are: a growth depressant in raw soya bean meal (Saxena 35 31., 1962) and in kidney beans (Kakade and Evans, 1965), and trypsin inhibitors in raw soy bean meal (Ham gt 21., 1944), and anti thyroid compounds in legume seeds (Greer and Astwood, 1948). Although peas have not been reported to contain toxic factors, as evaluated by improvement of growth following heat treatment, it is significant that Greer and Astwood reported that unlike the anti thyroid compounds of other legume seeds the anti thyroid activity of peas was not destroyed by heating. 2. Effect of Heat Processing.--Heat produces both bene- ficial and deletrious effects on the nutritive value of proteins. The beneficial effects are due to inactivation by heat of the trypsin and growth inhibitors, hemagglutinins and other toxic factors present. The adverse effects are due to the decrease in the availability of certain essential amino acids, such as lysine and methionine, as a result of reaction with reducing sugars and carbonyl compounds present in foods. This mechanism (Maillard's reaction) has been 26 extensively studied (Patton, 1955; Ellis, 1959). The loss of amino acid depends on the severity and length of heat treatment and the moisture content of the food. In general, foods, such as legumes containing trypsin and growth inhib- :itors, show marked improvement in the nutritive value of their proteins on processing. 3. Amino Acid Imbalance, Deficiency, Toxicity, and Antag- onisnu-mPartial deficiency of any one essential amino acid adversely affects the utilization of proteins for the maintenance of nitrogen equilibrium and growth. Certain amino acids when present in quantities that greatly exceed the requirements, adversely affect the growth of the an- imals. These effects have been ascribed to: (a) amino acid imbalance, (b) amino acid toxicity and (c) amino acid antagonism (Food and Nutrition Board, 1963). Amino acid imbalance may be defined as a condition in which increase in the concentration of certain amino acids brings about an increase in the amount of other amino acids needed to main- tain a given growth rate when protein intake is low (Harper, 1959). In a two week study using gluten, the requirements for lysine as a percentage of the diet and the total amount cm lysine required to support growth were both increased (Munaver and Harper, 1959). The quantity of excess amino acid required to cause mndcity vary greatly. Methionine,cystine, tyrosine, tryp- tOphan, and histidine present in excess quantities are 27 more toxic than arginine, lysine, threonine, isoleucine and valine (Sauberlich, 1961; Salmon, 1958). The signs and symptoms of toxicity include depression in growth, increased mortality, and histological changes in the liver, skin and pancreas. In amino acid antagonism, excess of the amino acid depresses the utilization of a structurally similar amino acid; e.g., excess of leucine depresses utilization of isoleucine, as in corn proteins (Harper SE _a_l” 1955) and excess of lysine depresses utilization of arginine (Jones, 1962). 4. Amino Acid Availability.--This depends on (a) digest- ibility coefficient of a protein and (b) rate of release of amino acids during digestion. There is evidence that some of the essential amino acids present in proteins may not be fully released after digestion, and the rate of release of different amino acids varies from protein to protein during digestion (Mauron _e_t_ _a_1_., 1955) . For example, the ten essential amino acids were all highly available from peanut flour and wheat (92-100%) , whereas in cotton seed meal the availability ranged from 64.5—93.4%. The utilization of lysine from 19 food proteins ranged from 49-98%, and of methionine from 48-83% (Guthneck 32 11., 1953; Schweigert and Guthneck, 1954). Raw soy bean meal gave the lowest values for the availability of both lysine and methionine. 28 From the review of literature several indications towards the direction of research were obtained. The focal points of interest are: 1. In peas and legume seeds sulfur containing amino acids and tryptophan are limiting. 2. The quantity and quality of seed proteins have been shown to vary with varietal difference. 3. An alcohol insoluble protein fraction contained a high concentration of the limiting amino acid of maize seeds. The genes responsible for the synthesis of this fraction were located and used to obtain nutritionally superior variety of maize. 4. work with pea proteins indicated that 2 or 3 distinct proteins could be obtained from it. If sulfur in pea proteins could be utilized as an index 5. of sulfur containing amino acids, the albumin fraction which contained twice the amount of sulfur might be expected to have a high biological value. In addition the albumin fraction contained two to three times the) concentration of lysine and tryptOphan present in either of the globulins. 6. The availability of the amino acids from the water soluble albumin fraction might be higher. Most indications point to albumin fraction as one of good protein quality. REFERENCES «Altschul, A. M., Yatsu, L. Y., Ory, R. L. and Engleman, E. M. (1966). Seed proteins. Ann. Rev. Plant Physiol. 11, 113-136. Boulter, D. and Thruman, D. (1968) . Acrylamide gel elec- trOphoresis of proteins in plant systematics. Systema- tics Assoc. Special volume 2, 39-48 (ed.) Hawk, J. G. Bolourchi, S. (1967) . Wheat as a source of protein for human adults. Effect of wheat protein on nitrogen balance, urea, and amino acid metabolism. Ph.D. Thesis Michigan State University . Campbell, J. A. (1963). Methodology of protein evaluation. A critical appraisal of methods for evaluation of pro- tein in foods. Div. Food Tech. and Nutr. Faculty of Agric. Sci. American Univ. of Beirut, Lebanon. Dalby, P. L. and Lillevik, H. A. (1969). Taxonomic analysis of electrOphoretic blood serum patterns in cotton rat, Sigmoidon. Biological series 4 (3), Publications of the Museum, Michigan State UniversIty. 1r Danielsson, C. E. (1949a). Investigations of legumin and vicilin. Acta Chem. Scand. 3, 41-49. ”Danielsson, C. E. (1949b). Seed globulins of the Gramineae and Leguminoseae. Biochem. J. 4:1, 387-400. Danielsson, C. E. (1950). An electrOphoretic investigation of vicilin and legumin from seeds of peas. Acta Chem. Scand. 4, 762-771. Danielsson, C. E. (1952). A contribution to the study of the synthesis of the reserve proteins in ripening pea seeds. Acta Chem. Scand. _6_, 149-159. v" Danielsson, C. E. (1956). Plant proteins. Ann. Rev. Plant Physiol. 7_, 215-236. v Danielsson, C. E. and Lis. H. (1952). Differences in the chemical composition of some pea proteins. Acta Chem. Scand. 3, 139-148. Dean, R. F. A. (1953). Plant Proteins in Child feeding. Med. Res. Counc. (Brit.) Special Report 279. 29 30 Dimler, R. J. (1966) . Alcohol insoluble proteins of cereal grains. Fed. Proc. _2__5_, 1670-1675. Ellis, G. P. (1959) . The Maillard reaction. Advances in carbohydrate chemistry at 63. Esh, G. C. (1958). J. Proc. Inst. Chemists (India) 30, 21. Esh, G. C., De, T. S., and Basu, U. P. (1959). Influence of genetic strain and environment on the protein content of pulses. Science 129 (3342), 148-149. Evans, R. J., St. John, J. L., Craven, P. M., Hadock, J. L., Wells, D. G., and Swenson, S. P. (1947). Some factors influencing the protein cystine and methionine content of dry peas. Cereal Chem. 24 (2), 150-156. Evans, R. J. and Bandemer, S. L. (1967). Nutritive value of legume seed proteins. Agric. Food Chem. l_5_ (3), 439- 443. FAO, Food Balance Sheet, FAO, Rome, Italy (1959). Food and Nutrition Board (1963). Evaluation of protein quality. Public. 1100 Natl. Acad. Sci. - Nat. Res. Counc. Washington, D. C. Fowden, L. (1964). The chemistry and metabolism of recently isolated amino acids. Ann. Rev. Biochem. 33, 173-204. Fox, D. J., Thruman, D. A., and Boulter, D. (1964). Studies on the proteins of seeds of the Leguminoseae. l. Phyto- chemistry _3_, 417-419 . Goffman, Y. Y. and Vaintraub, I. A. (1960). The use of paper electrophoresis in checking homogeneity of seed proteins. Biochemistry (U.S.S.R.) 22 (6), 1049-1054. v” Grant, D. M. and Lawrence, J. M. (1964). Effects of sodium dodecyl sulfate and other dissociating reagents on glo- bulins of peas. Arch. Biochem. Biophys. 108, 552-561. Greer, M. A. and Astwood, E. B. (1948). The antithyroid effect of certain foods in man as determined with radio- active iodine. Endocrinology 43, 105-119. Guthneck, B. T., Bennet, B. A., and Schweigert, B. S. (1953). Utilization of amino acids from foods by the rat. J. 31 Ham,VL E. and Sandstedt, R. M., and Mussehl, F. E. (1944). A.pnfi£olytic inhibiting substance in the extract from unheated soyabean meal. J. Biol. Chem. 154, 505-506. Hausa» J. D. L., Schendel, H. E., Wilkins, A. and Brock, J.IL (1960). Pediatrics. 22, 258. Amino acid balance and imbalance. J. 1. Harper, A. E. (1959). Dietary level of protein and amino acid imbalance. Nutr. Q, 405-418. Harper, A. E. (1964). 22_Allison, J. B. and Monroe, H. N. protein metabolism, 2, 119. Amino acid toxicities and imbalances. (eds.) Mammalian 14, 257. In Bonner, J. and (1856). Botan. Z. Plant BIochemistry. Acad. Hartig, T. ___ Varner, J. E. (eds.), p. 775. Press. Hatz, C., Leuthardt, F. (1967). Isolation and characteriza- tion of fructose diphosphate aldolase from Pisum sativum. Biochim. Biophys. Acta 139, 460-468. The role of wheat in the World's Hegsted, D. M. (1962). food supply. Report of conference. April 30th-May 2nd at Albany, California. Jones, J. D. (1962). Observations on the toxicity of (Abstract) Fed. Proc. 22, l. lysine. Kakade, M. L. and Evans, R. J. (1965). Growth inhibition of rats fed navy bean fractions. Agric. Food Chem. 22, (5) 450—452. The protein components of primary (1965). Lamport, D. T. A. Botan. Res. 2, cell walls. Advanc. Mattern, P. J., Johnson, V. A., Schimdt, J. W. (1968). Improvement of the nutritional quality of wheat through increased protein content and improved amino acid bal- Fifth semiannual progress report. Dept. Agronom. ance. Nebraska, Lincoln. Univ. Mauron, J., Mottu, F., Bujard, E. and Egli, R. H. (1955). The availability of lysine, methionine and tryptophan in condensed milk and milk powder. In vitro digestion stu- dies. Arch. Biochem. BiOphys. a, 433-451. Nitrogenous constituents of plants. McCalla, A. G. (1949). Ann. Rev. Biochem. 22, 615-638. 32 Mertz, E. T., Bates, L. S. and Nelson, 0. E. (1964). MuUHm.gene that changes protein composition and in- creases lysine content of maize endosperm. Science 222,279-280. L. S. and Nelson, 0. E. Mertz, E. T., Vernon, O. A., Bates, Science U965). Growth of rats fed opaque-2 maize. 258,1)41—1742. Mertz,1L T. (1966). Improving the nutritional value of cereal grains by the use of gene mutations and selective breeding. Fed. Proc. 2_5_, 1662. Meyer,IL F3 (1960). Infant foods and feeding practices. Thomas, Springfield, Illinois. Mickelsen, O. and Yang, M. G. (1966). Naturally occurring toxicants in foods. Fed. Proc. 22 (1), 106-123. and Block, R. J. (1946). Some relationships Mitchell, H. H. between amino acid content of proteins and their nutri- tive value for the rat. Biochem. J. 163, 599-620. Mossé, J. (1966). Alcohol soluble proteins of cereal grains. Fed. Proc. 2;, 1663-1669. Munaver, S. M. and Harper, A. E. (1959). Amino acid balance and imbalance. 11. Dietary level of protein and lysine requirement. J. Nutr. g2, 58-64. Niyogi, S. P., Narayana, N. and Desai, B. P. (1932). Ind. J. Med. Res. 22, 1041. 23 Influence of genetic strain and environment on the protein content of pulses. Esh, G. C., De, T. S. and Basu, U. P. (1959). Science 129, p. 148. / Osborne, T. B. (1924). The vegetable proteins. 2nd edi- titni. (Longmans, Greens and Co., London, England) 168. (1896). Legumin and v Osborne, T. B. and Campbell, G. F. J. Am. Chem. Soc. 22, other proteids of pea and vetch. 583-608. T. B. and Campbell, G. F. lentil, horse bean and vetch. (1898). The proteids of ~ Osborne, J. Am. SOC. 2_0.’ 410— pea, 427. Parpia, H. A. B., Narayana Rao, M., Rajagopalan, R. and Swaminathan , M. 83 (1956). Ind. J. Med. Sci. 10, 1. In Patwardhan , V . N . _ _ Part 1. Phans- Nutritive value of vegetable proteins. alkar, S. V., Ramachandran, M. and Patwardhan, V. N. (1957). Ind. J. Med. Res. 22, 612. Paranu S. (1955). J. Dairy Sci. 22, 457. In Albanese, A.ZL (ed.) Newer methods of nutritional Biochemistry. _3_, 210 Acad. Press, N.Y. Protein content of field pea seeds Pesola, V. A. (1955). Acta Agral. Fenn. 22, 125-132. as a varietal character. Phansalkar, S. V. and Patwardhan, V. N. (1956). Utilization of animal and vegetable proteins. Nitrogen balance at marginal protein intakes and the determination of minimum protein requirements for maintenance in young Indian men. Ind. J. Med. Res. 22, 1-10. and Patwardhan, V. N. Ramachandran, M. Part 1. Phansalkar, S. V., Nutritive value of vegetable proteins. (1957). Protein efficiency ratios of cereals and pulses and the supplementary effect of addition of leafy vegetables. (l957a,b,c). Protein synthesis in ripening pea seeds. 1. Analysis of whole seeds. 2. Development of embryos and seed coats. 3. Study of the pods. Biochem. J. 22, 101-110, 110-113, 113-116. Raacke, I. D. Ries, S. K., Chmiel, H., Dilly, D. R. and Filner, P. (1967). Increase in nitrate reductase activity and protein con- tent of plants treated with simazine. Proc. Natl. Acad. Sci. 23 (2), 526-532. Russelq v1. G., Taylor, M. W., Mohroff, T. G. and Hirch, (1946). The nutritive value of the protein of R. Rm ‘varieties of legumes and the effect of methionine sup- plementation. J. Nutr. 22, 313-325. Saxena, H. C., Jensen, L. S. and McGinnis, J. (1962). Failure of amino acid supplementation to completely overcome growth depression effect of raw soybean meal in chicks. J. Nutr. _7_Z_, 259-263. Sachs, J. (1874). Text book of Botany. 51. Leipzig. Salmon, W. D. 34 Sauberlich, H. E. (1961). Studies on the toxicity and antagonism of amino acids for weanling rats. J. Nutr. 22., 61-720 Schweigert, B. S. and Guthneck, B. T. (1954). Utilization of amino acids from foods by the rat. III. Methionine. Jo NUtr. 5_4-’ 333-334. Schweizer, C. J. and Ries, S. K. (1969). Protein content of seed: Increase improves growth and yield. Science 165, 73-75. Scrimshaw, N. S., Behar, M., Wilson, D., Viteri, F., Arroyave, G. and Bressani, R. (1960). Am. J. Clin. Nutr. 2, 196. In Albanese, A. A. (ed.) Newer methods of nutritional Biochemistry. 2, 229. Shadmanov, R. K. (1964). Evolyutsionnya Biokhim. Rast. 128-137 (N.V. Tsistsin, ed. Moscow) Smirnova, T. I., Poglazov, B. F. and Kretovitch, V. L. (1959). Biochemistry (U.S.S.R.). 22, 695-697. 22' Altchul, A. M., Yatsu, L. Y., Ory, R. L. and Engleman, E. M. (1966). Seed proteins. Ann. Rev. Plant Physiol. 21, 116. Stahman, M. A. (1963). Plant proteins. Ann. Rev. Plant Steward, F. C. and Street, H. E. (1947). The nitrogenous constituents of plants. Ann. Rev. Biochem. 22, 471-502. Swaminathan, M. (1967). Availability of plant proteins. In Albanese, A. A. (ed.) Newer methods of nutritional bIochemistry 2, 197-237. Acad. Press. Swaminathan, M. S. and Austin, A. (1968). Studies on pro- tein quality of sorghum, millet, wheat and other cereal grains for the period 1967-1968. Ann. Progress Report Cummings Laboratory, Ind. Agric. Res. Inst. New Delhi, India. ‘ Van Etten, C. H., Miller, R. W., Wolff, I. A. and Jones, Q. (1961). Agric. Food Chem. 9, 79-82, 433-435. In Stahman, M. A. (1963). Plafit proteins. Ann. 125?. Plant Physiol. 22, 144. Varner, J. E. and Schidlovsky, G. (1963a). Intracellular distribution of proteins in pea cotyledons. Plant Physiol. 22, 139-144. 35 Varner, J. E. and Schidlovsky, G. (1963b). Intern. Seed Protein Conf. U.S.D.A., New Orleans. “1 Wetter, L. R. and McCalla, A. G. (1949). Electrophoretic properties of pea proteins. Canad. J. Res. C 27_, 96-102. Young, J. L. and Varner, J. E. (1959). Enzyme synthesis in the cotyledons of germinating seeds. Arch. Biochem. Biophys. 8_4_, 71-78. PART I BIOLOGICAL VALUE OF THE PROTEIN IN DIFFERENT VARIETIES OF PEAS BIOLOGICAL VALUE OF THE PROTEIN IN DIFFERENT VARIETIES OF PEAS ABSTRACT The biological value of 32 varieties of peas (22ggm_ sativum L.) grown under similar field conditions was deter- mined using weanling rats, by means of net protein utiliza- tion (NPU) and protein efficiency ratio (PER). Different varieties of peas when fed as the sole source of protein at a 10% level in an otherwise adequate diet varied, in com- parison to casein, from 0.18-0.78 in their ability to sup- port growth and nitrogen retention as estimated by NPU and PER. From analysis of rat growth curves, the pea varieties were separated into those that produced fairly good growth and those which barely maintained the initial weight of the rats. Carcass protein as a percentage of body weight was higher in pea fed (20-22%) than in casein fed rats (18-l9%). This was probably associated with a difference in the body fat content. The total nitrogen content in different varieties of peas provides a poor indication of their biological value. 36 h“ 1 q 37 INTRODUCTION Legumes are an important source of dietary protein in developing countries for several reasons. In a few countries a large number of people are vegetarians for one reason or another and so depend on seeds as a primary source of protein. Legumes, in their dry state, can be stored for long periods of time and can be transported for human con- sumption with a minimum amount of special treatment. The ease of storage, preparation and transport of dry leguminous seeds makes them ideal protein sources for use in disaster areas. Although India is the biggest consumer of legumes (FAO, 1957) with a per caput intake of 60.8g per day; the inhabitants of the United States and United Kingdom consume a fairly large quantity (10.6 and 11.0g respectively) of legumes per day. In the United States peas are used in baby foods and if special foods for the aged are marketed they might find use in them. Most of the work on peas has been directed towards improving their appearance, yield or freezing and canning qualities. The little work that has been done on the nu- tritional value of peas (Esh 22.22., 1959; Pesola, 1955) has been limited to increasing their nitrogen content. Almost no work has been reported on the biological qualities of legume proteins from different varieties of seeds, al- though it appears logical to assume that such differences exist. 38 New interest arose in improving the nutritional value of seed proteins when Mertz and his collaborators (Mertz 22'223, 1964, 1965, 1966) showed that the biological value of maize protein could be improved to such an extent that it approached or equalled that of casein. This re- sulted from a threefold increase in lysine and tryptophan content of the maize. The present study was conducted to evaluate the biological value of the nitrogen components of 32 varieties of peas, hoping thereby to find some varieties with superior biological value to be used in future breeding experiments. EXPERIMENTAL The peas were grown at the Horticulture Research Center, East Lansing in 1966. Rainfall was adequate for growth. The peas were dried on the vine and later harvested by hand. The seeds were stored until used at 35-40°F and low humidity. Alaska peas, a commercial variety (lot number 54401) grown under standard conditions of commercial seed production were used for comparison. The peas were ground in.a Wiley mill until the powder passed through a 20 mesh screen. .All samples were analyzed for nitrogen by the Kjeldahl method (AOAC, 1965). The nitrogen value was multiplied by a factor of 6.25 (Hegstedq 1964) to facilitate the comparison of the 39 nitrogenous fraction of the peas with casein, which was used as the reference standard in the bioassays. For the bioassays, the pea meal was added to a gnuified rationl at the expense of sucrose. All the samples were incorporated into the ration to provide a total nitrogen equivalent to 10% protein level. Weanling male Sprague Dawley rats were maintained I I 2 O O on grain rat1on for a week. Five rats were ass1gned to each group in such a way that the group weights did not vary by more than 3g. An extra group of 5 rats in each assay was used as controls for the evaluation of changes in body protein content. Since the 32 varieties of peas could not be assayed in the same experiment, four separate exper- iments were performed. Casein was repeated in every exper- iment, and a few varieties of peas were assayed 2-3 times. One group of rats in each assay (body composition controls) was sacrificed at the start of the experiment. The carcass (minus the gastro intestinal contents) of each of those rats was analyzed for nitrogen thereby providing values for the initial quantity of protein in the carcasses. lPurified ration: Corn oil 5%, Salt mix 4%, Vitamin Hug: 2.2% (obtained from Nutritional Biochemicals), Alpha cel 2%, Peas or casein to provide 10% protein (based on nitrogen x 6.25) , Sucrose to make up 100%. 2Grain ration: Ground corn 60.7%, Soy bean meal (50% protein) 28%, Alfalfa (17% protein) 2%, Fish meal (12.5 protein) 2.5%, Dried whey (67% lactose) 2.5%, Lime stone (38% calcium) 1.6%, Dicalcium phosphate (18.5% P, 22-25% Ca) 1.75%, Iodized salt 0.5%. 40 The remaining animals were fed the experimental diets for a period of 3 weeks. Food was provided ad libitum. The weight gains and the feed intake was measured every week. At the end of the 3 week period the animals were sacrificed, the contents of the gastro intestinal tract removed, the animals were autoclaved, homogenized (Mickelsen and Ander- son, 1959) and a sample thereof analyzed for nitrogen. From this value, the total nitrogen in the carcass was calculated. Biological values of the proteins in the diets were calculated as: PER = Weight gain / g protein eaten Nitrogen retained in carcass X 100 Nitrogen intake NPU RESULTS AND DISCUSSION The protein content based on the nitrogen values, of vine dried peas varied from 21-28% (Table 3). The nitrogen content may not be an accurate estimate of the total protein. It is possible that different varieties of peas contain different quantities of non-protein-nitrogen (NPN) . The NPN may contain peptides and amino acids which are nutritional equivalents of protein, and other compounds which cannot be utilized by monogastric animals. To reduce the non-utilizable NPN (other than the nucleic acids neces- sary for protein synthesis) and improve the biological 41 Table 3.--Percent Protein in Peas, NPU, PER, Percent protein in carcass and the weight gain on pea diets in comparison to casein diet. Percent Protein Percent in the Protein Protein Experimental Protein Supplement Efficiency in Weight Gain Source4 (Nx6.25) NPU5 Ratio Carcass of Rats Casein 87.5 52.34 2.78 18.83 75.61 66-1 24.4 32.57 1.443 22.623 33.8 66-2 24.4 25.95 1.283 20.28 31.6 66-3 25.9 16.11 0.763 21.213 23.4 67-4 25.5 39.53 2.113 18.73 39.2 66-5 25.8 9.66 0.463 21.023 7.73 66-6 25.4 30.59 1.403 21.853 33.8 66-7 27.2 36.84 1.803 20.47 42.61 66-8 27.2 26.36 1.163 22.733 33.0 66-9 27.2 22.26 1.003 22.273 21.6 66-10 28.1 32.63 1.613 20.27 20.4 66-11 28.2 30.93 1.463 21.193 34.6 66-12 22.9 31.00 1.563 21.603 37.4 66-13 28.5 28.10 1.493 18.86 28.8 66-14 26.9 33.66 1.703 19.8 35.8 66-15 26.7 37.50 1.863 20.20 34.8 66-16 28.3 27.18 1.343 20.30 27.2 66-17 27.7 31.99 1.693 20.40 32.0 66-19 25.2 31.55 1.573 20.10 24.8 66-20 23.5 33.75 1.673 20.21 27.8 42 Table 3.--Continued Percent Protein Percent in the Protein Protein Experimental Protein Supplement 5 Efficiency in Weight Gain Source4 (Nx6.25) NPU Ratio Carcass of Rats 66-22 21.6 35.48 1.833 19.39 60.856 66-23 24.8 37.70 1.883 20.06 35.8 66-24 26.3 41.28 2.053 20.14 35.2 66-26 25.1 35.71 1.763 20.29 35.2 66-27 23.6 26.67 1.283 "20.84 22.0 66-28 23.7 23.47 1.163 20.24 24.6 67-28 21.2 32.30 1.653 19.58 20.8 66-31 26.7 44.44 2.203 20.20 37.1 66-33 26.9 33.62 1.803 18.68 36.6 66-34 28.5 34.37 1.853 18.58 38.2 66-38 28.2 25.58 1.283 19.99 13.0 66-52 23.5 26.27 1.313 20.06 43.29 Alaska 22.9 13.65 0.6353 21.53 3.00 3Significantly different from casein at (P<0.05). 4Casein was vitamin-free assay protein - secured from General Biochemicals. The numbers refer to the dif- ferent varieties of peas with the first two figures indi- cating the year when the peas were grown. 5NPU = Nitrogen retained in carcass X 100 Nitrogen’intake 6The weight gains are higher in this experiment because the animals were fed experimental diets for a period of 5 weeks instead of the 3 weeks in all other experiments. 43 value of protein in peas should be an important considera- tion in any pea breeding study. The protein efficiency ratio is a function of weight gain and food intake. On the basis of weight gain, the best pea varieties (67-4, 66-24, 66-31; Table 3) were approximately half as effective as casein. The PER of these varieties was closer to casein than anticipated, possibly due to the lower food intake of pea-fed rats. The reduced intake of food on pea diets may be due to depression of appetite brought about either by poor protein quality or the presence of toxic substance(s) in the un- cooked peas. That there is a difference in the response of the animal to the two types of diets (casein vs. peas) is suggested by the observation that the size of the stomach and cecum of the rats fed pea diets were 2-3 times larger than those of the casein fed rats. The PER of the same sample when assayed more than once remained fairly constant (Std. dev. for PER = 0.12, and NPU = 2.12, Table 4). Small deviations occurred when periods as long as 6 months intervened between 2 assays. Similar results have been observed by other workers (Chapman 32 31., 1959; Campbell, 1963; Jansen, 1962). The variation was much greater when the value of the PER was low, for the higher values the deviation in the results of the same sample between assays was surprisingly small. The greater deviation in the values for peas of low biological value stems from the fact that small fluctuations in body weight gains of rats markedly influence the values. Table 4.--Variation in PER and NPU Between Experiments. Source of Experiment 1 Experiment 2 Experiment 3 EXperiment 4 Protein PER NPU PER NPU PER NPU PER NPU Casein 2.77 52.16 2.67 56.0 2.89 50.20 2.80 51.00 66-5 - - 0.25 9.60 0.67 7.60 0.46 11.8 66-7 - - 1.80 34.90 1.79 36.60 1.74 39.06 66428 - - - - 1.30 23.52 1.02 23.42 66-14 - - - - 1.68 35.22 1.72 32.1 66-33 - - - - 1.82 35.20 1.78 32.70 66-24 - - - - 2.00 39.1 2.10 43.5 67-4 2.20 40.8 - - - - 2.02 38.40 Standard deviation: A common standard deviation was calculated for each parameter since the variances (P< 0.01) were equal. II 0 o l—I N Standard deviation for PER II N [.1 N 0 Standard deviation for NPU The net protein utilization is based on the fraction of consumed nitrogen retained in the carcass. Obviously nitrogen retention in the carcass is determined by weight gain and carcass composition. Since the ratio of weight gain to food intake is already determined by PER, the factor of importance in differentiating NPU from PER is 45 the percentage of protein in the carcass. The percentage of protein in the carcass of rats fed some varieties of peas was significantly higher (P<0.05) than in casein fed rats (Table 3). The higher protein percentage in the car- cass of pea fed animals suggests that they had a smaller amount of fat in their carcass. The PER and NPU values for casein are within the range of reported values (Morrison, 1964). From the rat growth curves (Figure 1), pea diets can be designated as supporting: (l) fairly good growth, (2) mediocre growth and (3) bare maintenance of initial weight. Rats fed poor quality diets usually showed an initial loss of weight which was regained in the latter part of the assay. For the peas assayed, both PER and NPU varied 3-4 fold between the best and poorest quality peas. Further- more, the best and the poorest varieties (67-4 vs. 66-5) had similar quantities of nitrogen showing that the super- iority in protein quality is not related to total nitrogen content. The observation that some varieties of peas (66-24, 66-15, 66-7, 66-34) contain large amounts of high quality protein suggests that it might be possible to improve these qualities by selective crossing. The peas should be bred to be appealing from a visual and nutritional point of View. 46 .muoHG swwmmo paw mom Um“ wumu mo mm>Hso fluzouw .H whomwm 8.. “w A.» Um?» Zn m0< 1.. r"! aura—g.“- l1 ZEm sflasnoau CHEDQH4 zmz samuoum Hmscflmmm manmuomuuxm z pamouwm mxmmHe mmumm Gmumo emumm mmlmm mamamm .Hmwfi mom unmonmm msflmuonm mom cw coausnwuumflc somOHUHzII.m manna 60 varied from 1.29-1.85 (66-1, 66—28) and neither this nor any other fraction studied correlated with (R<0.3) PER. To evaluate the quality of protein in fractions, an initial comparison of the biological value of the intact pea meal estimated by the microbiological method was made with the PER values obtained by rat assay (Bajaj et_al., 1969a). Since the values of the microbiological assay are expressed as a percentage of bacterial growth produced by the pea meal when compared with casein, a similar expression was obtained by dividing the PER or NPU values for these pea samples by that of casein. The NPU and PER ratios thus obtained were not significantly different (P<0.05) from the microbiological values of pea meals (Table 6). The microbiological values of the albumin fraction ranged from 117-128% of the bacterial growth produced by casein (Table 7). These values were not significantly different (P<0.05) since the proteins are not absolutely pure and the assay involves a certain amount of variation. On the other hand the differences in the values of the globulin fraction, which ranged from 64-100% were signif- icant (P<0.05). The microbiological values of the residue ranged from 28-56%. The peas of high biological value (66-24, 66-7, 66-23) had a residue of low biological value on the basis of the microbiological assay, while those that supported poor growth of rats (66-5 and Alaska) had a res- idue that supported good growth of the microorganisms. 61 Table 6.--A comparison of the biological values of pea meals as estimated by the microbiological assay and the values obtained by rat assay PER2 Expt./ NPU3 Expt./ Sample Microbiological PER Casein NPU Casein 66-1 63:2 51 62 66-3 32:6 27 31 66-5 40:6 l6 18 66-7 65:6 65 70 66-9 44:4 36 42 66-23 73:6 68 72 66-24 74:3 74 79 66-26 65:7 63 68 66-28 45:2 42 45 Alaska 30:8 23 26 1Values of PER and NPU were obtained from Bajaj 32 al. (1969a). 2PER = Weight gain of experimental rats / g protein eaten at 10% protein level. 3NPU = Gain in carcass nitrogen of experimental rats / g nitrogen consumed. A graphical representation of the quantity of al- bumin nitrogen plotted against the microbiological values is a curve (Fig. 3), the shape of which is similar to the curve between PER and the albumin content of pea meals. MICROBIOLOGICAL VALUE 0:: PEA MEALS 62 75" /066-24 6623 70- 65--—- 66-26C‘ 366‘? 664 60"“ 55—— 50.... 45'- 66—28 66-9 40— 35“" 66-3 30 ALASKA. :/ | l l _ J I 0.5 0.6 0.7 0.8 0.9 7.0 ALBUM”? i“~flTRCEJET£ ‘33 07 PE: ”2:35... Figure 3 Relation of albumin to the microbiological values of pea meals. 63 Table 7.--Microbiologica1 assay values* of protein frac— tions isolated from pea meals. Protein fraction Sample Albumin Globulin Residue 66-3 118:10 85:9 48 66-5 119:8 64:2 56 66-7 117:8 71:6 29 66-9 121:7 90:5 28 66-23 126:9 94:6 29 66-24 128:9 76:3 28 66-26 120:3 93:6 32 66-28 120:5 100:6 48 Alaska 128:11 97:4 52 *All the values are expressed as the percentage of response produced by the sample when compared with the response secured with casein. A possible explanation for the high biological value of the albumin fraction and its correlation to the protein quality of pea meals may reside in the sulfur con- tent of the albumin fraction. Osborne and Campbell (1898) reported that the legumelin fraction (albumin) contained 1.06% sulfur whereas the globulins, legumin and vicilin, contained 0.5 and 0.16% respectively. The sulfur content in an isolated protein is present almost exclusively as amino acids, and sulfur containing amino acids limit the biological value of many legumes (Phansalkar gt $1., 1957). 64 The albumin fraction is reported to contain 3—4 times more lysine and 3 times more tryptophan than the globulin frac- tion (Danielsson and Lis, 1952), since both these are essential amino acids and tryptOphan is the second limiting amino acid of legumes they may contribute to the superiority of the albumin fraction. The possibility that the amino acids in the albumin fraction might be more easily available should also be considered. The albumins are water soluble proteins and may be composed of small proteins easily accessible to the proteolytic enzymes. The proportion and the availability of amino acids in the egg albumin has been shown to be perfect for mammalian nutrition (Delhumeau, 1962). Genetic selection of pea varieties with high albumin content, and the biological value of these varieties should form basis of future experiments concerned with the incor- poration of nutritional quality in peas. REFERENCES AOAC (1965). Official methods of analysis (Association of official Agricultural chemists, Washington, D. C. edit. Bajaj, S., Mickelsen, 0., Baker, L. R. and Markarian, D. (1969a). Biological value of the protein in different varieties of peas (in preparation). ‘Bajaj, S., Mickelsen, O., Lillevik, H. A., Baker, L. R. and Gill, J. L. (1969b). Prediction of protein quality of peas from albumin content (in preparation). Danielsson, C. E. (1949a). Acta Chem. Scand. 3, 41-49. Danielsson, C. E. (l949b). Biochem. J. 33, 387-400. Danielsson, C. E. (1950). Acta Chem. Scand. 3, 762-771. Danielsson, C. E. (1952). Acta Chem. Scand. 3, 149-159. Danielsson, C. E., and Lis, H. (1952). Acta Chem. Scand. 3, 139-148. Delhumeau, G., Velez Pratt, G., and Gitler, C. (1962). J. Nutr. 11, 52. Dimler, R. J. (1966). Fed. Proc. 33, 1670-1675. Ford, J. E. (1960). Brit. J. Nutr. 33, 485—497. Mertz, E. T., Bates, L. S. and Nelson, 0. E. (1964). Science 145, 279-280. Mertz, E. T., Vernon, O. A., Bates, L. S. and Nelson, 0. E. (1965). Science 148, 1741-1742. Nelson, 0. E., Mertz, E. T. and Bates, L. S. (1965). Science 150, 1469-1470. Osborne, T. B. and Campbell, G. F. (1896). J. Am. Chem. Soc. 33, 583-608. 65 66 Osborne, T. B. and Campbell, G. F. (1898). J. Am. Chem. Soc. 33, 410-427. Phansalkar, S. V., Ramachandran, M. and Patwardhan, V. N. (1957). Ind. J. Med. Res. 33 (4), 611-621. Raacke, I. D. (1957). Bichem. J. 33, 101-110. PART III PREDICTION OF PROTEIN QUALITY OF PEAS FROM ALBUMIN CONTENT PREDICTION OF PROTEIN QUALITY OF PEAS FROM ALBUMIN CONTENT ABSTRACT Varieties of peas (Piggm sativum L.) with higher protein efficiency ratios (PER) contained more albumin nitrogen than those of lower PER, as determined by rat bioassay. For data from 21 varieties of peas, a quadratic function of the albumin content had a high coefficient of correlation (R = 0.949) with PER. The equation provides a simple quick and accurate method for the evaluation of protein quality of peas. The genetic selection of varieties of peas containing more albumin may be important for future improvement of PER values of peas or other legumes. INTRODUCTION Until recently, the primary emphasis in plant breeding was on improving the appearance, yield or disease resistance of seed crops. Mertz and his collaborators (1964) showed the possibility of improving the nutritional quality of cereal protein by genetic selection. Since a large part of the world's population is dependent on seed proteins, a redirection of effort in plant breeding appears justified. 67 68 Previous work with maize bred for high lysine indi- cates that most of this amino acid is associated with the alcohol insoluble protein (glutelin) fraction (Dimler, 1966). However, a significant correlation between the concentration of glutelin or some other protein in maize and its biolog- ical value has not been reported. EXPERIMENTAL Albumins as defined by Osborne (1924) are the water soluble heat coagulable proteins. The original procedure for the extraction of pea proteins was elaborated by Danielsson and Lis (1952) and later modified by Raacke (1957). This procedure was further modified for the present study to obtain better extraction. The results secured depended on the extraction procedure employed (Table 8). Based on the results shown in columns 1 and 2 of Table 8, the procedure was devised whereby the data shown in column 3 were secured. The field-dried peas used in this study were ground in a Wiley mill until the powder passed through a 20 mesh screen. The pea meal was homogenized for 4 minutes, with Standard buffer (pH = 6.8-7.0)1 at room temperature (25°C). The homogenate was centrifuged to separate the residue from 1Standard buffer: pH 6.8-7.0, 0.2M NaCl, 0.3M HPO4, 0.02M NaH PO . Na 2 4 2 Table 8.--Effect of extraction time and procedure on the 69 albumin nitrogen extracted. Albumin nitrogen percent of pea meal Extracted Extracted Extracted Sample2 Overnight3 2 minutes4 4 minutes5 66-16 0.461 0.528 0.630 66-19 0.431 0.469 0.650 66-4 0.514 0.560 0.614 66-34 0.525 0.331 0.702 66-24 0.514 0.603 1.060 66-14 0.513 0.580 0.626 66-12 0.606 0.485 0.660 66-31 0.513 0.651 0.779 Alaska 0.514 0.523 0.528 Average 0.510 0.525 0.694 2Sample: The first two figures represent the year of harvest, others the variety number. 3Extracted overnight with the magnetic stirrer, homogenizer was not used. 4Extracted 2 minutes, three times in a homogenizer. 5Extracted 4 minutes, three times in a homogenizer, as described in text. the extract. To ensure maximum extraction, the residue was homogenized twice more with fresh buffer. When Alaska peas were thus extracted, 70.7% of the total nitrogen was ob- tained in the first extract, 7.6% in the second, and 4.4% in the third extract. The combined extract was filtered 70 through a Whatman-No. 1 filter paper. The filtrate was dialyzed (dialysis tubing 27/100 - Union Carbide) against distilled water (1:100) for 48 hours at 4°C. During dial- ysis the water was changed 5-6 times. At the end of the dialysis period the globulins precipitated leaving the albumins in solution. The dialyzed solution was centrifuged for 1 hour at 5000 rpm. The albumin (supernate) was de- canted into a graduated cylinder. The volume was recorded for use in subsequent calculations. The protein content of the albumin fraction was determined by the micro Kjeldahl method (Bradstreet, 1965). The PER was determined according to the AOAC pro- cedure (1960). For this, the pea meal was incorporated into a purified ration2 to provide 10% protein. Weanling (3-week-old) male Sprague Dawley rats were fed the exper- imental diet for a period of 3 weeks. Casein3 at 10% level was used as a standard of comparison. The low coefficient of variability of PER (Chapman 32 33., 1959; Campbell, 1963) was one reason why it was chosen for the evaluation of protein quality. 2Purified ration: Corn oil 5%, Salt mix 4%, vitamin mix 2.2%, Alpha cel 2%, peas or casein to provide 10% protein, Sucrose to make up 100%. 3Casein: Vitamin free assay protein was obtained from General Biochemicals. 71 RESULTS AND DISCUSSION The PER values of the varieties of peas studied ranged from 0.46-2.20 g weight gain / g of protein eaten, while the albumin content ranged from 0.52-1.06 g nitrogen / 100 g pea meal (Table 9). The direct use of nitrogen for the expression of protein eliminates the necessity of choosing a conversion factor. The moisture content was 6-7% in each of the pea varieties.4 In the first experiment only 9 varieties of peas were analyzed for albumin nitrogen. In these varieties of peas the PER was proportional to the albumin content. When a quadratic curve was fit to the initial data, 99% of the variation in PER could be explained by measured differences in albumin (R = 0.998). This high correlation suggested that the equation of the quadratic curve in albumin content as (X) could be used for the prediction of PER (Y). Y = 23.2x - 12.8x2 - 8.2 -------- 1. To check the validity of the equation, additional samples were selected, coded to conceal their identity, and analyzed for albumin. The predicted values of PER were calculated from equation 1. The differences of the pre- dicted values from those obtained by rat assay ranged from 0.0 - 0.18, and the correlation was still high (R = 0.97). The data from both the experiments were pooled and a new curve with the following equation was obtained: 4Moisture determinations were made in a vacuum oven at 70°C. 72 Table 9.--Protein efficiency ratio and albumin content of pea'varieties. Albumin N Percent Sample Pea Meal PER 66-5 0.526 0.46 Alaska 0.528 0.63 66-3 0.573 0.74 66-38 0.607 1.30 66-28 0.624 1.16 66-14 0.626 1.45 66-9 0.628 1.00 66—16 0.630 1.34 66-1 0.640 1.44 66-19 0.650 1.57 66-12 0.660 1.56 67-28 0.666 1.65 66-4 0.674 1.82 66-26 0.679 1.76 66-7 0.705 1.80 66-34 0.707 1.86 66-17 0.721 1.70 66-15 0.724 1.90 66-31 0.779 2.20 66—23 0.819. 1.88 66—24 1.060 2.05 y = 24.7x - 13.6x2 — 8.8 -------- 2. The correlation of the data from 21 varieties of peas to the final curve was R = 0.949. At the lower levels of albumin, a unit change in albumin content produces a large change in the PER. The behavior of the curve outside the range of values studied cannot be predicted. The effect of increments in albumin above the values studied can be revealed only by further studies. Examples of predictions across the range of albumin studied will illustrate the preciseness of the procedure. For albumin nitrogen content of 0.5, 0.7 and 1.0%, the predicted values (: standard errors) are 0.11 : 0.37, 1.79 : 0.06, and 2.27 : 0.74, respectively. As in all regression analyses, predictions near the center of the rangeistudied are considerably more precise than those made at the ex- treme values. The probable validity of predictions made from average albumin content may be expressed as a confi- dence interval about the predicted PER value. Given an albumin nitrogen of 0.7%, for example, one may have 95 percent confidence that the actual PER value for such cases will fall between 1.67 and 1.91. The length of this in- terval is well within the range of errors normally encoun- tered in measuring PER. Correlation such as those reported here have been observed in experiments relating biological values to the 74 availability of lysine and methionine (Ford, 1960; Campbell, 1963). It is possible that the albumin contains consid- erable lysine and methionine in a readily available form. Danielsson and Lis (1952) showed that the tryptophan and the lysine content of albumin fraction was 2-3 times higher than that of the globuline (tryptOphan is one of the limit- ing amino acids of legume seeds). Much effort has been spent searching for simple methods to evaluate protein quality. The method presented here, because of its simplicity and accuracy, merits use in routine assays. Increase in albumin content above present levels and its effect on the nutritional quality of protein should provide an interesting area for continued research, not only in peas but in other legumes as well. REFERENCES AOAC (1960). Association of official agricultural chemists. Official methods of analysis. 9th edit. Washington, D. C. Bradstreet, R. B. (1965). The Kjeldahl method for organic nitrogen. Acad. Press. Campbell, J. A. (1963). Methodology of protein evaluation. A critical appraisal of methods for evaluating proteins in foods. Div. of Food Tech. and Nutr. Faculty of Agric. Sci. American Univ. of Beirut, Lebanon, pp. 19 and 70. Chapman, D. G., R. Castillo and J. A. Campbell (1959). Canad. J. Biochem. Physiol, 33, 679. Danielsson, C. E. and H. Lis (1952). Acta Chem. Scand. 3, 139-148. Dimler, R. J. (1966). Fed. Proc. 33, 1670. Ford, J. E. (1960). Brit. J. Nutr. 33, 493. Mertz, E. T., L. S. Bates and O. E. Nelson (1964). Science 145, 279-280. Osborne, T. B. (1924). The vegetable proteins. 2nd edit. Longmans Green and Co., London, England. Raacke, I. D. (1957). Biochem. J. 39, 101-110. 75 PART IV A STARCH GEL ELECTROPHORETIC ANALYSIS OF THE NUTRITIONALLY SIGNIFICANT AND SALT-EXTRACTABLE PROTEINS OF PEAS ABSTRACT The salt-extractable proteins of peas (33ggm.§3337 XEE.L') when analysed by starch gel electrOphoresis showed 3 or 4 major components. The two slower moving bands cor- responded to the globulin fraction, whereas the faster ones related to the albumin fraction. The unfractionated globu- lins of the pea protein extract migrated as a single streak with 2 bands faintly discernible. When an isolated prep- aration, of the two separated globulins, legumin and vicilin was subjected to electrophoresis, each migrated as a single band. The inclusion of dissociating and reducing agents (urea, 2-Mercaptoethanol) in the gel produced dissociation and/or scission of the globulins (3 new bands from vicilin, and 4 new bands from legumin) but had no effect on the major albumin component. INTRODUCTION According to the classical definition given by Osborne (1924) the proteins in peas are classified as al- bumins and globulins. The salt soluble proteins have been suggested to function as reserve proteins (Bonner and Var- ner, 1965). The two well defined proteins of the globulin fraction, legumin and vicilin, were isolated from peas by 76 77 Osborne and Campbell (1896, 1898). These researchers also isolated a water soluble protein fraction (albumin), and called it legumelin. The physico-chemical properties of globulins from peas have been studied extensively, and have been reported to be homogenous by both free boundary (Daniels- son, 1950) and zonal electrophoresis (Grant and Lawrence, 1964) as well as by analytic ultracentrifuge studies (Daniels- son, 1949). Tests on the albumin fraction indicated that several enzymes such as, phosphatase, amylase, and protease, were associated with it, and consequently it might be re— garded as micro-heterogenous on polyacrylamide gels (Fox and Thruman, 1964). The primary concern in most of the aforementioned electrophoretic characterizations of the pea proteins and their fractions has been homogeneity, number of protein com- ponents, and their distribution in plant seeds, to ascertain taxonomic relations. Significant differences in the nutritional value of the total proteins from various varieties of peas have been reported from this laboratory (Bajaj §E_g3., 1969a). Fur— thermore, the nutritional quality of meals from different varieties of peas was highly correlated to their percentage of albumin nitrogen (Bajaj §3_§3,,1969b). The biological value of the albumin fraction was found to be higher than that of any other protein fraction of peas assayed by a 78 microbiological method using Streptococcus zymogenes (Bajaj 33.33., 1969c). The aim of the present study was to make a prelimi- nary survey of the distribution, homogeneity, association behavior and polypeptide chain structure of the proteins in peas as detected by starch gel electrophoresis. It is also desired to obtain any relation between the electro- phoretically identifiable components of the pea protein fractions which were previously tested for biological values. EXPERIMENTAL Materials and Reagents Starch.--Solubilized potato starch (partially hydrolysed) was obtained from Connaught Medical Research Laboratories, Toronto, Canada, and the same lot was used in all experiments. Buffers.--The buffers contained in the starch gels or the buffer tanks were; 0.76M tris-citrate pH 8.6, 0.3M borate pH 8.6, 0.05M formate pH 3.1 (prepared according to Poulik, 1966). §E§3g.--The dye used for staining proteins was amido Schwartz and was prepared and used as described by Poulik, (1966). 79 Proteins.--The sample proteins (total extractable protein and fractions thereof) were prepared from different pea varieties by procedures outlined in Part II. Apparatus and Equipment.--The apparatus employed for horizontal starch gel electrophoresis was similar to that described by Sargent (1965). Other standard electro- phoretic equipment such as a constant volt power supply, and electrolytically destaining unit were used are described by Bloemendal (1963). Procedure Electrophoresis was performed on 300 m1 batch of starch gel under the following conditions: 1. A starch gel with no dissociating or reducing agent, under conditions of discontinuous buffers. Tris-citrate was used on the gel and borate in the tanks (method des- cribed by Scandalios, 1969). 2. A starch gel with no dissociating or reducing agent, in a continuous tris—citrate buffer (method described by Scandalios, 1969). 3. A starch gel with no dissociating or reducing agent in continuous formate buffer at acidic pH of 3.1 (method des- cribed by Scandalios, 1969). 4. A starch gel with 7M urea, in a discontinuous buffer tris-citrate on the gel and borate in the tanks (method des- cribed by Wake and Baldwin, 1961). 80 5. Starch gel with 7M urea and 2-Mercaptoethanol in a dis- continuous buffer, tris-citrate on the gel and borate in the tanks (method described by Wake and Baldwin). Samples.-—The samples for electrophoresis were pre- pared by dissolving in a solution composed of 2% starch prepared in the same buffer that was used for making the gel (7M urea was included when used on the gel). 0.1-0.3ml of dissolved protein sample was directly micropipetted into the slots and covered. Electrophoresis.--Electrophoresis was conducted at 10-20 milliamperes and 250 volts at 4°C. After electrophoresis for 8-12 hours, the gel was sliced into 3 layers and the middle layer was retained for staining. The gel was submerged in the dye solution for 15 minutes after which it was electrolytically destained in 7% acetic acid. RESULTS AND DISCUSSION Condition I.—-Electrophoretic analysis of the total salt-extractable proteins are seen in Figure 4. The two nearest the origin are heavily streaked (Plate I, slots 1-3) and are regarded as the pea globulins. The fastest migrating band suggests that it comes from the faster moving albumin component (Plate I, slots 4-6; Plate III, slots 1-6). The electrophoretograms of isolated preparations of legumin and vicilin show that legumin migrates ahead of vicilin. 81 219239.: Starch gel electrOphoresis of pea proteins (No dissociating agents) EXTRACTABLE PROTEIN A I iii? lCentrifuge :33: + éséia L I Albumin Globulin I'— “”"“‘0, IDialyse pH 4.7 + d' -————-mrmmfizmumnlmmu_n Legum3g Vicilin ' ‘--—---O .---",""-0 Ear: V - L + 1 *Note the letters, 0, V, L, A, refer to origin, vicilin, legumin and albumin resPectively. 82 The presence of one or two major components in the albumin fraction has importance from a nutritional stand- point. If the genes controlling the synthesis of the albumin components could be identified, pea varieties could be selectively crossed to produce an increase in the nutri- tionally significant protein components. Although various enzymes have been identified in the water soluble extract of peas, these might not account to any great extent for the quantitative distribution shown by the albumin bands. Condition II: The phenomenon of boundary self sharpening frequently occurs in systems with discontinuous buffers (Ornstein and Davis, 1962). Therefore to check the homogeneity of the albumin fraction, the electrOphoresis was performed in a continuous tris-citrate buffer. Under such conditions at pH 8.6 the albumin fraction separated indiscretely, but for the varieties of peas studied two bands were faintly discernible in a heavy streak (Plate II, slots 1-3). The globulins did not resolve under these con- ditions as seen in Plate II, slots 4-6. Condition III.--Starch gel without added reagents but run at acid pH of 3.1 in formate buffer, show that the albumin fraction appears mainly as one band (Figure 5) sug— gesting either (1) the albumin is a single complex at this pH, or (2) one of the two components observed at pH 8.6 has not acquired sufficient mobility. 83 Figure 5 Starch gel electrOphoresis of pea albumins (Continuous buffer) Formate buffer pH 2.1 V""' Tris citrate buffer pH 8.6 84 Condition IV.--The gels run with urea stained very poorly and did not yield much information and therefore they were discontinued. Condition V.--Electrophoresis of the total salt- extractable protein under these conditions yielded 8 dis- cernible bands (Figure 6; Plate VI, slots 1-3). The 3 closest to the origin can be identified as derived from vicilin (Plate V, slot 6). The major albumin band which migrated the furthests from the origin (Plate V, slot 5) still remained unaltered by the action of urea or 2-Mercap- toethanol. Among the remaining 4 bands which were obviously derived from legumin (by elimination), 2 represent proteins which have changed net charge and migrated towards the cathode in contrast to other proteins (Plate IV, slots 1-6). From the data obtained it may be concluded that the albumin is composed of l or 2 major proteins depending upon the pea varieties studied. Since the albumin fraction was found to have nutritional importance, the identifica- tion of the component containing the more essential and available amino acids is important for selecting varieties containing proteins of good nutritional quality. 85 Figure 6 Starch gel electrOphoresis of pea proteins (With 7M urea and 2—Mercaptoethanol) EXTRA CTABLE PROTE IN ' - :Centrifuge ! A lbw-.3311 Globu lin _. I I _, o I . lDialyse pH 4.7 ! 11.91.1111... A 86 Plate I mm mm a. . 2 I) w E23 fl 4 '13 5 IHEI A. I Starch gel electrophoretograms of the albumin and globulin fractions from 3 varieties of peas. Medium: Starch with no dissociating or reducing agents. Gel buffer: 0.76M tris- citrate at pH 8.6. Electrolyte buffer: 0.3M borate at pH 8.6. Running time: 8 hrs. at 4°C. Slots 1,2,3 contained 3% globulin from pea varieties Alaska, 66-24 and 66-15. Slots 4,5 and 6 contained 7% albumins from the corresponding pea varieties. The diagrammatic representation of the gel photo— graphs show the location of the pea protein fractions from the origin (0) and are identified by the letter in the fol- lowing plates as; vicilin (V), legumin (L) and albumin (A). 87 Plate II 0 G A A h," .., . I. 'c' l O I. c. .,.'. ‘a,’ ... . - . _ a . .5 ‘ + # Starch gel electrophoretograms of the albumin and globulin from Alaska peas. Medium: Starch with no dissociating or reducing agents. Gel buffer: 0.76M tris-citrate at pH 8.6. Electrolyte buffer: 0.76M tris at pH 8.6. Running time: 8 hrs. Slots 1, 2, 3 contained albumin (A) at l, 3, 5% concentration. Slots 4, 5, and 6 contained globulin (G) at 5, 3, and 5% concentration. 88 Plate III 0 y 3: :v w (A) N .—l + 01 mm 1 HI n 2 HE ; 3 BB 4 BB 5,3 HE HEI Starch gel electrophoretograms of albumin fraction from 3 varieties of peas. Medium: Starch with no dissociating or reducing agents. 333 buffer: 0.76M tris-citrate. Electro- lyte buffer: 0.3M borate at pH 8.6. Running time: 8 hrs. at 4°C. Slots 1, 2, and 3 contained 2% albumin (A) from variety Alaska, 66-15, and 66-24. Slots 4, 5, and 6 con- tained 3% albumin from corresponding pea varieties. 89 Plate IV 0 A5A4A3A2A1 A mm Inn-ram E35 EC: E3355: mat—"Imam manna: Emmmm “Ef‘”‘”““;w Starch gel electrophoretograms of the albumin fraction from 6 varieties of peas. Medium: Starch with 7M urea and 2—Mercapto ethanol. Gel buffer: 0.76M tris-citrate at pH 8.6. Electrolyte buffer: 0.3M borate at pH 8.6. Running time: 8 hrs. at 4°C. Slots 1-6 contained 4-5% albumin from Alaska, 66—24, 66—15, 67-4, 66-5, 66-7. 90 Plate V L1 L2 OV1V2V3 L3L4A1 A ' I l. 10 0 000 0000 2 El fl 0001 0000 i-. a 0 0 0000000 :0 4 0 I 000 0000 5 6 BER Starch gel electrophoretograms of pea proteins. Medium: Starch with 7M urea and 2—Mercaptoethanol. Gel buffer: 0.76M tris—citrate at pH 8.6. Electrolyte buffer: 0.3M borate at pH 8.6. Running time: 8 hrs. at 4°C. Slot 6 contained 3% vicilin, slot 5 contained 2% albumin, slot 4 contained 5% globulin, slot 3 contained 2% globulin, slot 2 contained 6% globulin, and slot 1 contained 3% globulin from Alaska pea. 91 Plate VI L1 L2 OVlV2 V3 L3L4A 100 000 200|||000 3ElElIII EEIEI 400IIEEI s00m0 600|0|0 + Starch gel electrophoretograms of pea proteins. Medium: starch with 7M urea and 2—Mercaptoethanol. Gel buffer: 0.76M tris-citrate at pH 8.6. Electrolyte buffer: 0.3M borate at pH 8.6. Running time: 8 hrs. at 4°C slots 1, 2, and 3 contained total salt extractable protein from varie- ties Alaska, 66-24 and 66-15 at 3% concentration. Slots 4, 5 and 6 contained corresponding globulins at 4% concentration. ‘Efi?’ ““ “”710“ REFERENC E S Bajaj, S., Baker, L. R. and Mickelsen, O. (1969a). Fed. Proc. (Abstr.) 33 (2), 811. Bajaj, S., Mickelsen, 0., Lillevik, H. A., Baker, L. R. fil and Gill, J. L. (1969b). Prediction of protein quality - of peas from albumin content. (In preparation) Bajaj, S., Mickelsen, O., Lillevik, H. A. and Baker, L. R. (1969c). Nitrogen distribution and biological value of protein fractions in different varieties of peas. (In ‘ preparation) 03 Bloemendal, H. (1963). Zone electrophoresis in blocks and columns. p 6.‘ Bonner, J. and Varner, J. E. (1965). Plant biochemistry. Acad. Press, p 774. Danielsson, C. E. (1949). Biochem. J. 33, 387-400. Danielsson, C. E. (1950). Acta Chem. Scand. 3, 762-771. Fox, D. J., Thruman, D. A. and Boulter, D. (1964). Phyto- chemistry 3, 417-419. Grant, D. R. and Lawrence, J. M. (1964). Arch. Biochem. Biophys. 108, 552-561. Ornstein, L. and Davis, B. J. (1962). Disc electrophoresis. Distillation Products Industries, Div. Eastman Kodak Co. Osborne, T. B. and Campbell, G. F. (1896). J. Am. Chem. Soc. 33, 583-608. Osborne, T. B. and Campbell, G. F. (1898). J. Am. Chem. Soc. 33, 410-427. Osborne, T. B. (1924). The vegetable proteins. 2nd edit. Longmans Green and Co., London, England, p. 168. Poulik, M. D. (1966). Methods of Biochem. Anal. 33, 455- 495. 92 .11“.-. 93 Sargent, J. R. (1965). Starch gel electrophoresis 33 Methods in zone electrophoresis. p. 55. Scandalios, J. G. (1969). Biochem. Genetics 3, 37-79. Wake, R. G. and Baldwin, R. L. (1961). Biochim. Biophys. SUMMARY AND CONCLUS IONS SUMMARY AND CONCLUS IONS The 32 varieties of peas grown under similar field conditions were provided by the Department of Horticulture. The peas were vine ripened, harvested by hand, and stored (until used) at 35-40°F under conditions of low humidity. The peas were ground into meal in the Wiley mill, and analyzed for nitrogen by the Kjeldahl method. The nitrogen was multiplied by a factor of 6.25 to obtain the quantity of crude protein in peas. The total protein thus calculated varied from 21-28% for different varieties of peas. All 32 varieties of peas were assayed for protein quality using 3-week-old Sprague Dawley rats. Since the number of the pea varieties to be assayed was large, four separate experiments were performed. Based on the simplic- ity of the method and the reliability of results obtained in the first experiment, net protein utilization (NPU) and protein efficiency ratio (PER) were selected for use in subsequent experiments. Casein at a 10% level, was used in the standard diet for comparison in all experiments. Some of the pea varieties were assayed more than once to obtain an indication of the precision of methods. The PER of the 94 95 best and the poorest variety of peas varied fourfold (0.46- 2.11), both these varieties (66-5, 66-4) had a similar quantity of protein (25.8, 25.5%). The PER values of some pea lines (66-4, 66-24) were close to that of casein, but the weight gain of the rats on pea diets was approximately half that secured on casein diets (429 vs 75g on casein). The reason for this difference was decreased food intake of pea-fed rats. Both PER and NPU yielded repeatable results, however PER had a lower standard deviation than NPU (0.12 vs. 2.12). The percentage of protein in the carcass of rats fed some varieties of peas (66-5, 66-7, 66-52 and Alaska) was significantly higher than casein fed rats (P< 0.05), this difference might be associated with a lower percentage of fat in the carcass of these animals. To determine the distribution of nitrogen in the major extractable protein fractions, varieties of peas were selected to represent high, low and mediocre PER. Such a selection was expected to reveal contrasts in the distribu- tion of nitrogen in different protein fractions as related to the PER of the pea meals. The percentage of albumin nitrogen in the pea meal correlated highly with PER (R = 0.949). The equation of the quadratic curve which fit the data relating PER to albumin content could be used to predict PER values of pea meals. The percentage of nitrogen in the other protein fractions did not correlate with PER (R<0.3). 96 The biological values of the pea meals were esti- mated by comparing the growth response produced by adding pea meal to the media for Streptococcus gymogenes, with the growth response produced when casein was used. In the same way, NPU values for the pea meals secured by rat assays were calculated as a percentage of the values for casein. The resulting ratios and the microbiological values were not significantly different (P<0.05). When the protein quality of albumin, globulin and the residue was estimated by the microbiological method, albumin had a higher biolog- ical value (117-128) than either globulin (64-100) or the residual fraction (26-56). The reason for the high protein quality of the albumin fraction might be a superior amino acid composition in terms of essential amino acids, better availability or both. The biological value of albumin from different varieties of peas was not significantly different (P<0.05), but the differences in the globulin and residue were significant. The extractable proteins of the pea were represented by 3-4 major bands when observed on starch-gel electro- phoresis. The slow moving bands were enveloped in a heavy streak and represented the globulin fraction. An isolated preparation of vicilin (a globulin) attracted the globulins when they were in adjacent slots during electrophoresis. The preceding two observations might be interpreted to indicate that the globulin fraction is present as a complex, 97 consisting of two proteins which can be separated on the basis of solubility. The albumin fraction was represented by one major and several minor bands. The addition of reducing agents (7M urea and 2-Mercaptoethanol) dissociated the globulins (3 bands - vicilin, 4 bands - legumin) but not the major albumin band. From the evidence presented it can be inferred that: l. Nitrogen is a poor index of protein quality. 2. Variation in the quantity and quality of protein exists in different varieties of peas even if they contain the same percentage of nitrogen. 3. That 94% of the variation in the protein quality (PER) of the peas studied can be explained on the basis of dif- ferences in albumin content. 4. The albumin content of peas can be used to predict their PER. 5. The biological quality of the albumin is better than the other fractions. 6. The albumin fraction contains one or two major proteins. APPENDIX A. AMINO ACID CONTENTS OF SELECTED VARIETIES OF PEAS Choice of Method With the advent of the amino acid analyzer complete analysis of amino acids which required a long time became possible to obtain in 24 hours. The complete analysis of the essential amino acids can be used to give a fairly accurate estimate of the biological value (Oser, 1951) in most cases except where the proteins are undigestible. Mitchell and Block (1946) devised a system of chemical scores based on the amount of essential amino acid in greatest deficit in a protein compared to the level present in a reference protein selected for its nutritional excel- lence. Despite the emperical nature of this system the chemical scores were shown (Mitchell, 1954a) to have a high degree of correlation (r = + 0.95) with published biological values for a series of proteins. However as Mitchell pointed out in a review (Mitchell, 1954b) the chemical score is an index of the biological value of the protein for growth only, since it assumes that absence of an essen- tial amino acid renders the protein completely unavailable for tissue maintenance, an observation not consistent with 98 99 reported observations (McCollum, 1911, Mason and Palmer, 1935). For scoring proteins by Mitchell's method the limiting amino acids of legume seeds need to be determined by methods other than the use of the amino acid analyzer, since some methionine, cystine, and tryptophan are lost by acid hydrolysis. Oser's method of the integrated amino acid index for predicting biological value of proteins accounts for several factors not considered in Mitchell's methods. Oser's method (Oser, 1959) is based on the observation that i for optimum utilization of food proteins all essential amino acids must be available and liberated for absorption at rates permitting mutual supplementation. The probability of such events occurring simultaneously is considered a function of the product of the probability of each essential amino acid being present at the site of protein synthesis. Since the essential amino acid index is a product function, the absence of an amino acid would yield a biological value of zero, which is not consistent with observations with most deficient proteins such as gelatin and gluten. In short term experiments the absent amino acids may be secured from the products of tissue catabolism. This method therefore requires that one assume the presence of small quantities of missing amino acids, to be able to correlate the estimates with reported biological values. Experiments 100 The amino acid content of four selected varieties of peas was estimated on the Spinco 120-C Amino acid analyzer. The main purpose of this experiment was to determine if differences the observed differences values reported in Table alone cannot explain the in amino acids could account for in the biological values. The 10 indicate that methionine values differences in the biological value since the variety 66—5 contains the most methionine and has the lowest BV, similarly the essential amino acid index cannot explain the low biological value of this variety. The low BV of 66-5 appears to be due to some other reason than poor amino acid content. The BV of 67-4 as observed by the balance method was 70.7 which is quite like the value obtained by the essential amino acid index (70.9). The observed values of the amino acids are within the range of reported values for Alaska pea (Biological data handbook) although they are frequently at the upper range. “Ml—J... 0“- 101 Table 10.--Amino acid composition of pea meals g/100g nitrogen. Alaska Alaska Amino Acid 67-4 66-5 66—7 pea peal Alaska pea2 Tryptophan 12.1 9.3 12.3 10.2 11.9 - Lysine 49.5 43.8 43.1 50.2 43.1 29.5, 8.9-43.6 Leucine 51.7 49.9 47.0 47.3 40.6 39.0, 29.3-54.6 Tyrosine 25.4 23.1 23.6 24.0 11.25 15.2, 10.0-25.8 Phenyl Alanine 34.8 31.2 31.2 32.5 24.0, 15.1—34.8 Valine 38.2 34.8 31.9 32.2 25.6 25.6, 19.4-32.8 130 Leucine 30.8 35.1 29.0 28.1 23.1 28.7, 23.0-36.7 Methionine 2.6 3.9 2.1 2.4 4.4 5.0, 1.7- 9.7 1/2 Cystine - - - - - 6.8, 3.9- 9.6 Threonine 29.9 25.6 27.3 25.6 25.6 22.9, 17.3-29.8 Arginine 48.7 56.4 63.1 66.9 59.4 55.5 Histidine 15.9 14.8 15.1 13.1 13.7 10.2, 6.1-19.8 Aspartic Acid 82.5 83.0 77.9 75.1 - - Serine 34.5 33.1 37.9 30.8 - - Glutamic Acid 86.0 131.1 103.0 110.9 - — Proline 31.1 22.0 26.0 26.5 - — Glycine 29.6 25.9 26.0 30.8 Alanine 31.1 27.1 30.5 26.5 - — *EAA Index 70.9 71.8 64.5 64.1 - - ‘FU O ' , . 7‘ ,' P“. 5"" 1 "Q" _ 1Alaska peas--Values from Evans and Bandemer (1967) converted to g/100g nitrogen. 2Alaska peas--Values from Biological data handbook. *Egg protein was used as a standard for calculating essential amino acid index. REFERENCES Evans, R. J. and S. L. Bandemer (1967). Agric. Food Chem. 33, (3) 439-443. Mason, I. D. and L. S. Palmer (1935). J. Nutr. 3, 489. McCollum, E. V. (1911). Am. J. Physiol. 33, 215. Mitchell, H. H. (1954a). Wiss. Abhandl. Deut. Akad. Land Wirtsch. 3, 279. Mitchell, H. H. (1954b). Biological value of proteins and amino acid interrelationships. Quartermaster Food and Container Inst. Surveys. Natl. Res. Counc. Washington, D. C. Mitchell, H. H. and R. J. Block (1946). J. Biol. Chem. 163, 599. Oser, B. L. (1951). J. Am. Diet. Assoc. 33 (5), 396-402. Oser, B. L. (1959). An integrated essential amino acid index. 32 Albanese, A. A. (ed.) Protein and amino acid nutrition., p. 281, Acad. Press. 102 1 B. MOISTURE CONTENT OF PEAS Table 11 Percentage Percentage Sample Moisture Sample Moisture 66-1 6.81 66-17 6.95 66-2 6.90 66-19 6.82 66-3 6.93 66-20 7.00 66-4 7.01 66—22 6.93 66-5 7.00 66-23 6.86 66-6 6.61 66-24 6.90 66-7 6.89 66-26 6.87 66—8 6.99 66-27 7.11 66-9 6.78 66-28 6.67 66-10 7.12 67—28 6.70 66-11 6.88 66-33 6.80 66—12 7.00 66-34 6.81 66-13 6.69 66-38 7.15 66-14 6.87 66-52 6.77 66-15 7.10 Alaska 6.99 66-16 6.86 lMoisture was estimated in a vacuum oven at 70°C for 48 hours. LBR IIWWIIIIIIll!HIIIWIIHWIHIIVIIINMli 5706 WIN